<?page no="0"?> ISBN 978-3-381-15521-7 This volume presents the results of an intervention study involving 9 th -grade high school students and apprentices in the hospitality industry, which empirically examines writing in a professional context. The study focuses on the question of what writing performance the students demonstrate when drafting job applications and advertising letters during the course of the writing project using “self-regulated strategy development”. To answer this question, both the written products and the writing processes were recorded and evaluated primarily using quantitative methods. The goal of the study is to improve the writing competencies of the participating students based on the selected text types, as measured by content, linguistic, and formal aspects. Giera Promoting career-oriented writing skills through SRSD Winnie-Karen Giera Promoting career-oriented writing skills through SRSD An evaluation of a school writing project <?page no="1"?> Promoting career-oriented writing skills through SRSD <?page no="3"?> Winnie-Karen Giera Promoting career-oriented writing skills through SRSD An evaluation of a school writing project <?page no="4"?> Prof. Dr. Winnie-Karen Giera Universität Potsdam · Philosophische Fakultät · Institut für Germanistik Am Neuen Palais 10 · 14469 Potsdam https: / / orcid.org/ 0000-0001-8485-8570 This publication is an English translation of the work Berufsorientierte Schreibkompetenz mithilfe SRSD fördern (ISBN 978-3-8233-8373-4), which was published in 2020 as volume 1 of the peer reviewed series Kommunizieren im Beruf. Die Veröffentlichung wurde aus Mitteln der Universitätsbibliothek Potsdam für Open-Access-Monografien gefördert. / This publication was supported by funds for Open Access Monographs from the Potsdam University Library. DOI: https: / / doi.org/ 10.24053/ 9783381155224 © 2026 · Winnie-Karen Giera Das Werk ist eine Open Access-Publikation. Es wird unter der Creative Commons Namensnennung - Weitergabe unter gleichen Bedingungen | CC BY-SA 4.0 (https: / / creativecommons.org/ licenses/ by-sa/ 4.0/ ) veröffentlicht, welche die Nutzung, Vervielfäl‐ tigung, Bearbeitung, Verbreitung und Wiedergabe in jeglichem Medium und Format erlaubt, solange Sie die/ den ursprünglichen Autor/ innen und die Quelle ordentlich nennen, einen Link zur Creative Commons-Lizenz anfügen und angeben, ob Änderungen vorgenommen wurden. Die in diesem Werk enthaltenen Bilder und sonstiges Drittma‐ terial unterliegen ebenfalls der genannten Creative Commons Lizenz, sofern sich aus der am Material vermerkten Legende nichts anderes ergibt. In diesen Fällen ist für die oben genannten Weiterverwendungen des Materials die Einwilligung des jeweiligen Rechteinhabers einzuholen. Narr Francke Attempto Verlag GmbH + Co. KG Dischingerweg 5 · D-72070 Tübingen Internet: www.narr.de | eMail: info@narr.de Lektorat: Anitha Rose George Druck: Elanders Waiblingen GmbH ISBN 978-3-381-15521-7 (Print) ISBN 978-3-381-15522-4 (ePDF) ISBN 978-3-381-15523-1 (ePub) Umschlagabbildung: © AndreaObzerova (istockphoto.com) Bibliografische Information der Deutschen Nationalbibliothek Die Deutsche Nationalbibliothek verzeichnet diese Publikation in der Deutschen Nationalbi‐ bliografie; detaillierte bibliografische Daten sind im Internet über http: / / dnb.dnb.de abrufbar. <?page no="5"?> 1 9 2 19 2.1 19 2.2 33 2.3 40 2.3.1 41 2.3.2 51 2.4 65 2.5 81 2.6 98 2.7 119 2.8 121 2.9 126 2.10 128 3 137 3.1 137 3.2 147 3.3 152 3.4 160 3.5 164 3.6 167 3.6.1 167 3.6.2 171 3.7 174 4 179 4.1 179 4.2 185 Contents Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Theory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Writing skills and their development . . . . . . . . . . . . . . . . . . . . . Writing skills before vocational training at the end of lower secondary level . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Writing skills in dual training . . . . . . . . . . . . . . . . . . . . . . . . . . . Dual vocational training in Germany . . . . . . . . . . . . . . . Writing in dual training . . . . . . . . . . . . . . . . . . . . . . . . . . Text (Type) competence in the professional field . . . . . . . . . . . Writing process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Writing texts using the SRSD Method . . . . . . . . . . . . . . . . . . . . Preliminary investigations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Summary of the state of research . . . . . . . . . . . . . . . . . . . . . . . . Research idea . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Research questions and hypotheses . . . . . . . . . . . . . . . . . . . . . . Research design & methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Sample . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Test instruments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Procedure of the writing project . . . . . . . . . . . . . . . . . . . . . . . . . Data collection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Data preparation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Course of the text rating training . . . . . . . . . . . . . . . . . . Data preparation of the individual test instruments . . . Data evaluation in the sub-areas . . . . . . . . . . . . . . . . . . . . . . . . . Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Key results from the affect research area . . . . . . . . . . . . . . . . . Key results from the behavior study area . . . . . . . . . . . . . . . . . <?page no="6"?> 4.3 189 4.4 198 5 211 5.1 211 5.2 218 5.3 222 5.4 227 229 251 253 Key findings from the content learning study area . . . . . . . . . Key Findings from the metacognition research area . . . . . . . . Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Discussion of the Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Critical reflection on the study . . . . . . . . . . . . . . . . . . . . . . . . . . Outlook . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Epilogue . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . List of Figures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . List of Tables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 Contents <?page no="7"?> For my wonderful son, Friedrich <?page no="9"?> 1 In the text of this paper (except the project title, the questions, the hypotheses, quotes and illustrations), a gender-neutral and reader-friendly designation is used, which takes into account both men, women and intersexuals. Quotations and illustrations have not been changed unless indicated, even if they contain typographical errors. Terms relevant to the work have been written in italics or bold. 1 INTRODUCTION “Text is all around you” (Dürscheid, 2007: 3). Writing enables individuals to put their thoughts on paper and to develop thoughts through writing. 1 Writing is a creative and emotionally driven process that conjoins the body, mind, and soul. Furthermore, writing connects the writer with a reader - an addressee or a circle of addressees. Each individual’s personality influences the writing process, guiding it through their own effort and motivation. Everyone writes differently and develops their unique style. Even on typed pages, a writer’s ‘handwriting’ can be identified through their word choices and their connections to individual word groups and sentences. Writing is, in essence, always individual. The use of language generally has a communicative and cognitive function (Quasthoff & Heller, 2014: 7). Consequently, writing occurs within communi‐ cative situations that are embedded in a socio-cultural ‘space’. Anyone who writes addresses their text to a real or fictional recipient. Writing constitutes a fundamental aspect of cultural expression and transmission. As a cultural technique and key competence, writing is a skill that performs various functions in the development of our society. When a text is intended for others, it can convey knowledge, prompt action, and serve practical purposes. Texts are thus created in social contexts (Ehlich, 1994: 20). Writing is an integral part of everyday life, involving several communicative tasks such as composing job application letters, submitting complaints, or completing bureaucratic forms. It is a significant aspect of the cultural dynamics of daily life. Contemporary social, linguistic, and professional activities are characterized by texts, both oral and written. Writing and handling professional text types such as letters is commonplace in later (professional) life because these occassions also require “social interaction” (Heinemann, 2008: 114, translated by WKG) between the writer and the addressee. The associated professional goals and resulting actions should be fulfilled as a “communicative function” <?page no="10"?> (Heinemann, 2008: 115, translated by WKG). Both the terms everyday culture and writing culture should be viewed critically, as writing is still characterized by the power dynamics of a socially higher class, which poses challenges for some students pursuing social advancement (Koch & Pielow, 1984: 66). However, the global labor market increasingly requires skilled workers with writing skills (Becker-Mrotzek & Böttcher, 2006: 23). As a result, the European Parliament has formulated eight key competences for lifelong learning (Euro‐ pean Union, 2006, translated by WKG). Four of these skills are relevant to professional writing and form the background of this paper: the first skill is “communication in the mother tongue”, which comprises both oral and written skills (ibid.), the second is “digital literacy”, e.g., writing on a PC. The third is “learning to learn” (ibid.), which particularly involves self-organization and application of diverse methods. The fourth competence is “social and societal competence” (ibid.), which is a critical factor for success in professional life. Both the European Union and the Organization for Economic Cooperation and Development (OECD) recognize the significance of writing skills. Within a professional context, writing skill is classified as an “informa‐ tion-processing skill” (OECD, 2013: 143, translated by WKG) alongside key competencies such as “co-operating or collaborating with co-workers” (ibid.) and “self-organizing skills” (ibid.). Professional writing therefore requires a combination of social, methodological and technical skills that should ideally be taught at school. For example, business and semi-private letters are often a collaborative effort rather than the work of a single individual. This may involve designing the layout, offering suggestions for formulating these letters, conducting repeated read-throughs, providing proofreading suggestions, or, finally, obtaining the signature of the management. Firstly, writing is a basic tool with a range of intertwined functions (Becker-Mrotzeck & Böttcher, 2006: 16). Therefore, it has a social and cultural significance and promotes cognitive development, psychological relief and interpersonal communication (Fritzsche, 2003: 201 f.). However, writing is also determined by institutional frameworks and enables students’ “[…] participation in the ‘educational language of schooling’ […]” (Schründer-Lenzen, 2009: 122, translated by WKG). For beginners, the initial stages of writing often involves free or random writing where they may not clearly observe norms and conventions. It is only through systematic training at school that students begin to consider spelling and grammar rules, use punctuation correctly, and learn various text types along with their associated structures. This phase facilitates the development of beginners into motivated or even expert writers. The aim of the school as an institution is to integrate 10 1 Introduction <?page no="11"?> writing into the curriculum, focusing on everyday texts that are relevant for long-term use, such as, sales contract or employment contracts. Therefore, didactic concepts for writing are required to ensure that no pupil is left behind and to continuously develop their writing skills. For students with low literacy skills, it is imperative to “[…] compensate for the lack of learning opportunities in the extracurricular environment through institutional support” (Stanat et al., 2010: 202, translated by WKG). As school progresses, writing and text skills therefore become the primary focus of nearly all subjects. Success at school is linked to an “[…] adequate command of the German language. It is the medium of communication in and with the majority society” (Steinmüller, 2006: 322). Writing skills, therefore, play an important role in academic success. Most theoretical final examinations are written, whether the training examination of the vocational chambers, the school-leaving examination for the Hauptschulabschluss or Realschulabschluss or the (technical) Abitur, and the same applies to university or company entrance examinations: Those who cannot write in school language are at a disadvantage in the assessment in all subjects - even if the answers are correct in terms of content - also because the proportion of written work in the overall assessment is extremely high (Neumann, 2010: 14, translated by WKG). In German lessons at lower secondary level, pupils primarily write for the teacher. Writing conventions and norms play an important role as the grade level increases. Added to this is the aspect of time pressure. In contrast to elementary school, the number of German lessons is reduced. However, writing takes time: time to plan, discuss, revise, present and reflect. In secondary school, the competence area of writing includes teaching “writing stances (narrative, informative, argumentative)” (Abraham & Frederking, 2017: 58, translated by WKG) and “text and writing patterns (narrative, reporting, describing, explain‐ ing, discussing etc.)” (ibid.). Text patterns are taught using text examples, and writing instructions include sub-processes of planning and revising as well as writing strategies linked to school text patterns. Longer-term writing projects promote the listed areas of writing competence, the setting of writing goals and individual writing development (Abraham & Frederking, 2017: 70, translated by WKG). Writing skills can therefore mature sustainably over the long term: This area of competence makes a significant contribution to the ability to study, train and work by familiarizing learners with writing and introducing them to text types that are used for communication in everyday life and at work (e.g. factual letters, minutes) and for acquiring knowledge (e.g., summaries, excerpts) (ibid.). 1 Introduction 11 <?page no="12"?> The competence areas of Handling media and Writing and reading are closely linked. Writing on a PC requires and encourages repeated reading. Digital media texts, such as text messages, emails, chats or texts in social networks, have become an integral part of the media diversity in German lessons and are now indispensable in our daily lives (Abraham & Frederking, 2017: 64). Writing occurs in nearly every subject and the skills learned in German can and should be used in other subjects as transferable skills, e.g., in the form of the communicative or knowledge-expanding function (heuristic-epistemic). Thus, the acquisition of writing skills is not only the task of German lessons, even if these skills are specifically promoted there. German didactics, often regarded as the cradle of German teaching is subject to constant social change. German was first introduced as an A-level subject in 1812 (Abraham & Frederking, 2017: 53). In the 1970s, it was divided into literature and language didactics. In 1984, writing didactics was still described as a “white spot” or an area of neglect (Koch & Pielow, 1984: 1, translated by WKG). Writing didactics has, at times, faced criticism: As many examples show, writing can be such a productive, humane, social, democratic, in a Writing didactics was sometimes criticized has, at times, faced criticism: word, cultural-revolutionary force - if only it is truly productive writing and is not immediately pressed theoretically and practically back into schemes that make it available to ruling economic, political, scientific powers (Koch & Pielow, 1984: 2, translated by WKG). The advent of internet and its possibilities expanded the above-mentioned dichotomy to include media didactics to the field. From that point, German lessons were referred to as “linguistic, literary and media education” (Abraham & Frederking, 2017: 54). The introduction of new media now also integrated audio books, films, PC research and “writing and reading on the PC” into the subject (KMK, 2004, translated by WKG). Today, German lessons aimed at the intermediate school leaving certificate encompass the four areas of competencies of “speaking and listening” (ibid.), “writing” (ibid.), “examining language and language use” (ibid.) and “reading - dealing with texts and media” (ibid.), which apply to all pupils in Germany. Writing didactics is now interlinked with the disciplines of psychology, language didactics and linguistics, among others. It necessitates research networks and writing studies from the primary school level to professional writing (ibid.). The associated research area of Writing at Work has only been established since the 1980s, predominantly in the Anglo-American language area ( Jakobs, 2005: 13). The research field of writing at work is gaining in importance with the predicted transition from the information society to the knowledge and service society and 12 1 Introduction <?page no="13"?> - more broadly - to the health care society. Each of the three development trends demands considerable communicative skills from those practicing the profession ( Jakobs, 2005: 15, translated by WKG). For experts and researchers in German didactics focusing on the writing process, assessing writing competence with professional text types on the PC as well as evaluating writing projects as interventions, is highly valuable. German didactics requires authentic German lessons with entire classes, assessed using a variety of test instruments, including writing tasks, questionnaires, tape and video recordings and recordings of keyboard writing, because “empirical didactics of writing in German-speaking countries is a young interdisciplinary field dedicated to learning and teaching writing” (Steinhoff et al., 2017: 9, translated by WKG). Every individual must handle application letters when applying for jobs. In today’s labor market, there is a tendency for employees to change careers more frequently than 100 years ago. The traditional career path, completing training in a company after school, being hired there, and working until retirement, has largely disappeared in the fast-paced modern job market. Almost every citizen in the country writes job applications, which companies often require even for school internships. These applications are embedded in the German subject curriculum, yet they have so far received little attention in the field of German didactics research. The application letter is required at the start of a career. It is a form of letter that, on the one hand, adheres to the DIN 5008 standard, which imposes a high degree of standardization, yet it also demands a personal style and careful alignment with the addressee (Grün, 2013) - a balancing act that offers a compelling area of study from a text linguistic perspective. German teachers, parents, careers advisors and business teachers are not only asking for insights into this subject matter, but also want guidance on how to teach it: How should application letters be taught? This challenge requires a thoughtful approach to both didactics and methodology. While German teachers should be the primary contacts for these questions, German and business textbooks offer numerous practice opportunities, such as sample letters illustrating the components of an application. However, this alone is insufficient. Writing effective application letters demands skills to develop independently, especially as application formats evolve. Where starting a new job might once have been secured with a handshake, companies are now legally required to archive cover letters in case of employment law claims. Even today, offers are still predominantly made verbally. The offer is legally the first declaration of intent, whether in the supermarket or a bakery. As consumers, we can make the second declaration of intent, and thus conclude a 1 Introduction 13 <?page no="14"?> contract, by making a purchase at the checkout or leaving the store. However, the increasing flood of e-mails obscures declarations of intent. Consumers accidentally click on a confirmation button and often make a second declaration of intent without realizing it. Young people in particular are easy victims for this type of contract and usually have huge cell phone bills. Many young people are not yet aware that this is an unsolicited offer, as the legal background to contracts is only taught at vocational schools. Lower secondary schools have not yet dealt with this (Nds. KM 2006a-e). In later everyday life, consumers may receive unsolicited offers online via email and through traditional mail. It is important to close a curricular gap here. Letters following the DIN standard, especially letters of application and unsolicited offers, are often written to meet legal requirements. In professional contexts, employees, especially those in commercial roles within the service sector, frequently draft unsolicited offers in order to attract more customers for a service or product and thus increase sales. Every entrepreneur, regardless of the industry, must face this writing challenge. As a German teacher, you not only have to teach the theory, but also practice writing letters with pupils on a PC from the eighth or ninth school year at the latest. A German teacher is required not only to teach content related to letter writing, but also to be familiar with word processing programs, and to be able to teach not just one pupil, but an entire class. In addition to text-linguistic and computer-based know-how, a further teaching quality is required: keeping an eye on the writing process. This can be visually supported for the pupils with the help of the PC and projector. Furthermore, due to the heterogeneous student body, the students should be provided with several learning approaches to this topic - with all senses, so to speak (Akoun/ Pailleau, 2014: 40). For this reason, a teaching professional is required who is proficient in various social forms with teaching methods - and also wants to implement them. These initial explanations show that entry into training must also be suppor‐ ted more strongly through the writing of relevant professional text types. This should be prepared not only by vocational schools, but also by general education schools ( Jakobs, 2005, 2008). Statements made by trainees in a study on the linguistic and communicative requirements of trainees (Efing, 2010) show the discrepancy between general education schools and in-company training in the area of written communication: When asked whether and how well they felt their general schooling prepared them for the linguistic and communicative requirements of vocational training, a consistent tenor of the responses was that school had prepared them extensively for linguistic skills such as spelling and grammar (analysis), which are perceived as irrelevant for 14 1 Introduction <?page no="15"?> the profession, but not (sufficiently) for communicative skills and typical professional text types and language action patterns […] (Efing, 2012: 9, translated by WKG). Insufficient text and writing skills on the part of school leavers can also mean that applicants are not prepared for training. The number of companies that see the lack of training maturity of school leavers as an obstacle to training is stagnating at a high level. The lack of training maturity thus remains the number one obstacle to training (Deutscher Industrie- und Handelskam‐ mertag, 2012: 32, translated by WKG). For example, half of the training companies in the Chamber of Industry and Commerce complain about trainees’ inadequate written expression skills (Deutscher Industrie- und Handelskammertag, 2012: 33). However, the dilemma surrounding the demand for and supply of training places is becoming even more acute: vacant training places cannot be filled, particularly in the catering and hotel industry. This situation shows that a student who wishes to train in the catering/ hotel industry has a wide choice of training places, provided they are ready for training. In order to successfully counter inadequate text type and writing skills using a prototypical example, a doctoral project is to be carried out as part of a school writing project with students from general (n = 77) and vocational schools (n = 26) (specializing in the hotel and catering industry). The background is the author’s personal wealth of experience in the hotel industry as a trained hotel manager, as an IHK examiner for the intermediate and final examinations for hotel managers, as a self-employed consultant and business economist (IHK) in the hotel and catering industry and as a trained and civil servant secondary school teacher (M. Ed.) for the subjects German, politics and economics. The subject of teaching in German include writing cover letters and unsolicited offers following the DIN standard 5008 on the computer as an example and to gain knowledge about the form of an initial declaration of intent for contracts, as in the case of offers. Therefore, this doctoral thesis with the project title Evaluation of a career-oriented writing project: Lower Saxon pupils write letters of application and offers with the help of SRSD and feedback interviews at the Chair of German Didactics under the supervision of Prof. Astrid Neumann (University of Leuphana, Lüneburg) and Prof. Christian Efing (Bergische University, Wuppertal) was created with the fundamental question: How does the writing competence of students from general and vocational schools develop with the help of the SRSD approach using the text types “cover letter” and 1 Introduction 15 <?page no="16"?> 2 This project title was shortened in the course of publication and slightly changed for the book title. “unsolicited offer” in the course of a writing project?   2 The aim of the work is to present the strengths and weaknesses of a tested and evaluated writing project for secondary level I teachers, trainee teachers, German students, (vocational school) teachers and trainers. The writing project is intended to provide an initial insight into teaching with professionally relevant text types that are didactically prepared and action-oriented. This thesis is divided into five chapters: This Introduction (Chapter 1) is followed by the Theory and scientific foundations (Chapter 2). Theoretical foundations and the current state of research are not explained separately, but in an integrative manner. It begins with the precise definition and the current state of research on writing competence and its development (Section 2.1). The chapters on writing competence before vocational training at the end of lower secondary level (Section 2.2) and writing competence in dual training (Section 2.3) provide insights into writing didactics and curricular writing requirements. Text (type) competence in the vocational field (Section 2.4) is then presented in detail with text linguistic and didactic references. This chapter is particularly essential for the evaluation of student products. The writing process (Section 2.5) is considered separately from writing competence. The two areas are deliberately treated separately, even if there are overlaps. Nevertheless, the complexity of both concepts is too great to explain them in just one chapter. The SRSD approach is then explained in detail, in the chapter Writing texts using the SRSD method (Section 2.6). This is followed by a presentation of the preliminary research carried out (Section 2.7), which is bundled together with the other sections in a summary of the current state of research (Section 2.8). From here, the research idea (Section 2.9) is formed, setting the stage for the final chapter on research questions and hypotheses (Section 2.10). The third chapter, Research Design & Methods, is classically divided into method (Section 3.1) and the presentation of the sample (Section 3.2) and the test instruments (Section 3.3). This is followed by the process of the writing project (Section 3.4) and the triad of data collection (Section 3.5), data preparation (Section 3.6) and data analysis (Section 3.7). The fourth chapter presents the central Results from the study areas of Affect (Section 4.1), Behavior (Section 4.2), Content Learning (Section 4.3) and Metacognition (Section 4.4). The results lead to the final chapter, Discussion, which describes the consequences of the writing project (Chapter 5). For a better overview, this chapter is also divided into a discussion of the results (Section 16 1 Introduction <?page no="17"?> 5.1), a critical reflection on the study (Section 5.2), an outlook with research desiderata (Section 5.3) and an epilogue (Section 5.4). 1 Introduction 17 <?page no="19"?> 2 THEORY The second chapter provides a comprehensive analysis of the current state of research, as well as the scientific foundations and the associated theoretical concepts. In addition to the concept of writing competence, writing develop‐ ment is explained using models and research findings. This is followed by a specification of writing competence before and during vocational training. The explanation of text type competence in the professional field is intended to bring the professional writing requirement closer. This is followed by an explanation of the theory of the writing process, which includes the development of several models. The chapter continues with a presentation of the SRSD method and previously conducted studies. Finally, the chapter concludes with a summary of the state of research, followed by the research idea, which introduces research questions and hypotheses. 2.1 WRITING SKILLS AND THEIR DEVELOPMENT Competence includes abilities and skills that a person possesses and demon‐ strates in certain situations, but cannot always be regarded as performance. Weinert (2001: 27) further specifies that competencies can also be learned and that the underlying abilities and skills are linked to motivational, volitional and social dispositions that are used variably in situations. Example: A skier will not be able to demonstrate the ability to ski on the beach in summer. It is a skill that he has learned and can use as a performance in the situation of a winter vacation with snow. Competence is therefore a secure knowledge and ability (Thomas, 2007: 94). Psychologists classify this term as “dispositions” (Sieland & Rahm, 2007: 199, translated by WKG) or “latent response readiness” (ibid.). A learned behavior can therefore be present in the “behavioral repertoire” (Winkel/ Petermann & Petermann, 2006: 191, translated by WKG), but only becomes visible in the “behavioral performance” (ibid.). Whether this performance is shown again depends on the intrinsic motivation. This symbiosis is explained as: <?page no="20"?> A person must be motivated to perform a behavior; in other words, there must be a reason to act. Motivation is determined by the expectation of receiving a reward from the behavior. These expectations are learned through experience (Winkel/ Petermann & Petermann, 2006: 191, translated by WKG). In linguistics, the concept of competence and, more specifically, communicative competence, is defined as a construct with specific characteristics and definite indicators: By definition, competencies are not directly observable as cognitive processes, but must be examined on the basis of derived constructs. They then become visible on the (linguistic) surface in the performance. Within the framework described above, this ‘showing’ of competence must be initiated in a comparable way for all and evaluated using common, defined indicators (Neumann, 2014: 58 f., translated by WKG). Against this background, communicative competence can be divided into sub-com‐ petences as a multi-modular construct: a language system competence (grammatical competence), a sociolinguistic competence (repertoire, mastery and awareness of varieties and registers, ability to code-switch flexibly), a pragmatic competence (to achieve goals), a text/ discourse competence (text structuring competence) as well as a strategic and a social/ sociocultural competence (Efing, 2015: 21, translated by WKG). In summary, competencies are not directly observable, but can only be seen in their performance. Our school system, in turn, does not assess competence, but evaluates performance, a specific feature of competence (von Saldern, 2011: 38). Writing competence is therefore a latent construct, as it has latent characteristics that cannot be observed in their entirety (Grabowski, 2017: 317). The following section discusses this in more detail. As writing competencies are not directly observable, specific manifest indicators must be chosen in order to measure latent writing competence. ‘Against this background, the special feature of writing didactic research involving text corpora is that texts are not simply examined, but texts are regarded as indicators of an individual’s writing competence’ (Steinhoff, 2017: 354). Furthermore, writing tasks exhibit text indicators through texts produced: “How competent a writer is can be recognized by the writing process, but always also by the writing product” (Steinhoff, 2017: 352, translated by WKG). Writing competence is therefore closely linked to text production: The term writing competence focuses on effective, targeted and addressee-oriented text production. In addition to graphomotor and (meta-)cognitive processes, which form the core of writing, this requires a certain motivation to write, and writing has a social dimension in that you typically write for someone or - increasingly - with someone (Philipp, 2015b: 9, translated by WKG). 20 2 Theory <?page no="21"?> Writing competence is thus a product of the empirical connection between the text product and the text production process (Philipp, 2015b: 33). This is also the case with writing in general. Text types such as invitations, which are less common in everyday professional life, are learned at school or at home. Nevertheless, this skill can also be used in everyday situations, such as planning a birthday party. In the educational standards (KMK, 2004, 2005), German didactics also considers the development of pupils’ writing skills in addition to the promotion of writing skills. This is defined as follows: The term ‘writing development’ describes the development of writing skills as a continuous process ‘after the acquisition’ of basic skills in the first years of school, which are crucial for first reading and first writing. It encompasses the processes and developments that are constitutive for further writing (Bachmann, 2002: 59, translated by WKG). A model for writing development that is still fundamental today is the Bereiter’s writing development model (1980; see Fig. 1). Fig. 1: Writing development model (Bereiter, 1980: 85) Bereiter’s model shows that writing development is not complete beyond the end of school. This writing development model should therefore not be understood as linear, but focuses on various parameters such as the process, the product or the reader and relates them to cognitive developmental writing phases. This orientation initiates various writing modes - such as associative, performative, communicative or epistemic writing. Various cognitive tasks are solved in the process: Cognitively oriented learning and memory research focuses on complex concepts such as perception, problem solving, decision-making behavior and information processing 2.1 Writing skills and their development 21 <?page no="22"?> instead of simple correlations between stimuli and reactions - as in behaviorism (Winkel/ Petermann & Petermann, 2006: 145, translated by WKG). This explains Bereiter’s model in terms of cognitive tasks. For example, the writer can focus on his readership, thereby fulfilling social-cognitive tasks that produce communicatively oriented writing. Bereiter’s model is repeatedly used as a template for writing research, as Fix (2000) has also taken up. Fix (2000: 316, translated by WKG) confirms in his study that in a free writing task, pupils “use epistemic writing functions without practising communicative writing”. “In principle, the individual components are possible at all levels of acquisition, but at different levels” (ibid.). Writing competence is therefore also a form of problem-solving competence, because texts written in the media refer to a situation, a problem that needs to be solved, depending on the communication framework. In this context, the addressee must be addressed appropriately (Antos, 2008: 241). Writing therefore requires many different skills, such as text production and the associated formulation (Antos, 2008: 242). The school writing process begins with an assignment, which indicates the start of a writing process. They control the writing process through a required function that must be fulfilled, addressees and the problem are recognized, world knowledge and linguistic knowledge are acquired, the text should be created in a social context and the effect on readers should be checked, e.g. with the help of writing conferences (Bachmann, 2014: 47). During revisions, the task is revisited. It is the linchpin of all revision steps. The task definition forms the basis of all revisions. The author must interpret the writing task for themselves within their long-term memory and taking into account the task environment and use this to determine a writing goal, which takes on the role of a monitor during the revision process. Without a defined writing goal, the revisions lack a clear line, which can lead to text deterioration (Fix, 2000: 29, translated by WKG). This quote emphasizes that the task is immensely important for writing. The more concrete the task, the more likely the writer can refer back to it and use it as a guide. However, writing tasks also fulfill another function - they can be used to guide writing motivation. Situated tasks, in particular, which have long been used in vocational training through work with learning fields, lead the writer into a specific situation that they must engage with. In this context, Bachmann once introduced the term “tasks with a profile” into writing didactics: ‘Tasks with a profile’ are understood as writing tasks that embed writing as a social practice in a context of social interaction that is as ‘authentic’ as possible by giving 22 2 Theory <?page no="23"?> writing a recognizable meaning for the students and - at product level - allowing it to be questioned and assessed in a plausible way with regard to textual qualities (Bachmann, 2014: 46, translated by WKG). He examined indicators of writing development based on the Bereiter model. In his seven-week project “Tasks with a profile”, he had pupils (n = 78) from the fourth, eighth, and tenth grades write six instructive texts, such as a room description, game sequence, or folding task. “The texts in the sample were examined to determine whether and to what extent the occurrence of coherence-creating elements and the use of cohesive devices in the texts can be indicators of writing development” (Bachmann, 2002: 15; [modified by W.-K. G.]). His sample, comprising 12 test subjects and 36 texts, shows that the fre‐ quency of the use of cohesive elements is not an indicator of the “differentiation of social cognition and writing ability” (Bachmann, 2002: 158). However, a lack of these elements or their terse occurrence is particularly evident in the fourth grade. In Bachmann’s view, this rather indicates the beginning of social cognition. However, when these elements are used, it is attributed more to norm orientation, conventions, and text patterns than to addressee orientation, as the pupils are predominantly still in the norm-oriented writing phase, as can be seen from Bereiter’s model. In the tenth grade, the students show a rather low use of these elements. This is linked to the assumption that tenth graders are more critical than younger writers. According to Bachmann, the distribution of the categories instructing, positioning, differentiating, and referring are reliable indicators of writing development. In all three age groups, it is confirmed that positioning is “more demanding” than instructing, while differentiation is the weakest indicator. The author shows, as Fix already did, that there is no linear writing development for a text type or text pattern: According to Bereiter/ Scardamalia, developed writing ability is expressed in a funda‐ mentally different strategy of text production. Knowledge is not simply retrieved and passed on without further processing. It is transformed in the process of writing (knowledge transforming). This transformation of knowledge requires at least two things: (a) a task representation that anticipates the needs of the reader, and (b) the definition of goals that are to be achieved with the text in the respective context of action […] (Bachmann, 2002: 44, translated by WKG). Bachmann concludes by advocating the use of peer group discussions about text samples so that text sample knowledge can be built up as part of writing competence. Writing conferences should not be perceived as time-consuming; instead, they should be seen as valuable opportunities for students to present their texts to real readers, engage in discussions, and learn from the process. “For 2.1 Writing skills and their development 23 <?page no="24"?> language to be acquired and developed, however, its means, structures, forms and patterns must be internalized, linked to and integrated into individual prior knowledge” (Bachmann, 2002: 85, translated by WKG). In this context, Feilke coined the term “text procedures”: Literal procedures are text routines. They are functionally related to recurrent communicative tasks; these routines are literal insofar as they are typical or […] even specific to communication by means of written texts (Feilke, 2010: 4, translated by WKG). Text procedures are socially schematized actions in texts that are appropriated as tools of writing in the culturally established practices of dealing with texts (Feilke, 2014: 15, translated by WKG). In terms of learning psychology, the term is derived from “procedural learning” (Winkel/ Petermann & Petermann, 2006: 219, translated by WKG): “In procedural learning, cognitive or motor skills are acquired step by step through practice by gradually replacing conscious processes with unconscious processes” (ibid.). With regard to the profiled writing tasks, he states: “In this case, task schemata refer to text procedures as action schemata available in the competence, which provide textual solution alternatives for writing” (Feilke, 2014: 17, translated by WKG). Thus, the task itself should not only be placed on the control level, but should also be recognized as a resource, because “competent writers have text procedural action schemata to which they can assign differentiated inventories of textual structure patterns, syntactic constructions and lexical collocations” (Feilke, 2014: 19 f., translated by WKG). For this reason, German didactics is also interested in text procedures. Feilke equates the routine approach to writing with the term writing strategy (Feilke, 2014: 20). Text procedures, however, refer to the text itself: When it comes to a letter of application, even the normative specification of sample templates is legitimate because it is associated with social opportunities; the conven‐ tions are then rules of the game that identify the writer as a connoisseur of the rules and thus a potential player (Fix, 2000: 329, translated by WKG). Fix explicitly advocates the use of sample text modules and sample texts (Fix, 2000: 339), which are common in professional contexts and support the text procedure (Wergen & Wörner, 2005). At this point, declarative and implicit learning (see Fig. 2) should be discussed in more detail in order to understand the relevance of patterns in the form of text modules and texts. Figure 2 therefore visualizes the learning processes and outcomes. 24 2 Theory <?page no="25"?> Fig. 2: Learning processes (Winkel/ Petermann & Petermann, 2006: 229, translated using AI) Text pattern knowledge and text type norms can be classified as declarative knowledge (‘knowing that’), as this is part of world or factual knowledge. “Procedural learning refers to the acquisition of motor and cognitive skills” (Winkel/ Petermann & Petermann, 2006: 16). Implicit learning happens uncon‐ sciously and is an integral part of the learning process (ibid.). Figure 2 illustrates that strategies can be learned both implicitly and declaratively through practice and proceduralization. For teaching, this means that rule learning also occurs implicitly through the teaching context. However, text pattern knowledge alone does not characterize a successful writer (Fix, 2000: 333). Young writers tend to write texts “just-in-time” and in “flow”. They tend to write linearly, sentence by sentence, and find it difficult to deviate from this strategy (Ortner, 2013: 69 f.). Writing strategies are therefore relevant parameters of writing competence and solution aids for the task at hand, as Nitz (2010) highlights in his study: 2.1 Writing skills and their development 25 <?page no="26"?> Nitz’s (2010: 123 f.) study on the ability of third and fourth graders to revise texts in writing conferences shows that even fourth graders demonstrate selectiveness in accepting feedback to improve texts. The ‘author’ decides what to incorporate. Writing conferences facilitate text revision, particularly with regard to coherence. However, not all difficulties can be eliminated. This cannot be achieved by the teacher’s suggestion alone, rather it is a highly emotional matter whether advice is accepted or not. The who and how are also decisive: “In the writing conferences, the consultants show themselves to be extremely critical readers and in many cases help the texts to become more coherent through intensive discussions about certain text passages” (Nitz, 2010: 124, translated by WKG). “The ‘authors’ are usually very committed to the quality of their texts, of which they are obviously proud and which they want to bring to a satisfactory conclusion” (Nitz, 2010: 125, translated by WKG). Coherence problems could often be solved, even if Fix (2000) interestingly rates writing conferences as less successful for writing success in his study. Nevertheless, the communication of model knowledge is recognized as part of a writing strategy: Since pattern knowledge can also be used as a writing strategy, it can also be ‘knowing how’, which is either consciously used to deal with the problem situation (problem-solving knowledge) or is already automated (procedural knowledge). If, in the case of communicative writing, text is understood as a macro speech act, this knowledge is applied as illocutionary knowledge with the aim of writing in a functionally appropriate way for an addressee. In traditional essay teaching, the declarative component of this prior knowledge has been overemphasized (Fix, 2000: 332, translated by WKG). It is clear that writing skills can be promoted and guided in various ways. The task and the writing strategy are important influencing factors for writing competence. This realization has has been incorporated into the framework curricula for writing in German. The challenge here is to teach writing to a mixed-ability class. Particularly at elementary school, secondary schools, com‐ prehensive schools, and similar institutions, content, process-related and social competence goals have been formulated in favor of at least three competence levels. It should be noted that additional competence levels have also been formulated for pupils with integration status. This means that German teachers can teach a heterogeneous and inclusive student body in their German classes, which often requires them to address tasks across four competence levels (minimum, regular, expert standard and at least one level for integration pupils). 26 2 Theory <?page no="27"?> In addition to using the competence level model, it is advisable to provide ad hoc criteria for analysis for the respective writing task and then relate these back to the competence descriptions […] (Eichler, 2017: 151, translated by WKG). Some study seminars, e.g., in Verden (Lower Saxony), prepare very precisely for the formulation of competence levels and their evaluation during the traineeship before the second state examination. The formulation of competencies by Kessler and Ziener (2004), shown in Figure 3, is used for this purpose. Kessler and Ziener have identified four requirement areas for the formulation of com‐ petencies for all subjects with reference to the educational standards: Category I “Perceiving, knowing and understanding” (Kessler & Ziener, 2004: 1, translated by WKG) is assigned to the cognitive area, category II “Speaking and providing information” (ibid.) to the communicative area, category III “Developing and designing” (ibid.) to the methodological-creative area and category IV “Planning and working together” (ibid.) to the personal and social area. Fig. 3: Competence grid according to Kessler & Ziener (2004, own illustration) Both co-authors of the Baden-Württemberg educational standards examined the academic standards for this categorization and divided all operators (also known as predicates) into these four categories (ibid.). Each area of competence is then 2.1 Writing skills and their development 27 <?page no="28"?> divided into three levels, with requirement area A being the lowest level and C the highest level (ibid.; Ziener & Kessler, 2012: 29). It is similar to the classifica‐ tion according to the minimum standard, regular standard and expert standard. The model aims to enable pupils to learn in a competence-oriented, cooperative, and self-organized manner (Giera, 2014: 5 f.). Regarding this interdisciplinary competence grid, a competence level model for the area of writing was developed in Germany at the Institute for Quality Development in Education (IQB) (Böhme et al., 2017: 57 ff.). The basis for this was the Intermediate School Leaving Certificate at secondary level I and the ducational standards of the Standing Conference of the Ministers of Education and Cultural Affairs for the subject of German from 2004. The idea of the levels emerged from the competence level models of the science subjects. The standardization was based on a sample of 2,996 pupils from the ninth and tenth grades in 2011 with twelve developed writing tasks. For each task, which was completed in 20 minutes, 500 student texts are available. The texts were first rated on a holistic five-point global scale. The task difficulty was then estimated, taking into account background variables such as age. This resulted in threshold values for the individual levels and, ultimately, five levels of equal width. These individual levels were described in more detail and labeled as levels - from minimum standard failed, minimum standard, regular standard, regular standard plus to optimal standard. These standards were precisely described and scored as competence levels for the argumentative texts. As a result, it could be stated: “However, around a quarter of ninth graders aiming for the MSA or higher do not yet meet the expectations formulated in the educational standards of the KMK (regular standard)” (Böhme et al., 2017: 69, translated by WKG). There is another study worth highlighting that empirically investigated the development of pupils’ writing skills, not cross-sectionally but longitudinally; to test text type competence in pupils’ writing genesis, Augst, Disselhoff, Heinrich, Pohl and Völzing (2007) conducted a long-term study of primary school pupils with the text patterns narrative, report, instruction, description, and argumentation over several years. The question of text type development was the focus of this study, with 39 pupils and 585 pupil texts, as Bereiter’s writing development model had not yet been empirically tested. Each pupil wrote 15 texts per year over a period of three years: a narrative based on a picture stimulus, a report on Christmas, instructions on a favorite sport, a description of the classroom, and an argumentation with a proposal. Over the years, it was observed that primary school pupils wrote longer and more complex texts. The variety of sentence structures increased. At the end of fourth grade, narration 28 2 Theory <?page no="29"?> is the most successful. Describing and arguing and, finally, reporting remain in the middle range: In any case, it is important for the children to experience the work on the text as something natural and self-evident that is inherent to the writing process. […] This is why children’s discussions about texts (or even a writing conference) can be very beneficial; this is also for an ontogenetic reason: at their respective developmental stage, the children will question and stimulate the ‘writing child’ in a more experiential (child-like) way than a teacher might be able to in view of the normative pattern (Augst et al., 2007: 359 f., translated by WKG). The research team states that all text types in elementary school are in demand for developing writing skills. It is striking that a recognized structural feature in one type of text is transferable to another (Augst et al., 2007: 358). “Producing a text with a certain text quality is therefore the result, i.e., an expression, of the individual writing competence of writers” (Schmitt & Knopp, 2017: 239, translated by WKG). This excursus shows that pupils at one grade level are competent in writing at different levels. The competence grids with various level descriptions must increasingly take priority when planning writing lessons. A learning starting point is absolutely essential. The results of the study showed that writing competence can vary depending on the type of text. To measure and promote writing skills in the classes, several tasks would have to be set and tested for each type of text (Böhme et al., 2017: 70 ff.). The resulting writing products are indicators of writing competence. Furthermore, the writer, the writing process, the writing procedure, and the potential reader are parameters of holistic writing competence measurement and promotion. This complexity is reflected in a summary by Feilke (2017: 158), which subsumes didactic writing impulses relevant to teachers and writing researchers. The basis of Feilke’s “parameters of writing didactic concepts” (Feilke, 2017: 158, translated by WKG) is the writing development model according to Bereiter (1980), but the levels of the process, reader, and product have been supplemented by the levels of “procedure” (ibid.) and “writer” (ibid.) (Feilke, 2017: 157, translated by WKG). The number of writing levels shows how complex the parameters of writing competence development in writing didactics are. The combination of goals, task types, task modes, and writing didactic concepts alone requires extensive research work, and the creation of free space for many lessons in German. Consequently, several factors influence writing competence and thus text quality and can act as predictors, as the following definition and the associated Figure 4 illustrate and, in my opinion, summarize well, as both the basic writing 2.1 Writing skills and their development 29 <?page no="30"?> models and the associated predictors are related. The areas of volition, motiva‐ tion and metacognition are linked to writing motivation and self-regulation, which are fundamental to this work: As the results presented show, writing competence is a (highly) complex construct, which is made up of numerous sub-skills from different skill areas (constitutional, cognitive-skill-related, motivational-volitional and metacognitive as well as specifi‐ cally language-related areas) that interact in an ensemble-like manner (Schmitt & Knopp, 2017: 248, translated by WKG). Fig. 4: Overview of candidate predictors (Schmitt & Knopp, 2017: 241, translated using AI) 30 2 Theory <?page no="31"?> Writing skills are still primarily taught in the subject of German. However, Norway shows that the essential skill of writing should be anchored in the curric‐ ulum in all subjects: “Every teacher, regardless of subject, is now responsible for the teaching of writing” (Berge et al., 2016: 180). To this end, the Scandinavians developed the theory-based writing model “The Wheel of Writing” in two versions (Berge et al., 2016). All rings can be moved in relation to each other (see Fig. 5 and 6). Figure 5 is read as follows: The outer ring focuses on the six writing or language acts: to interact, reflect, describe, explore, imagine and convince. The middle ring focuses on six writing purposes: persuasion and knowledge development are just two examples. The wheel is, in turn, held together by semiotic mediation. It is exposed to and influenced by cultural and situational contexts. In Figure 6 of the Wheel of Writing, the outer and inner rings are the same, while the middle ring focuses on four tools and resources for writing. Fig. 5: Wheel of Writing I (Berge et al., 2016: 180) 2.1 Writing skills and their development 31 <?page no="32"?> Fig. 6: Wheel of Writing II (Berge et al., 2016: 182) Modalities include the letters, and the illustrations, pictograms, tables and images. Text structure refers to the local and global means of cohesion. Writing tools include writing media, such as pen or PC, and others. Vocabulary and grammar refer to vocabulary on the one hand and grammar on the other. It also includes syntax, which is more complex in written than spoken language. To summarize, the Wheel of Writing starts with the writer himself. The writer aims at a writing action, which could also be equated with the writing goal. To achieve this writing goal, the writer has various personal, linguistic, and media resources at his disposal, which he uses to fulfill a writing purpose for himslef and in the interest of a circle of readers. For this chapter, competencies mean secure knowledge and skills in problem‐ atic situations. They can be functionally classified either as an objective or as an operationalization. Writing competence is linked to the writing tasks and the associated writing occasions, as presented by the tasks with profile and the Wheel of Writing. However, Schmitt and Knopp’s predictors emphasize that writing competence is influenced by other indicators such as working memory, reading ability, perspective taking, writing motivation, self-regulation, writing fluency, text pattern knowledge, vocabulary, and coherence production, among others. Riener’s model makes it clear that writing development has various phases, 32 2 Theory <?page no="33"?> providing a rough orientation. Still, the student must be considered alone with his or her writing genesis and procedural and declarative writing knowledge. Especially after leaving school, former pupils develop an “ego-development” (Koch & Pielow, 1984: 82, translated by WKG): Writing frustration can turn into a desire to write, which is a motivating factor for the genesis of writing competence. If writing competence is to be supported, Feilke shows in his list that this can be promoted through different tasks, working modes and writing goals. Although the definitions by Efing and Neumann provide definitions of the term competence, the term writing competence is very complex, as this subchapter makes clear. For this reason, the following two sub-chapters will deal with writing competence before and during vocational training in greater depth and also show what writing didactics recommends for teaching, which could only be touched on in Section 2.1. 2.2 WRITING SKILLS BEFORE VOCATIONAL TRAINING AT THE END OF LOWER SECONDARY LEVEL “Those who cannot write in school language are at a disadvantage in the assessment in all subjects - even if the answers are correct in terms of content - also because the proportion of written work in the overall assessment is extremely high” (Neumann, 2010: 14, transla‐ ted by WKG). Success at school is linked to an “[…] adequate command of the German language. It is the medium of communication in and with the majority society” (Steinmüller, 2006: 322). Writing, along with reading and arithmetic, are basic requirements and competencies that are institution-bound and essential for so-called educational maturity (Neumann, 2010: 13). These skills are the entry ticket to upper secondary level for school-leavers of all school types, and thus, is necessary for many pupils to vocational training. According to the Federal Employment Agency, both writing competence and text type competence (see Section 2.4) are prerequisites for employment, which requires the following criteria for the characteristic of (legal) writing (Federal Employment Agency, 2006: 22, see Fig. 7): 2.2 Writing skills before vocational training at the end of lower secondary level 33 <?page no="34"?> Feature: (Correct) Writing Description: Adolescents can write simple texts without errors and for‐ mulate them clearly. Indicators/ Criteria: He/ she writes texts in legible handwritten form. He/ she knows the basic rules of spelling and punctuation and can apply them. He/ she can spell frequently used words correctly. He/ she can write texts that are clear and coherent. He/ she can produce standardized texts, e.g., letters, résumés, cover letters, and fill out forms. He/ she structures texts appropriately for the purpose and audience, with a logical and coherent structure. (Federal Employment Agency, 2006: 22, translated by WKG) Fig. 7: Criteria catalog for training maturity (Federal Employment Agency, 2006: 22, translated by WKG) The aforementioned criteria are taught in German lessons on the basis of the framework curricula starting from the first grade until the end of school. The subject of German teaches four basic skills included in the educational standards for the intermediate school leaving certificate. Among these, writing is one of the four skills and represents a basic qualification (KMK, 2004: 8): Pupils are familiar with the diverse possibilities of writing as a means of communica‐ tion, presentation and reflection and write texts that are appropriate for the target audience (KMK, 2004: 8, translated by WKG). The focus on action in writing and process-oriented German didactics in the area of writing is also required by the Standing Conference of the Ministers of Education and Cultural Affairs of the Länder in the Federal Republic of Germany: “have writing skills” (KMK, 2004: 11, translated by WKG), “write correctly” (ibid.), “design a writing process independently” (KMK, 2004: 12), “write texts” (ibid.) and “revise texts” (KMK, 2004: 13). At this point, it becomes unmistakably clear that the writing process and not the writing product alone is the subject of instruction in German lessons (Becker-Mrotzek & Böttcher, 2006: 24 f., translated by WKG). The core curriculum for German in Lower Saxony addresses these skills. The curriculum for writing lessons in schools today is based on the original writing model by Hayes and Flower (1980) (see Section 2.5). In this model, as in the more recent model by Hayes (2012), the writing process is broken down into phases of planning, writing and revising. These steps should also be made visible in the context of German lessons (Becker-Mrotzek & Böttcher, 2006: 24 f.; Bachmann & Becker-Mrotzek, 2010: 195). Knowledge of schemata is an influencing factor for the development of writing skills and the associated strategies and procedures. Feilke and Bachmann (2014: 8) point out that there are “textual action schemata” for this. These 34 2 Theory <?page no="35"?> text procedures are therefore taught and learned within the school context. In German lessons, in particular, the training of recurring procedures for writing texts should be taught, as they lead to a writing routine (Feilke, 2010: 1, see Section 2.1). A systematic approach involves first recognizing the type of text referred to in the writing task and classifying the “text action type” (Feilke, 2014: 26, translated by WKG), e.g. arguing. This is followed by certain “action patterns” (Feilke, 2014: 23, translated by WKG), such as reasoning. Finally, the incorporation of “procedural expressions” (Feilke, 2014: 26, translated by WKG), e.g., “I think that” (ibid.), should be implemented as a linguistic textual action. Thus, writing lessons require “genre-specific approaches” (ibid.). The use of text modules can help support this: “Text modules as action units can be recognized and distinguished in class and group discussions by relevant procedural expressions” (Feilke, 2014: 28, translated by WKG). Further reasons for this approach are explained in the following quotes: The central argument for this approach is that the text modules are relatively independent units from which texts of a text type can be composed in different ways, as the word ‘composition’ aptly expresses. While the traditional didactics of school text types fixes global text patterns in a normative way and encourages adaptation to and imitation of these patterns, focusing on the middle level of text modules offers the opportunity to develop text creativity and a functional understanding of the linguistic nature of texts. This means that in such an approach, writing tasks are no longer only related to whole texts to be written in a communication context, but also preferably to work on text modules (Feilke, 2014: 28, translated by WKG). The degree of orientation towards characteristics of text types - whether a writer relies more on content-, addresseeor text type-related writing strategies - is more closely related to the specific writing function. Someone writing a cover letter for a job application is more likely to follow a template and work more closely with standard specifications than someone writing a private letter, although the focus on the addressee is also decisive in both writing tasks (Fix, 2000: 49, translated by WKG). In addition to text modules, other didactic writing measures are used: Text planning should lead to developing an explicit writing plan based on a collec‐ tion of material and the generation of ideas. Text features are taught using texts produced on this basis. Teaching explicit text type knowledge relieves the working memory when writing, as existing text routines can be used. Routines/ procedures develop through everyday situations and same action sequences (Djik, 2008: 68). This development of skills continues to evolve throughout writing development. 2.2 Writing skills before vocational training at the end of lower secondary level 35 <?page no="36"?> In the context of text revision, the “writing conference” (Spitta, 1992: 13, translated by WKG), the use of criteria lists, and other methods to support a feedback process are required by the curriculum. The writing conference method is frequently used in the German core curriculum and should, therefore, be explained briefly so as not to confuse the term with the peer group discussions in the writing project to be presented here: the term originally goes back to the research work of Graves (1983). Classmates first read the texts, and then give feedback. Spitta adapted this procedure in Germany and advocates it because the quality of the text is improved by the revision instructions (Spitta, 1992: 13). The exchange between pupils about their texts is important, as this leads to an “approximation of literary norms when revising” (Dehn et al., 2011: 89, translated by WKG). Although the use of criteria lists restricts a text type as a “prototypical concept” (Adamzik, 2004: 47, translated by WKG), it should also provide impulses for revision to improve text quality. Furthermore, criteria lists can be used as discussion guidelines in cooperative writing settings for objective feedback and thus, as in the context of quality management in companies, can also be used profitably for text quality: Cooperative writing is about contexts and arrangements in which two or more people work together to produce a text. The collaboration can relate to the joint planning, formulation and revision of a text. However, it can also be limited to providing feedback on a text that has not been produced by the students themselves, for which ideas and suggestions for optimization are formulated (Lehnen, 2017: 299, translated by WKG). The lists of criteria should not be rigid but should be reviewed repeatedly and supplemented if necessary. Therefore, criteria lists are also the basis for discussion per se. Feedback discussions are always challenging for those involved, who usually react emotionally. Objective and non-personal feedback should, therefore, be practiced at school, as it requires feedback givers and feedback receivers to adhere to conversation rules, such as finding the appropriate tone, taking into account the change of speaker, and starting and ending the conversation appropriately (Wagner, 2003: 754). Furthermore, the use of feedback in the classroom is time-consuming. It requires cognitive and linguistic skills from classmates, which nevertheless positively influence the quality of the text (Lehnen, 2017: 299). This makes it clear that the writing process, based on the writing process model by Hayes and Flower (1980), is divided and taught in the 36 2 Theory <?page no="37"?> sub-areas of planning, writing, and revising. The didactic integration of text feedback from classmates is a common procedure in writing lessons. In his research on text revision by eighth graders (n = 63), Fix (2000) was able to prove empirically that the revision phase in particular is very complex in the classroom, as the students often do not want to muster the motivation for it: Difficulties in social interaction with each other made joint revision in writing conferences difficult, and listening to each other and discussing fairly did not yet work. There were also major differences in the time required by the individual writers, which is why a variety of differentiation options had to be organized. In particular, rewriting by hand (as working on the computer was only possible to a very limited extent) was reluctantly accepted (Fix, 2000: 6, translated by WKG). When revising, the writer will reread their text and the task again if they intend to revise the text. Revision can only succeed with motivation, which is why the role of the writing task regarding revision is also important (Fix, 2000: 45). The following quote also shows how complex revision is despite all the methodological and didactic recommendations: The study has shown that revising texts at school makes didactic sense, leads to text optimization in the majority of cases, does not level out performance differences between pupils and is generally well received by pupils […]. At the same time, difficulties became apparent: the ability to diagnose problems is weak in many students, they did not recognize deeper problems in their texts - even despite working together in writing conferences - they lacked suitable revision strategies and executive skills, the revisions carried out often referred only to local corrections and to text type norms, whereby there were clear differences between the two writing tasks (Fix, 2000: 320, translated by WKG). One of the reasons for the complexity of revision is the language itself, as the language of schooling is caught between the language of education, written language and other language fields (Feilke, 2012a, see Fig. 8, translated by WKG). The written language influences the language fields at school. On the other hand, external norms such as those of literary language and academic language also influence the demands placed on written language. In this context, Feilke advocates the “transit norms thesis” (Feilke, 2012b: 155, translated by WKG). He argues that the language of schooling is in itself an area of demand that a pupil has to meet, and thus presents them with a barrier to be overcome when writing: In this respect, language norms for didactic use are transitory norms. In the sense of a social constructivist approach, they are not intended as a goal, but as a supportive 2.2 Writing skills before vocational training at the end of lower secondary level 37 <?page no="38"?> means (scaffolding) of competence development (Feilke, 2012b: 155, translated by WKG). It should be noted here that the pupil cannot and must not simply write a text in his or her language, but that the pupil is in the context of his or her school, class, teacher, and subject. Writing lessons must, therefore, be guided from the pupil’s perspective to accompany them appropriately. Fig. 8: Language fields at school (after Feilke, 2012a: 2011: 6, simplified representation, translated by WKG) The “communicative turn” brought about in the 1960s (Eichler & Henzel, 1998: 105, translated by WKG) called for a focus on the surrounding situation, which was increasingly pursued in language didactics in the 1980s. Lessons, and thus also German lessons, should integrate a focus on action and pupils as characteristics of good didactics (Eichler & Henzel, 1998: 109), which is now included in the lesson plans for state examination papers in teacher training, but also in the core curricula for German. A further requirement in addition to the methods presented in writing didactics is the use of new media. The age of new media has also arrived in schools. Working with computers has long been a requirement in school curricula and, therefore, in the classroom. The core curriculum for German in Lower Saxony requires the use of new media, whether it is creating PowerPoint presentations, creating tables, researching online or writing with WORD or other programs. It should be emphasized that although the PC is often used as a research tool in all subjects, writing business letters at the end of lower secondary level is most likely only carried out by 38 2 Theory <?page no="39"?> writing application letters. At some schools, this is also anchored in business lessons and the associated vocational orientation (see Section 2.4). At many schools, students complete a work placement in the ninth and/ or tenth grade. For this, the students write an application with a cover letter and CV. It is the first situation in which the student applies for a job and gets a taste of the working world. Some companies are in such high demand that the list of applicants is very long and an initial selection of internship candidates is required regarding the application documents. Other types of text that prepare students for professional writing include the protocol, process description, ob‐ ject description, invitation, form, writing plan, application, apology, complaint, test procedure, description, or loan bill (ibid.). This is just a selection of school text types that can also be described as “trans‐ fer text types” (Giera, 2010: 40, translated by WKG), as they prepare students for the transition into the world of work and the writing requirements there. Through these school-based “text forms” (Pohl & Steinhoff, 2010: 6), students acquire text-type knowledge that they can transfer to professional writing and the text types anchored there. Although professional text types such as the offer or the complaint are not yet anchored in the school curriculum, argumentation can generally be regarded as an important competence requirement at lower secondary level. In addition, it must be pointed out that children can argue orally from an early age. In doing so, they defend their position in order to achieve certain outcomes such as consuming sweets or staying up late. However, writing arguments is a challenge for secondary school pupils, and it is regularly required in the MSA examination in the tenth grade in the form of a dialectical discussion. Argumentation as a communicative competence has its origins in classical rhetoric, in which the speech is first prepared in writing and then presented to a public audience (Ossner, 2006: 75). In the school context, this is prepared in writing with the help of linear and dialectical discussions. In these, students develop pro and contra arguments and support them with examples and evidence. To do this, sub-arguments must be related, linked, and weighted. In the ninth and tenth grades, a discussion question and the associated thesis statement are also taught. A dialectical discussion, which includes the thesis, antithesis, and synthesis, is written for this purpose. In the context of writing lessons, the discussions, especially the dialectical ones, represent a challenging writing task. Difficult writing tasks are characterized by the fact that the texts only fulfill their function when what is known is changed for presentation under a certain perspective. The logic of what is known must be changed with a view to the communicative function (Becker-Mrotzek & Böttcher, 2006: 61, translated by WKG). 2.2 Writing skills before vocational training at the end of lower secondary level 39 <?page no="40"?> 3 Brinker (2010) states that several of the five basic functions can play a role for a text type, but one usually dominates (Brinker, 2010: 78). The primary communicative function of argumentation is an “appeal function” (Brinker, 2010: 98, translated by WKG), 3 in which the addressees are convinced by a well-founded fact and called upon and influenced to take action (Hei‐ nemann & Heinemann, 2002: 187; Brinker, 2010: 101). This function is also congruent with application letters and unsolicited offers. For example, consumers and HR managers are to be convinced of a product, a service or of the person applying, which is expressed by the appeal function of both text-types. The applicants appeal to invite them, and the consumers are called upon to buy the service or product in the case of unsolicited offers. Therefore, these text types also contain arguments that speak in favor of the service or person. Students can thus draw on the skills of argumentation when writing letters of application or offers. This subchapter explains the importance of written argumentation in German lessons at the end of lower secondary level. It shows the challenges of written language in the school environment and the associated writing difficulties for pupils. By formulating the writing task, the German teacher has the opportunity to promote writing motivation, but must pay attention to the heterogeneity of their writers. Therefore, sample texts, lists of criteria, text modules, writing conferences and other feedback procedures should be used as support measures in writing lessons to accompany the writing process of all pupils and further develop their writing skills. 2.3 WRITING SKILLS IN DUAL TRAINING The research field of writing at work is gaining in importance with the predicted transition from the information society to the knowledge and service society and - more broadly - to the health care society. Each of the three development trends demands considerable communicative skills from those practicing the profession ( Jakobs, 2005: 15, translated by WKG). The following chapter elaborates the writing requirements in dual vocational training, and is first enriched with facts about dual vocational training in Germany. 40 2 Theory <?page no="41"?> 2.3.1 DUAL VOCATIONAL TRAINING IN GERMANY Politics influence new social challenges and requirements. Vocational education and training are closely linked to social and economic policy (Leischner, 1993: 7). Within the European Union, the EC 1992 Art. 127 makes vocational education and training a policy area and an important area of responsibility for all EU countries: “While EU countries are responsible for their own education and training systems, the EU supports them in setting common objectives and exchanging best practice” (European Union, 2018, translated by WKG). The EU countries should learn from each other in terms of harmonization. In 1993, guidelines for lifelong learning were also developed (Piehl & Sellin, 1995: 441 ff.; see Chapter 1). Education policy in Europe calls for reforms in the area of vocational training. In vocational education and training, there is a European strand of action with the following objectives, which has also been transferred to vocational education and training in Germany: Facilitating adaptation to industrial change, in particular through vocational training and retraining; improving initial and continuing vocational training to facilitate vocational integration and reintegration into the labor market; facilitating access to vocational training […] and promoting and cooperating on vocational training issues between educational establishments and enterprises; developing the exchange of information and experience on common problems within the vocational training systems of the Member States (Leischner, 1993: 10, translated by WKG). A comparison of school periods in the lower and upper secondary education sectors shows very different school systems in Europe: Lower secondary education lasts three years (as in Belgium, Italy, and Ireland) and six years (in some federal states in Germany). The upper secondary level lasts two years (as in Great Britain) or up to five years (in Italy and occasionally in Spain) (Leischner, 1993: 11). In principle, it can be stated that pupils in Europe predominantly attend a comprehensive or unitary school until the end of lower secondary level, in complete contrast to Germany, which for a very long time adhered to the tripartite structure of Hauptschule, Realschule, and Gymnasium, whereby the special schools should be added as a fourth branch (Leischner, 1993: 15). The majority of federal states no longer separate secondary schools into Hauptschulen and Realschulen, but have developed new school types with course systems for subject differentiation. As a result of this school development, the gymnasiale Oberstufe is not only offered at Gymnasien but also Gesamt‐ schulen, Oberschulen, or vocational schools. The term “school type” changes depending on the federal state. The fact is, however, that the general higher 2.3 Writing skills in dual training 41 <?page no="42"?> education entrance qualification can no longer only be achieved at grammar schools. It can also be stated that the German education system is divided into primary education (grades 1-4/ 6), lower secondary education (grades 5/ 7-10) at different types of schools, upper secondary education (from grade 11 at various types of schools such as grammar schools or vocational training at vocational schools) and tertiary education, i.e., university or technical (higher) education. In federally organized Germany, education is a matter for the federal states, so the number of years in the various school sectors and the names and types of schools differ from state to state. In Lower Saxony, primary school pupils move on to secondary school from the fifth grade onwards (Baethge, 2008; IHK Leipzig, 2011). In the year of the study (2013), children in Lower Saxony could transfer to grammar school, lower secondary school, intermediate secondary school, comprehensive school, or upper secondary school, depending on their performance. However, even then, the more homogeneous Realschule and Hauptschule were “outgrowths” and were replaced by Oberschulen (Secondary Schools) or Gesamtschulen (Comprehensive Schools). In both types of schools, after an initial orientation phase, German lessons are organized homogeneously in courses at the Realschule (Secondary Schools), Hauptschule (Middle Schools), or Gymnasium (Grammar Schools). This educational policy measure aims to increase the educational opportu‐ nities of all children, in contrast to the stigmatization of lower secondary school pupils that has persisted for years. Tenth graders from all school types write the final German, Mathematics, and English examinations for the central Intermediate School-Leaving Certificate (MSA) in all federal states. The skills required in these exams are intended to enable pupils to participate in society. Vocational school classes include pupils from all types of schools. As a result, vocational school classes are very heterogeneous in performance. Nevertheless, secondary general and secondary modern school leavers are more likely to pursue dual vocational training than grammar school pupils, most of whom focus on studying. Vocational education in Germany at the upper secondary level shows that Germany currently offers predominantly dual training. In contrast, most other European countries provide full-time vocational schools or full-company training (Leischner, 1993: 15): The ‘dual system’ or ‘dual vocational training’ refers to a form of training that includes training in a company and instruction at a vocational school. The common goal of the company and the vocational school is to impart the knowledge and skills required for the occupation to be learned (Leischner, 1993: 15, translated by WKG). 42 2 Theory <?page no="43"?> Countries with predominantly company-based training are Germany, Denmark, Austria, and Switzerland. Conversely, those with school-based training are France, Belgium, Italy, the Netherlands, Finland, Sweden, Norway, and the countries of the former Soviet Union. Company-based training is still in place in the UK, Luxembourg, and Ireland (Schaper et al., 2000: 262). Germany stands out in Europe due to its dual training system: In connection with the development of the common European internal market, the dual system is particularly emphasized in comparison with the education systems of other European countries and is even described as an ‘export hit’ of the Federal Republic of Germany (Leischner, 1993: 36, translated by WKG). As early as the Middle Ages, there were rules for vocational training in Germany through guilds. Still minimum legal standards were only introduced in 1869 by the Industrial Code and in 1897 by the Commercial Code. The chambers received legal legitimacy to regulate vocational training in 1897 through the Prussian law. In 1930, the chambers were given even more self-administration for vocational training. In 1970, vocational training was transferred to the state as a task (Leischner, 1993: 28 f.). The dual system has changed considerably over the last 200 years. Still, one part has remained - in-company training and vocational school teaching (Schaper et al., 2000: 26), which are also legally enshrined in the Basic Law: According to the Basic Law of the Federal Republic of Germany and the corresponding constitutions of the federal states, the entire school system is under the supervision of the state. Schools and universities are generally state institutions of the federal states. In the area of vocational education, the federal government is responsible for in-company training, while the Länder are responsible for the vocational school system (Leischner, 1993: 19, translated by WKG). Vocational training in Germany is also a constantly changing process, as theory and practice are coordinated and influenced from outside. The interest groups of the various dual partners are fundamental to this, namely “training planning by industry and education planning by the state” (Leischner, 1993: 36, translated by WKG). One challenge is maintaining the high quality of dual vocational training. Initial vocational training and retraining are controlled at the macro level of didactic planning via curricula that are developed centrally by committees of the Federal Institute for Vocational Education and Training. For the company-based part of training, these consist of the training regulations together with the oc‐ cupational profile and training framework plan; for the vocational school-based 2.3 Writing skills in dual training 43 <?page no="44"?> part, they consist of the framework curriculum for the four core theoretical subjects (Reetz & Seyd, 1995: 205, translated by WKG). In addition to the company and the school, three educational policy stake‐ holders influence dual training: the Federal Ministry of Justice, the Standing Conference of the Ministers of Education, and Cultural Affairs of the Counties and the Chambers. If all three education policy actors are summarized with their characteristic descrip‐ tions for vocational education and training in Germany, the following main tasks emerge: The Federal Ministry of Justice accredits the training occupations, the Standing Conference implements the framework curricula for vocational education and training in the vocational schools and the Chamber of Industry and Commerce is responsible for supervising the training companies and taking the examinations (Giera, 2010: 19, translated by WKG). The cooperation and interaction of all three education policy stakeholders is a prerequisite for the quality and structure of dual training in Germany. However, two further cooperation partners are decisive for the implementation: Vocational schools in Germany are to impart theory, and companies are to impart practice. Both cooperation partners focus on the ability to act professionally (§ 14 BBiG, 2005, see Fig. 9). This is referred to as dual training, as both the training company and the vocational school are cooperative partners during a trainee’s training period. While the training company tends to be responsible for practically teaching skills and abilities in the respective occupational profile, the vocational school focuses on the theoretical teaching of the subject (see Fig. 9). Fig. 9: Cooperation partners for dual training (own illustration) Vocational school attendance is compulsory as part of dual training, even if the trainees are of legal age. This is to ensure the professional quality of the training. Trainees are exempt from working in the company when they attend vocational 44 2 Theory <?page no="45"?> school (§ 15 BBiG, 2005). Apprentices attend vocational school once or twice a week and work at the training company for three to four days. Alternatively, block courses can be held. The teaching days at the vocational school are compressed, and pure school weeks take place at regular intervals. In this case, there are pure vocational school weeks and pure work weeks in the company. Vocational school lessons are divided into three phases by the provisions of § 26 BBiG: The first year comprises the basic training phase (basic skills and basic knowledge), followed by the general vocational training phase and, in the third year, the placement phase. Since the 1970s, the curriculum in vocational education has been character‐ ized by the guiding concept of action-oriented vocational education (Arnold et al., 1995: 13). To this day, action orientation has a special place in teaching at vocational schools. In retrospect, the history of vocational education as early as the Middle Ages shows that acting and doing during the training process are nothing new (Czycholl & Ebner, 1995: 41). Rather, this term is a reminder of the need to create learning arrangements in vocational training that “[…] enable authentic, simulated and/ or symbolically represented work actions” (Czycholl & Ebner, 1995: 40, translated by WKG). The teacher at vocational schools is seen more as a learning consultant. They are responsible for self-directed and exploratory learning to develop and try out solutions. In this role, the teacher has a high level of responsibility for planning but must exercise restraint during implementation (Pätzold, 1995: 169). The actions should then be evaluated in a reflection phase. This procedure combines thinking, ability, and knowledge (Pätzold, 1995: 168). Pätzold (1995: 168) recommends the following methods: problem-oriented learning, discovery learning, situational learning, modeling, project learning and learning in groups, simulation games, experiments, and explorations. In companies, the demonstration of specific skills and imitation with several practice phases and verbal repetition are standard (Pätzold, 1995: 164). In contrast to vocational school, learning by discovery can be a risk if it disrupts operational processes. Demonstrating new work processes requires the trainee to pay close attention and awaken their senses through hearing and seeing. The verbal repetition memorizes each work step, and the imitation itself thus complements learning with all the senses. “Frequent repetition of the learning process leads to the learning outcome - i.e., the newly learned behaviour - being stabilized” (Winkel/ Petermann & Petermann, 2006: 119, translated by WKG). Practical instruction in the workplace is carried out using various methods, including the four-step method developed by Ch. R. Allen (USA) in 1919, which 2.3 Writing skills in dual training 45 <?page no="46"?> is still used today, especially in smaller craft businesses. The four steps of instruction are as follows: “1. preparation of the trainee and the workplace, 2. demonstration of the training content by the trainer, 3. imitation of the training content by the trainee, 4. practice of the acquired skills by the trainee” (Schaper et al., 2000: 123, translated by WKG). Other action-oriented methods, such as simulation or role-playing, have become increasingly popular, especially in large companies. In many companies, for example, there is a project in which trainees manage a company for a day in the operational division. Independent learning is the basic principle of action orientation, which has been the guiding principle in vocational education and training since 1995 (Schaper et al., 2000: 134). Key qualifications during vocational training are: 1. the ability to act professionally, 2. the ability to solve problems, 3. the ability to cooperate and participate (Schaper et al., 2000: 132). A bundle of competencies is therefore required in the profession, which are aimed at further development after school: “If education is aimed at personality development and self-realization and competence at the ability to act in society, then every individual is to be wished both” (Efing, 2013a: 13, translated by WKG). Action competence is, therefore, the indicator of professional success. This is promoted, among other things, by the model of complete action: the first information phase, the second planning phase, the third decision phase (who does what in the team), the fourth execution phase (often a four-step method), the fifth control phase with self-monitoring and the sixth evaluation phase (training record such as learning diary as a sign of reflection) (Schaper et al., 2000: 140). The following procedure is very common to ensure that the training is legally secure: In a report booklet, the trainee records the training process in writing as a weekly or daily report. This booklet is submitted to the examination board at the intermediate examination in the middle of the training period and the final examination. All state-run dual vocational training courses are carried out in accordance with the respective training regulations. The training company plans the training by these regulations. It must document this transparently in the report booklet so the trainee knows which skills, knowledge, and abilities will be acquired in which training year (BMBF, 2010a: 3, 13; § 13 BBiG, 2005). Vocational schools, on the other hand, are based on the framework curricula of the Conference of Ministers of Education and Cultural Affairs (§§ 4,5; § 25 HwO, 1998). In most cases, dual vocational training lasts between two to four 46 2 Theory <?page no="47"?> years and ends after the final examination (§ 37 BBiG, 2005; § 31 HwO, 1998). This exam tests the trainee’s professional competence (BMBF, 2010: 25). Trainees are admitted to the final examination if they have successfully passed the intermediate examination, can provide written proof of training in the form of the report booklet, and have completed the training within the prescribed training period (BMBF, 2010: 25). The final examination consists of a written and a practical exam (BMBF, 2010: 26). The written final examinations of the commercial examinations of the Chamber of Industry and Commerce, for example, are planned centrally and uniformly throughout Germany by the task center for commercial final examinations and intermediate examinations (AkA for short), the central office for examination tasks of the chambers of industry and commerce (ZPA-Nord-West) and the Stuttgart Region Chamber of Industry and Commerce (IHK, 2010): The AkA is a joint institution for the supra-regional creation of examination tasks for commercial and commercial-related training occupations, such as the training occupation of insurance and finance clerk. It is managed by the Nuremberg Chamber of Industry and Commerce for Middle Franconia (AkA, 2011). The majority of IHKs obtain their written final examinations from this body. For example, 45 IHKs from 10 federal states have already joined the AkA association (ibid.). The Lüneburg-Wolfsburg Chamber of Industry and Commerce also uses part of its examinations from this institution (ibid.). The Central Office for Examination Tasks of the Chambers of Industry and Commerce (ZPA-Nord-West) is a company based in Cologne that together with the federal states of Berlin, Bremen, Hamburg, North Rhine-Westphalia, and Schleswig-Holstein, prepares standardized intermediate and final examinations for the CCIs. It also organizes the printing and distribution of these tasks to the IHKs (ZPA-Nord-West, 2011). The Stuttgart Region Chamber of Industry and Commerce creates task sets for the industrial-technical professions, such as electronics technicians for automation tech‐ nology (IHK Region Stuttgart, 2011) (Giera, 2010: 24) as part of the examination task and teaching material development center (PAL). During training, writing in certain professional situations and the associated text types are prescribed both by the framework curriculum for vocational schools of the Conference of Ministers of Education and Cultural Affairs and by the training regulations of the Federal Ministry of Justice (BMJ, 1998). The writing of text types also comes into play in certain final examinations. In addition to this, the course of a final examination will be illustrated in more detail - the 2.3 Writing skills in dual training 47 <?page no="48"?> 4 For legal reasons, no sample test may be used for illustration purposes. course and content of the final examination for hotel specialists in the IHK district of Lüneburg-Wolfsburg: 4 Example: First, the prospective hotel specialists have to take a written final examination on one day. This exam consists of several parts, including WISO (economics and social affairs), technical calculations, and technical theory. The practical part of the exam begins several weeks later. On the second day of the examination, the candidates are given a complex examination task, which is set by the examination board. A situation in the profession is specified. The candidates must first create a marketing concept for an arrangement with two overnight stays, meals and extras for a specific date, and a target group with an exact price and a list of all activities. The next step is to create an unsolicited offer (called an advertising letter in the task) on the PC and print it out. A further task, such as creating a room checklist, can also be added. The examination board evaluates this marketing concept and the fulfillment of the other tasks mentioned promptly on the same or the following day. There is a checklist for evaluating unsolicited offers. Two examiners evaluate these letters independently and summarize their results. A few days later, the candidate appears for the third day. On this day, they have to solve two practical tasks in one morning. This could be checking a room (using their own checklist) with the associated questions, working at reception in a typical professional situation such as check-in or check-out, or a telephone call with a guest who has received the written offer and now has questions they would like answered before booking. Another option is to work in a restaurant where guests from the region are invited, and the candidates have to assist the restaurant specialists as commis (helpers). The aim of all examination tasks is to be as situational and authentic as possible. For example, many vocational schools have a replica reception desk, a hotel room, a fully functional and equipped kitchen, a bar, and a replica restaurant. The training contract ends when the final examination is successfully passed. Trainees initially receive written confirmation from the examination board on the last day of the examination. The examination certificate, on the other hand, is issued by the regional chamber of industry and commerce for commercial professions and the training period ends with the ceremonial presentation of the certificate (BMBF, 2010b: 27). 48 2 Theory <?page no="49"?> 5 In other federal states, there are other names for this, which differ due to state sovereignty in the area of education. The first conclusion shows that the structure and process of dual training are dependent on education policy and the legal framework. The insight into a prototypical training process demonstrates the cooperation of several training partners and the application of action-oriented and independent learning. The current facts and figures on dual training in Germany are explained in more detail below: The training chambers such as the IHK and HWK (Chamber of Crafts) are struggling to find suitable applicants: “By 2030, there will be three million more academics and one million fewer skilled workers” (Schmickler, 2016: 3, translated by WKG). At the moment, training opportunities in Germany are good. Nevertheless, trainers repeatedly criticize the poor school performance of apprenticeship applicants - apprenticeships remain unfilled as a result (ibid.). In many places in Germany, there are training fairs, speed dating with applicants, sometimes in the schools themselves, taster days such as the Future Day, a wide range of internships in training companies, and extracurricular projects with training companies and chambers. This was unimaginable more than twenty years ago. The apprenticeship situation was very different from today. Even secondary school leavers competed with high school graduates for a training place. The training companies had a wide choice of applicants. This situation changed when the number of students increased. The new education policy is based on a higher number of school-leavers and students. The tripartite structure of the school system at secondary level I (Hauptschule, Realschule and Gymnasium), which is widespread in Germany, has been increasingly broken up in favor of more Gemeinschaftsschulen, 5 Oberschulen and Gesamtschulen. The aim is to make the school system even more permeable and to give pupils more opportunities to develop over the course of their schooling. The result is that secondary schools have less prestige if they do not offer the Abitur. Comprehensive schools, on the other hand, with their wide range of graduation options, are very popular, and grammar schools remain attractive. Many parents are keen to offer their children the best possible options, even if their child’s performance is below the requirements of the respective school. While the school landscape is increasingly flourishing with its school-based options for obtaining a qualification, training companies are struggling to find trainees. The attractiveness of completing a dual apprenticeship is de‐ creasing. In 2016 alone, around 480,000 training contracts were concluded in Germany. Between 2009 and 2017, the number of training contracts concluded in company-based training fell slightly from 518,505 (2009) to 507,772 (2017) 2.3 Writing skills in dual training 49 <?page no="50"?> (Federal Employment Agency, 2018). The decline in training contracts concluded between 2006 and 2015 can be put at 50,000 (BiBB, 2016). However, German companies need skilled workers, although it is estimated that only one in five companies in Germany still offers apprenticeships (Schmickler, 2016: 2). 80,000 training place applicants remain without an apprenticeship, and 270,000 young people complete internships or courses (AGJ, 2017). Some sectors, such as the hotel and catering industry, are less popular with school leavers and are barely able to fill their vacant training places (Schmickler, 2016: 3; BiBB, 2018). Certain professions are not attractive enough for young people due to the training content, time, and remuneration (Schmickler, 2016: 2). Schmickler (2010: 2) believes the supply-demand ratio has been unbalanced for over a decade. However, the cause is also to be found on the part of the companies. Companies are still reluctant to accept school leavers with a lower secondary school leaving certificate for training, even though these companies should look after the next generation of skilled workers (Schmickler, 2016: 3). Figure 10 illustrates this shortcoming visually. Fig. 10: Current training situation in Germany (own illustration, translated using AI) Furthermore, many young people with a higher level of education find studying more attractive due to the better financial prospects and higher standing in society (Schmickler, 2016: 3). This change is facilitated, among other things, by the increasing permeability of the German education system, which in Lower Saxony, for example, also allows subject-related studies without an Abitur under certain conditions such as a vocational training qualification and years of skilled work. The German labor market is also changing. Many jobs are no longer in the skilled trades, but in the service sector. The current figures and data on the status quo of vocational training for the years 2014 to 2017 show that there has been an increase in unfilled training places. While there were 41,040 in 2014/ 15, the number rose to 48,984 in 2016/ 2017 (Federal Employment Agency, 2017). Therefore, today’s school leavers have a wide range of training opportunities to choose from. In order to secure the next generation of skilled workers, the National Training Pact (Bundesagentur für Arbeit, 2006; KMK, 2010) was adopted back in 2004, which was intended to make training attractive again on the market for school leavers. 50 2 Theory <?page no="51"?> It was extended in 2006 and still exists today. Companies are encouraged to offer more training places to alleviate the lack of young skilled workers. Further figures on vocational training show that there are 1.3 million trainees in Germany, 770,100 of whom work in industry and commerce, and 362,400 in the skilled trades. Agriculture has 33,000 trainees, the public sector 37,500, the liberal professions 110,100, and home economics 5,600 trainees (Federal Statistical Office, 2017a). The number of students is significantly higher, at 1.7 million, and the trend is rising (ibid.). Since the 2012/ 2013 consultation year (annually from 1 October to 30 September of the following year), the Federal Employment Agency’s annual training market report has shown that the majority of school leavers do not have a university entrance qualification and are therefore in the secondary II sector at vocational schools. However, not all school leavers immediately start a course of study or vocational training (Federal Statistical Office, 2017a). When sorted by ranking, office management assistants are in first place, retail sales assistants in second place, and sales assistants in third place (BiBB, 2017). From the facts and figures on the distribution of school-leaving qualifications and vocational training, it can be concluded that most school-leavers do not have a university entrance qualification and therefore opt for vocational training. If vocational training is sought, the choice tends to fall on dual training in the Chamber of Industry and Commerce area. It can be seen that, from the school-leavers’ point of view, commercial professions and thus IHK apprentice‐ ships are more in demand. 2.3.2 WRITING IN DUAL TRAINING Writing in the training profession requires cooperation and integrates several participants in text production because the writer does not act independently and alone but is part of a company and must integrate into this hierarchical field, as shown in Figure 11, which is explained below ( Jakobs, 2005: 13 ff.): 2.3 Writing skills in dual training 51 <?page no="52"?> Fig. 11: Writing components in the workplace (after Jakobs, 2005: 17, translated using AI) The writer has the writing task in mind and uses his or her know-how, motivation, and experience to deal with this task. Writing, in turn, is part of the tasks at the workplace. Some parameters include supervisors, colleagues, resources, and specifications such as text samples. The workplace is, in turn, part of a company and, therefore, organizationally part of the company’s goals. The company is part of an industry or job-specific domain. In this domain, certain norms and values of written interaction apply: “The discourse communities in which or for which texts are created are found in this domain, and thus also potential addressees” ( Jakobs, 2005: 18). Each domain is part of a cultural space that is shaped socially, culturally and temporally, “[…] from which specific norms, aspirations and expectations result, legal and statutory requirements, specific value systems, etc.” ( Jakobs, 2005: 19, translated by WKG). For the professional text production process, skills and “[…] subject-related, rhetorical, text type, addressee or media-specific” ( Jakobs, 2005: 23, translated by WKG) competencies are only developed in the course of professional experience. Cooperative text production is a part of this, as drafts of texts circulate before they are sent out, especially in everyday working life - so-called “document 52 2 Theory <?page no="53"?> cycling” ( Jakobs, 2005: 24, translated by WKG) is common in order to meet the operational quality requirements of the texts: […] To put it more sharply, a lack of text and writing competence is not economically justifiable. Deficient presentations increase the time, cognitive and emotional effort required by the addressee. They not only cause frustration and additional working time, but also frequently result in poor performance due to imprecise, incorrect or incomplete information as well as additional time and money spent on ‘repairing’ deficient communication work ( Jakobs, 2008: 3, translated by WKG). Writing in vocational education arises from the combination of vocational school requirements, professional requirements in the training company, and industry requirements. It is supported by the prior knowledge of vocational school students from their previous schooling, their parents’ home and their own experiences. German at the workplace is concerned with all linguistic and communicative requirements. Although written communication is required in everyday working life, the responsibility for teaching it is primarily passed on to schools. While general education schools primarily teach academic text types, voca‐ tional schools are supposed to teach vocationally relevant text types in training classes (see the following figure): “Their teaching is — from the perspective of the school as an institution — the task of downstream training instances. The circle of mutual allocation of responsibility is difficult to break” ( Jakobs, 2008: 4, see Fig. 12, translated by WKG). Especially at vocational schools, there is little time for teaching professional writing skills in dual vocational training, and even less time in companies (ibid.). Writing skills are taught much more through learning by doing or copy and paste (ibid.). The phases of writing socialization in institutions illustrate that schools play a crucial role in developing necessary writing skills students need in their next stage, such as at the university or the workplace. This means that writing lessons at the lower secondary level play an enormously important role. Writing skills that have not been built up can only be developed further in limited time frames ( Jakobs, 2008: 4). 2.3 Writing skills in dual training 53 <?page no="54"?> Fig. 12: Phases of writing socialization ( Jakobs, 2008: 4, translated by WKG) The Fachstelle Berufsbezogenes Deutsch [Department for Job-Related German Language], which focuses primarily on employees with German as a second language and their support, explains this requirement in more detail: Vocational German lessons address various communicative requirements that are part of working life […]. In order to cope with such linguistic requirements, the respective speakers must have a whole range of communicative skills: They must be just as familiar with the content structure of specialist texts and work instructions as with the vocabulary they contain. In addition, it is necessary to know how to speak to customers, colleagues or superiors and to be able to use the appropriate colloquial, formal or polite language forms and registers (Fachstelle Berufsbezogenes Deutsch n.d., translated by WKG). This explanation makes it clear that this also refers to trainees in vocational training and the profession’s associated written and oral requirements. The writing requirements in vocational education and training are prescribed by the Conference of Ministers of Education and Cultural Affairs in the framework curriculum for vocational schools. This provides vocational school teachers with information on which text types and writing occasions must be taught in which year. Internal school curricula are based on the framework curricula. The Federal Ministry of Justice prescribes the teaching of specific text types directly and indirectly in the training regulations through the integrated factual 54 2 Theory <?page no="55"?> training framework plan - indirectly because sometimes only writing occasions are mentioned, which, however, require certain text types in professional practice. As the third educational policy actor in dual vocational training, the Chambers require writing certain text types as a testable performance in the final examination. It should be emphasized once again at this point that not only the production of various types of texts for certain typical professional writing occasions, but also receptive skills, such as reading and understanding the texts, should be promoted in the training (Giera, 2010: 25 f.). These three educational policy actors specify a large part of the writing requirements in dual training. The training company and the vocational school take up these writing requirements in various learning fields, which are structured in an action-oriented manner and have a professional, legally relevant reference (Czycholl & Ebner, 1995), as part of the learning field didactics introduced by the Standing Conference of the Ministers of Education and Cultural Affairs (KMK, 1996). The current state of research in the field of writing requirements and skills in vocational education and training shows that a relatively young field of research has been established to map the status quo on the one hand empirically and to recommend writing didactic measures on the other (for more details, see Neumann & Giera, 2018: 333 ff.). • In her dissertation, “What and how do apprentices write? An analysis of writing habits and selected formal features in essays” (Wyss-Kolb, 1995, translated by WKG), Wyss-Kolb was able to prove through a content analysis of 88 essays by Swiss-German apprentices in the 11th and 12th grades, 15 double-lesson essays, 13 Matura essays, and 30 one-hour essays by Zurich vocational school students that the texts averaged 285 words in length with grammar school students’ essays being significantly shorter. A qualitative error analysis based on the Zurich text analysis grid was used to examine the linguistic and orthographic accuracy of the student texts (Wyss-Kolb, 1995: 12). Although the grammatical error density of the vocational school students was twice as high as that of the grammar school students, 93 % of the texts were formally correct and contained few semantic errors (Wyss-Kolb, 1995: 107, 142, 272). The main errors included careless mistakes, comma errors, disposition errors (text development and coherence), as well as errors in the word choice, syntax or a reader-friendly structure (Wyss-Kolb, 1995: 22). A subsequent survey of apprentices (n = 128) illustrated that more pragmatic writing tasks such as writing applica‐ tions and complaints should be set by vocational school teachers, as shown in the following quote: 2.3 Writing skills in dual training 55 <?page no="56"?> There is also a clear postulate that the subject - similar to German lessons at vocational schools - should focus more on life skills and useful information. The young women and men would like to ‘learn techniques to find their way better in life’, would like ‘more everyday German (complaints, job applications)’, and more discussion and essay topics related to current problems and events or the profession […]. A certain discrepancy can therefore be observed between lessons that are more oriented towards humanistic educational ideals and the expectations of a student body that is interested in life and practical relevance (Wyss-Kolb, 1995: 41, translated by WKG). Furthermore, Wyss-Kolb points out in her study that students do not generally dislike writing longer texts, that they dislike writing essays, and are more likely to write business letters, exams or homework (Wyss-Kolb, 1995: 72). As a result, two out of five apprentices write reluctantly and rarely, even though they have to write every day at work (Wyss-Kolb, 1995: 66, 271). • Fleuchhaus (2004, translated by WKG) surveyed trainees in her doctoral project, “Communicative skills of trainees in vocational training: Character‐ istics, promotion and relevance in the judgment of trainers, teachers and trainees”, trainees (n = 1,360) in 21 occupational groups, vocational school teachers (n = 60) and trainers (n = 60) were asked in a questionnaire with a five-point rating scale what kind of texts and how and when they would write them (Fleuchhaus, 2004: 23). Significant results were that although the trainees frequently write on the PC from their perspective (M = 3.76), they rarely write business letters (M = 2.45) and, interestingly, the vocational school was rated more as a place for learning to write than the company (ibid.). Writing as a professional requirement is perceived more strongly by trainees who must develop and master this competence in the company than by trainees who are less likely to write texts professionally. The survey of trainers emphasized that although the computer is relevant as a work tool, there is little time for training and that this should be taught at vocational schools. On the other hand, the trainers consider formal language requirements such as grammar, spelling, and syntax necessary. These basics are not sufficiently developed in the trainees (Fleuchhaus, 2004: 262). • In the 2002/ 03 school year, the longitudinal study “Aspekte der Lernaus‐ gangslage und der Lernentwicklung - Klassenstufe 11” [Aspects of learning starting point and learning development - Grade 11] (Lehmann et al., 2004: 5, translated by WKG), LAU for short, was carried out at 100 Hamburg schools, as well as the parallel “Untersuchung der Leistung, Motivation 56 2 Theory <?page no="57"?> 6 This task was later used in the DESI study. und Einstellung zu Beginn der beruflichen Ausbildung” [Investigation of performance, motivation and attitude at the beginning of vocational training] (Lehmann et al., 2005: 5, translated by WKG). The aim was to record writing competence based on writing tasks that the pupils had to complete at general and vocational schools: In the first writing task (n = 3,517), pupils were asked to write a letter of complaint to the local council to oppose the closure of a youth club. 6 The second writing task (n = 1,539) was a response letter from the fictitious local council. The text quality was determined using a double-blind rating procedure and a five-point scale with content-related and linguistic-textual characteristics (Neumann, 2006: 24 f.). Almost two-thirds wrote a letter of complaint that was functionally, linguis‐ tically, and formally mature. In contrast, 34-% of the eleventh graders were found to have writing skills that required urgent writing support, as they had failed to achieve the writing task or goal (Neumann, 2006: 28). One-fifth of the test subjects wrote almost error-free letters. Overall, the vocational school students wrote better quality letters than the eleventh graders at general education schools, as the mean value determined (MEAN 500/ SD 100) was eleven points higher (Neumann, 2007: 194). • Efing and Janich (2006) analyzed the results of the Hessian pilot project “Vo‐ cational Literacy - Methodological and Linguistic Competencies in Vocational Education and Training” (Biedebach, 2006). To this end, the language skills of vocational school students (n = 624) aged 15 to 51 from various training occupations were examined (Efing & Janich, 2006: 7). Vocational literacy” (Efing & Janich, 2006, translated by WKG) is defined as “[…] the sum of the language skills required in specific occupational contexts” (Biedebach, 2006: 16, translated by WKG). The impetus for this study was to reduce the failure rates in final examinations. A “problem type test” (Efing & Janich, 2006, translated by WKG), a questionnaire survey with 48 teachers, and 97 qualitative interviews with students were carried out in ten vocational fields with 415 vocational school students. In the problem type test, the vocational students were asked to reproduce the content of a reading text on alcohol control in a few sentences, which 28 % of the test subjects either did not complete at all or started but stopped. Concerning the type of text, the students’ writing skills were highly deficient: the texts were hardly structured and showed little coherence, the content of the reading text did not come into play, the writer’s opinions were mixed up with the reproduction of the text, and the text pattern knowledge 2.3 Writing skills in dual training 57 <?page no="58"?> did not appear to be present (Efing, 2006: 39 ff., 42). Neither gender nor previous school education or career choice in this heterogeneous group of test subjects had a positive or negative effect on the text quality of the content statements. It was therefore concluded that “inner multilingualism”, strategic knowledge for understanding and writing texts, and concentration should be promoted in both processes (Efing, 2006: 61 f., translated by WKG). The questionnaire survey with 48 teachers identified low spelling compe‐ tence among vocational students as the main problem in writing proficiency. General writing competence ranked second, followed by expression/ style in third. Efing points out “[…] that the students already show such great problems in these basic areas in their text production that the promotion of writing competence has to start in the general area and not just in the job-specific area” (Efing, 2008: 20, translated by WKG). Efing (2008: 28 f.) cites the few opportunities to practise and the low moti‐ vation and partial aversion to text production and certain types of school texts as the reasons for the low level of writing competence. He calls for the teaching of text pattern knowledge, the development of writing routines, the training of language reflection, and the promotion of a feedback culture in German, and ultimately for a change in attitude on the part of teachers so that they do not just teach spelling. • In 2009 and 2010, Efing and Häußler investigated the “linguistic and communicative demands on trainees in the application and training phase” (Efing, 2013b: 126, translated by WKG). To this end, participant observations were carried out to record the professionally relevant text types, types of conversation, and semi-standardized interviews with trainees (n = 30) and trainers (n = 16) in a large company and small and medium-sized enterprises (n = 6) in the Stuttgart/ Heidelberg area. The test subjects came from the skilled trades (industrial mechanics, mechanics, mechatronics technicians, electronics technicians, technical drafters). The results of the survey of trainers showed, […] that in company-based training, language skills (in the sense of language system and language standard competence) are only explicitly relevant in the dimension of spelling competence - and this almost exclusively in the context of the application letter and when reviewing and correcting the report booklets. Other language system skills do not play a major role as long as they are present at a basic level and do not impair communication (Efing, 2013b: 127 f., translated by WKG). The text types to be produced by the trainees are primarily tables and lists (Efing, 2013b: 128). The survey of trainers revealed that communicative 58 2 Theory <?page no="59"?> action situations in the oral and written area are “hardly recognized” (Efing, 2013b: 129, translated by WKG). In contrast, the trainers rated active listening, structuring oral and written texts, and pointed formulation as relevant. The participant observation emphasized that in addition to writing and receiving less continuous texts, participants must also produce work plans, report books, protocols, self-reflections, presentations, documentations, reports, short notes, technical drawings, forms, tables, lists, and learning objective checks (Efing, 2010; Efing & Häußler, 2011; Efing, 2013b: 129). These short texts serve the purpose of “clear sense taking” (Efing, 2013b: 130, translated by WKG), effective communication with colleagues, and the execution of work assignments, usually in a chain of work processes: For example, an order is described as a short note, supporting drawings are made/ read, supplementary lists refer to the required material and tables provide an overview of costs […] (Efing, 2013b: 130 f., translated by WKG). When trainees write, the text products are usually addressed to multiple parties (colleagues, customers, superiors). This means that communicative rather than systematic language skills are required and promoted, especially during training (Efing, 2013b: 140). • The study “Deutsch am Arbeitsplatz - Untersuchung zur Kommunikation im Betrieb als Grundlage einer organizationsbezogenen Zweitsprachenförderung” [German at the workplace - Study on communication in the workplace as a basis for organization-related second language support] (Kimmelmann, 2013: 293, translated by WKG) conducted by the German Institute for Adult Education (DIE) was funded by the Volkswagen Foundation as an interdisciplinary research project from 2007 to 2013. The project aimed to document linguistic and communicative requirements in commercial professions, analyze authentic communication situations, and develop fur‐ ther training modules based on the results. The target group focused mainly on employees with German as a second language. The method used was ethnographic interviews in 15 companies in the catering, consumer goods, elderly care, electrical and metalworking, plastics, logistics, furni‐ ture manufacturing, and building cleaning sectors (Kimmelmann, 2013: 293 f.). In addition to the interviews on “[…] company organization, work processes and workplaces, employee structure, induction and, in particular, the communicative requirements in the company as a whole or at specific workplaces” (Kimmelmann, 2013: 294, translated by WKG), 70 audio record‐ ings and 150 written documents were obtained as data material, of which 56 2.3 Writing skills in dual training 59 <?page no="60"?> conversations were transcribed and 100 emails were linguistically analyzed for “[…] grammatical structures of vocabulary and speech acts” (ibid.). The result shows that the language requirements in every company, even in more supportive jobs, are increasing due to the flexibility of jobs, which constantly require new language handling situations. This makes it difficult for employees with language deficits to get a job (ibid.). Reasons for this include the growing number of regulations due to company certifications, which lead to enhanced language and communication skills (ibid.). An ex‐ ample from the kitchen area of the catering industry illustrates this through an interview with an operations manager: “We have so-called five-minute meetings every morning (in the kitchen; author’s note) in which the quality requirements are discussed” (Kimmelmann, 2013: 295, translated by WKG). In addition to this increasing control and documentation of work processes, another feature of the changed work situation is “flatter hierarchies/ work organization in the team” (Kimmelmann, 2013: 295, translated by WKG). In terms of language, this situation has the consequence of forming an opinion oneself, but also of describing parts of the work process and correcting errors independently (ibid.). On a linguistic level, it also leads to “auditing/ certification” (ibid.), as errors must be described and justified during inspections (ibid.). Furthermore, “more customer contact and greater customer orientation” (ibid.) are required of employees. They note that there is a “greater need for further training” (ibid.) due to the absorption of subject-specific information on an oral or written basis (ibid.). Further results are the authentic language acts, which were analyzed based on the the Common European Framework of Reference for Languages (CEFR) levels. The most important finding from this is: Linguistic and communicative activities in the workplace cannot be clearly assigned to a specific level. Instead, employees have to switch repeatedly (and sometimes very quickly) between low and high level language requirements. This means that it is no longer possible to define occupations, activities or work processes that require little or no language (Kimmelmann, 2013: 296, translated by WKG). The proportion and importance of written versus oral communication did not lead to any clear findings. Instead, oral and linguistic links were found that occur in a professional language situation (ibid.). From this, Kimmel‐ mann concludes that there is also a need for language support in industrial training occupations as a “cross-cutting issue” (Kimmelmann, 2013: 296, translated by WKG) and across the boundaries of occupations and target groups (ibid.): “Conversely, this requires teachers in vocational education 60 2 Theory <?page no="61"?> and training to specifically integrate speaking and writing occasions into lessons in order to train both areas of competence equally” (Kimmelmann, 2013: 296, translated by WKG). Kimmelmann, therefore, recommends draw‐ ing on authentic situations and material from companies and taking into account a company’s linguistic and communicative requirements (ibid.). • With the dissertation title “‘You have to be able to cope with writing a bit! Eine Studie zu den Schreibfähigkeiten von Auszubildenden (n = 175) im unteren Ausbildungssegment im Kontext der Ausbildungsreife” [“A study on the writing skills of trainees (n = 175) in the lower training segment in the context of training readiness”] (Baumann, 2014, translated by WKG), Baumann investigated whether the term Ausbildungsreife used by the Federal Employment Agency with the characteristic of (legal) writing has a strong influence on the transition, passing and employment after training (Baumann, 2014: 6, 58). The 175 Hamburg trainees from 49 occupations predominantly had a lower secondary school leaving certificate; in the survey period from June 2011 to February 2012, they completed three writing tasks: 1. writing a formal application letter, 2. writing a construction manual, and 3. the LAU-9 task on passive spelling knowledge. The first two tasks were evaluated according to the Zurich text analysis grid, and the text quality was assessed globally on a rating scale. Other evaluation tools used were the assessment of legibility, a qualitative model, and a text description model based on Sieber to determine closeness to the standard (Baumann, 2014). The results of the 133 student texts show the following picture: 11 (8.3%) texts were close to the standard and 42 texts (31.6%) were far from the standard. 67 (47.5%) application letters were written in accordance with the content and formal criteria, while 48 application letters (34 %) were written largely in accordance with the formal and content criteria (Baumann, 2014: 270). Baumann comes to the conclusion that lower (legal) writing skills of trainees do not influence the trainer assessment and that this characteristic of training maturity in the Employment Agency’s criteria catalog is therefore only an orientation and not a variable for training success (Baumann, 2014: 268). Furthermore, […] there is a very wide range of results with a mean value that can be regarded as rather poor. This finding is independent of the time of training (first/ last year of training) (Baumann & Siemon, 2013: 287, translated by WKG). Baumann and Siemon (2013), therefore, conclude that promoting language and communication skills during vocational training is important but should 2.3 Writing skills in dual training 61 <?page no="62"?> not be a barrier to starting vocational training (Baumann & Siemon, 2013: 288). • The “PROSAB” study by Hoefele, Konstantinidou, Kruse and Dieterich (2015) focused on the multilingualism of trainees and aimed to reduce drop-outs (Hoefele et al., 2015: 1). This controlled intervention study with pre-test (n = 287, 1st writing task: letter to the school principal), post-test (n = 278, 2nd writing task: letter to the education directorate) and a follow-up test four months later (n = 275, 3rd writing task: letter to the trainer requesting time off) showed writing results on products, the organization of writing processes, and the attitude towards writing itself. The test subjects (n = 317) were in their first year of training in nine training occupations in 2013 (ibid.). The content quality of the letters was examined according to the criteria of text type convention, text structure, communicative effect, language style, and linguistic accuracy. The DESI and VERA scales, and the Zurich text analysis grid were used for evaluation. The intervention group performed significantly better than the control group in all three writing tasks. Improvements were particularly evident in the linguistic-pragmatic area and the formal text features. Concerning communicative effect and addressee appropriateness, the intervention group showed significant improvements at all three survey points. The control group developed positively regarding writing technique but could not improve significantly. Asking for feedback from classmates occurred significantly more often in the intervention group than in the control group. Writing and school-related self-efficacy and self-regulation did not show any changes throughout the survey, but negative changes were observed in attitudes towards writing. Further results were: Pupils with German as their mother tongue developed more positively in the study than pupils with German as a second language. Pupils with a higher level of education achieved a higher text quality than pupils with a lower level of education (Hoefele et al., 2015: 8 f; Konstantinidou, Hoefele & Kruse, 2016). Overall, the researchers saw a high discrepancy between existing curriculum goals and the actual state of the students. Emphasizing writing instruction, including the writing process, was proposed as a solution (Hoefele et al., 2015: 10). Writing at work, on the other hand, has been a field of research both nationally and internationally since the 1980s ( Jakobs, 2005: 13). In this context, a workshop of the European Research Network on Learning to Write Effectively was held in Potsdam in 2010 based on the financial support from the EU (COST ACTION IS0703). The output of Working Group 3 - Design of Written Documents in the Workplace with the leading experts Franck Ganier (France) and Joyce Karreman 62 2 Theory <?page no="63"?> (Netherlands) was the Potsdam Triangle Model (ERN-LWE, 2011, see Fig. 13). This is a consensus on how writing is done in the professional environment today. The model focuses not just a single writer, but an entire communication team that collaborately writes and revises, with the writing task at its center. The multiple addressing is reflected in the form of the term user. Fig. 13: Potsdam Triangle Model (own illustration) The presentation of the current research field of writing competence in voca‐ tional education and training shows that both qualitative and quantitative evaluation methods are used. Evaluating writing competence through writing tasks and the associated evaluation with scales and grids is standard practice. The evaluation no longer concentrates so strongly on the sub-area of spelling but focuses equally on language-pragmatic and language-systematic sub-areas from a communicative perspective. Furthermore, the writing process is examined more closely, which is achieved using questionnaires on attitudes to writing. The use of quasi-experimental study designs shows the significance of the results, even if randomization and a larger sample of more than 1,000 vocational training participants are still lacking. The study results demonstrate a develop‐ ment of the research subject. The early language-systematic studies are now increasingly being supplemented by language-pragmatic studies. In addition, vocational school students are demanding job-specific writing tasks that are also linguistically pragmatic and do not just focus on language-systematic orthography as an assessment criterion (Wyss-Kolb, 1995; Efing & Janich, 2006; 2.3 Writing skills in dual training 63 <?page no="64"?> Baumann, 2014). If this perspective is broadened, it can be determined that almost two-thirds of (vocational) students can write a letter correctly in terms of language form (Lehmann et al. 2004, 2005). There is little time for the explicit teaching of business letters, both in companies and in vocational school lessons (Fleuchhaus, 2004). Particularly in commercial and craft occupations, writing extended or expository texts is replaced by tables and forms in everyday working life. Workplace texts are typically multi-addressed and embedded in multi-layered action situations (Efing & Häußler, 2011). Multiple addressees (users) in professional settings require writers to change their perspective - something that differs considerably from school writing. Therefore, voca‐ tional schools and preparatory writing tasks at lower secondary level must be action-oriented, address multiple audiences, and have language-systematic and language-pragmatic focuses within professional contexts. This is essential because research has shown that promoting these skills is the key to developing writing skills in the professional field (Hoefele et al., 2015). Having explained the theoretical background of vocational training and its writing competence, the following section will examine text (type) competence in the vocational field. The following subchapter will focus on professional text types, as all writing competence models emphasize the relevance of text patterns, which influence the writing process. If the text type associated with the writing task is known, prototypical text patterns can be used. If knowledge of text patterns is lacking, the working memory must adapt existing patterns. Knowledge of text types favors the writing process if the writing task requires a specific text type. To this day, the models also have an influence on writing research and writing didactics (Böhme et al., 2017: 57). Writing produces texts that the recipient implicitly and explicitly evaluates. In the next chapter, the concept of text and text type with the associated requirements will be explained in detail. 64 2 Theory <?page no="65"?> 2.4 TEXT (TYPE) COMPETENCE IN THE PROFESSIONAL FIELD “Emails have a lot in common with the little notes in the classroom back then: a few lines scribbled in tiny letters, folded twice, the name of the friend written on it and then fired off. […] Today, these little notes fly around the world: if they were visible […], our globe would look like it was covered by a snowstorm” ( Jäger, 2007: 149, translated by WKG). To be competent in writing, writers must have a potpourri of text (type) knowledge that they learn receptively and productively. This knowledge leads to text (type) competence, which can be called up in memory depending on the writing occasion and can be demonstrated in its performance. This creates a text routine (see Section 2.1). Every writen text can be mentally added to a treasure trove of text types, which is continually built up during the writing genesis and used as required to write the text type anchored in the writing task. Text competence is thus a “procedural meta-knowledge for text constitution” (Feilke, 2012c: 10, translated by WKG). The opening quote illustrates how writing changes over time in terms of medium and occasion. However, other writing occasions have changed less, and these have developed through certain communication situations and are still anchored in certain text types today. Text types have emerged from history, change and develop further, but remain the same at their core (Brommer, 2018: 65). A brief example from the historical genesis of text types shows that the Sumerians used cuneiform tablets for writing supply lists, invoices and contracts from around 3200 BC (Haarmann, 2007: 8). The use was aimed at processing information and agreements (ibid.). The medium, however, changed over time. Even today, non-literary texts are still professional or work-related texts (Brinker, 2010: 16). This group of factual and informational texts is primarily intended to inform the recipient rather than entertain them, as is the case with literary texts (Rosebrock, 2007: 50). The focus is on a “communicative-functional aspect” (Brinker, 2001: 112, translated by WKG). Consequently, text producers and recipients are in a communicative and business relationship where infor‐ mation is exchanged (ibid.). It should, therefore, be emphasized that, on the one hand, professional texts must be written clearly, concisely, and comprehensibly, and, on the other hand, the business relationship should not be disturbed by 2.4 Text (Type) competence in the professional field 65 <?page no="66"?> these texts but rather maintained. In addition to clarity, comprehensibility, and text density, maintaining a respectful tone is also essential in business correspondence ( Jäger, 2007: 8 ff.). According to Brinker (2010), a distinction is made between three forms of factual and informational texts: Instructional texts, which convey factual knowledge to transfer information; persuasive texts, which contain an inherent argumentative core and aim to convince the recipient through their appeal function (Brinker, 2010: 101); and instructional texts, which provide procedural knowledge to enable the recipients to perform specific actions (Brinker, 2010: 89). The term text must first be defined before addressing professional text types (Giera, 2010: 3): A prototypical text is perceived as a fabric of words and signs, categorized as written discourse or “textualization” (Ehlich, 1984, translated by WKG) (Koch & Oesterreicher, 2008: 204). Texts are also regarded “[…] basic units of linguistic communication, as instruments of communicative action” (Heinemann, 2008: 137, translated by WKG). It is a “[…] type of utterance that has arisen under the communicative conditions of distance […]” (Koch & Oesterreicher, 2008: 204, translated by WKG). On the one hand, text structures and formulations can be understood as linguistic factors; on the other hand, texts contain a multitude of non-linguistic factors that are not only recognizable in the text but also make the text recognizable in the context of an interaction process (Heinemann, 2008: 136). Text linguistics does not have a standardized concept of text (Adamzik, 2004: 32; Linke et al., 2004: 242). Today, the term text is prototypically used as a “[…] means of digesting the ephemeral linguistic basic event, the linguistic action […]” (Ehlich, 1994, translated by WKG) and as a “written, thematically coherent sequence of words, sentences, the wording of a speech, a piece of writing” (Basler & Schulz, 1981, translated by WKG). These definitions reveal two major traditions: one rooted in rhetoric and the other in grammar. The rhetorical tradition traces the concept of text from the speech, i.e., oratio, as recorded in the works of the Greek philosophers Socrates, Plato, and Aristotle. The written text, in this sense, was considered the transcript of the dialogues — the dead speech (Wolf, 2009: 87 f.; Adamzik, 2004: 38). In contrast, the grammatica tradition regards the text as derived from the sentence level (Adamzik, 2004: 35). In the German-speaking context, the term text has been in use since the 14th century. In the 16th century, this term was interpreted as the “linguistic part of a piece of music” (Adamzik, 2004: 33, translated by WKG) and also as the “linguistic part of a picture-text unit” (ibid.). The term lyrics originates there and refers to text written to accompany a picture or a piece of music. 66 2 Theory <?page no="67"?> It should be emphasized that a text forms part of a linguistic action and contains a sequence of words and sentences: They are “processes of producing and understanding utterance complexes” (Heinemann, 2008: 115, translated by WKG). The debate over what constitutes a text continues, including further discussions about the criteria of medium (written or spoken), the reference (product vs. user), and the scope or completeness (Adamzik, 2004: 41 ff.). A text can be either oral or written. Söll and Hausmann (1985: 17 ff.), Koch and Oesterreicher (1985: 17; 2008: 200 f.) and later Schwitalla (1997: 17) categorize texts medially according to their conceptual orality or conceptual writing (see Fig. 14). Fig. 14: Proximity-distance continuum (Koch & Oesterreicher, 2008: 201, translated by WKG) Example: Chat histories such as on Facebook or WhatsApp are written in media terms, but are more conceptually oral due to the use of colloquial language. In contrast, a speech, such as at a wedding, may be oral in media terms, but is conceptually written. The second point of discussion is the reference to the product. At this stage, a text typically consists of several sentences oriented towards the product. It is medi‐ ally graphic, but depending on the communication situation, it is written either in “communicative proximity” (Koch & Oesterreicher, 2008: 201, translated by WKG) or in “communicative distance” (ibid.). For business letters, the conditions are more likely to apply: conceptually written, graphically communicative, and communicative distance. This is formulated by the “strangeness of the 2.4 Text (Type) competence in the professional field 67 <?page no="68"?> communication partners” (Koch & Oesterreicher, 2008: 201, translated by WKG), even if the name is in the salutation, the “spatio-temporal distance” (ibid.), the “reflectiveness” and the “referential distance” due to the professional framework of an application process, or in the case of offers, due to the conclusion of a contract, and the “monologicity” (ibid.) due to the letter. Indeed, an applicant or even a hotelier tries to convince the potential guest or employer of himself or the hotel through an emotional connection to the addressee and the formulation of situations, but this is only psychological calculation and rhetorical skill and has nothing in common with “privacy” (Koch & Oesterreicher, 2008: 201, translated by WKG) or due to “strong emotional involvement” (ibid.) with the communication conditions of proximity: Such gradations on the proximity-distance continuum are part of the linguistic-com‐ municative knowledge that allows the producers to choose the appropriate linguistic strategies and thus give the utterances a suitable ‘style’, a different degree of elaboration and formality; the recipients also perceive these differences and compare them with their expectations (Koch & Oesterreicher, 2008: 202, translated by WKG). In terms of language system, coherence, i.e., the deep structure of the text, and cohesion, i.e., the use of repetition structures on the surface of the text, play a crucial role here (Linke et al., 2004: 254 f.). A text with several sentences may be grammatically correct but make no sense to the recipient if the sentences do not relate to each other on the surface of the text and in the deep structure. At this point, it is not clear whether a text must refer to the recipient (Adamzik, 2004: 95). Hurrelmann and Groeben (2006: 33, translated by WKG) classify this as a “result of the recipient’s interpretation.” The third point of discussion is scope. A text should be relatively self-con‐ tained (Adamzik, 2004: 44). New media such as text messages, e-mails, or chat histories stimulate this point of discussion because these media in particular have texts that are not always clearly self-contained and cannot be identified in terms of their coherence and cohesion, if sentences are written at all. At this point, further criteria are needed to define a text. De Beaugrande and Dressler (1981: 3 ff., translated by WKG) set out seven complete criteria for determining textuality, such as “cohesion”, “coherence”, “intentionality”, “acceptability”, “informativity”, “situationality”, and “intertex‐ tuality” (ibid.): We define a text as a COMMUNICATIVE occurence that fulfills seven criteria of TEXTUALITY. If any of these criteria is considered not to be fulfilled, the text is not considered communicative. Therefore, non-communicative texts are treated as non-texts (de Beaugrande & Dressler, 1981: 3, translated by WKG). 68 2 Theory <?page no="69"?> These seven criteria, which no longer claim to be selective or complete (Fix, 2008: 19), were supplemented by Sandig with the criterion of style (2006: 309). At this point, Heinemann’s statement about the text should be examined once again. On the side of language pragmatics, criteria such as “situationality, acceptability, and intertextuality” (Heinemann, 2008: 137, translated by WKG) are pragmatic criteria, while the criteria “informativity, cohesion, coherence” are textual features on the surface of the text. From the point of view of language pragmatics, there is no exclusive condition for the classification of a text, i.e. textuality (ibid.). The efficiency of a text depends on the lowest possible degree of effort and exertion on the part of the communication participants when using the text. Effectiveness depends on whether it leaves a strong impression and creates favorable conditions for achieving a goal. The appropriateness of a text is the correspondence of a text between its context and the way in which the criteria of textuality are upheld (de Beaugrande & Dressler, 1981: 14, translated by WKG). The writing process models show that pragmatic text linguistics has a strong connection to cognitive research. In its work, it is oriented towards text types (Fix, 2008: 16) and asks […] from the very beginning, she has been investigating what actually constitutes the text. […] In the course of her research, against the background of various theoretical approaches, she developed a series of different conceptions of the text, which expanded more and more. […] On the other hand, in a process that began later and is still ongoing and intensifying today, it endeavours to grasp the typology of texts, i.e., the exemplary characteristics of the phenomenon of text that a linguistic and cultural community has developed in common action, i.e., to determine the TYPES OF TEXT in which texts are realized (Fix, 2008: 17, translated by WKG). Texts manifest common characteristics and can, therefore, be categorized. Text types are created through the reception of the reader. However, texts cannot always be assigned to a specific text type, which Adamzik (2008: 147, translated by WKG) calls the “sorting problem.” The classification of a text to a text type is complex, as each text type demands certain expectations. The study of text types can be conducted in two ways: Either one starts with the traditionally defined text types as observable objects and attempts to reconstruct them via a consistent text theory, or one starts from a text theory that deductively offers theoretical types for comparison with empirical examples. For a science of texts as human activities, it seems impossible to dispense with traditional text types, which actually have heuristic value for text users in the 2.4 Text (Type) competence in the professional field 69 <?page no="70"?> processes of production and reception (de Beaugrande & Dressler, 1981: 189, translated by WKG). The term “text type” and the content of the term “text type” are repeatedly discussed in German didactics and text linguistics, including whether it should still be used at all (Koch & Oesterreicher, 2008: 211). What is certain is that knowledge of text types and patterns is a degree of written-language communi‐ cation experience consolidated as a person learns. The term text type used in this paper does not connote anything negative but is derived from pattern learning in terms of learning psychology. Since written communication also follows linguistic patterns developed by society (Brommer, 2018: 65; Brinker, 2010: 120), the recipient can recognize a pattern through a sum of text characteristics. The following definition is therefore used in this paper: As a rule, text types are highly variable but at the same time conventional patterns of composition of such action-related functional text modules that stand between the elementary syntactic and the global level of the whole text (Feilke, 2014: 23, translated by WKG). To identify a text type, two characteristics have become established in text linguistics: 1. the “text function” (Brinker, 2001, translated by WKG) and 2. the “textual pattern” such as “[…] narration, description, argumentation and explication […]” (ibid.). For a classification of text types, the higher instance of text class and text type must first be considered (Linke et al., 2004: 282 f.). However, it should be emphasized that the classifiying all texts into classes and types will never be sufficient, as new text types are constantly emerging. It is not always easy to distinguish between text types, even though Eckard Rolf (1993) attempted to do so in his work (Adamzik, 2008: 162). Ultimately, the situation in which a text is created is the guiding factor and allows all classes and types to be sorted anew (Adamzik, 2008: 162). Classification of the terms is repeatedly attempted, but in this work, we will work based on Adamzik, whereby the terms such as text type class, text type, text type variant, or text type do not lead to any distinctions or levels of abstraction (Adamzik, 2008: 170). Text-internal and text-external criteria are the main fundamental distinctions used to describe a text (ibid.). Even if text-internal and text-external classifica‐ tion criteria can categorize a text type, text types are still seen as “heterogeneous, prototypically structured categories with vague boundaries” (Linke et al., 2004: 289, translated by WKG) because classifications and categorizations are not always clearly possible. Text-internal criteria can be (Linke et al., 2004: 278 ff., translated by WKG): 70 2 Theory <?page no="71"?> • “the phonetic-paraverbal (or graphic) level” (Linke et al., 2004: 278), • “the choice of words” (Linke et al., 2004: 279), • “Type and frequency of sentence structure patterns” (ibid.), • “the topic connection and the course of the topic” (ibid.), • “the topic itself ” (ibid.) and • “Text structure patterns” (ibid.). In contrast, the following criteria can be counted as external to the text: • “the text function” (Linke et al., 2004: 280), • “the communication medium that ‘carries’ the text” (ibid.) and • “the communication situation in which a text is embedded” (ibid.). Depending on the linguist, whether Dressler, Brinker, Morris, Heinemann, or Adamzik, these aspects are among the two fundamental distinctions (Adamzik, 2008: 165). In this paper, Brinker’s integrative concept of text is used as a definition and basis: “A ‘text’ is therefore to be regarded as a linguistic and at the same time communicative unit, i.e., as a limited, grammatically and thematically coherent (coherent) sequence of linguistic signs, which as such realizes a recognizable communicative function (text function)” (Brinker, 2010: 19 f., translated by WKG). This means that the text function as a descriptive feature of a text is distinguished as a text-external feature, while the grammatical and thematic text structure are classified as text-internal features (Adamzik, 2008: 165). Nevertheless, texts can never be perfectly assigned to a text type. Instead, features can appear in the text, which, due to their frequency, lead to text types because they are close to them and prototypical. There will never be a ready-made list of textual features for a text type. It should be emphasized that a text can contain several functions, but one basic function usually predominates. Karl Bühler initially highlighted the three functions of expression, appeal, and presentation in his “Organon Model” (Bühler, 1934, translated by WKG). This was followed by further developments in text linguistics by Große (1976), Isenberg (1974), and the model currently used by Brinker (2010). The informa‐ tion and appeal functions have already been mentioned and are prevalent in professional text types. In addition, there is also the obligation function, which obliges the recipient to take a specific action (Brinker, 2010: 109), the contact function, to establish or maintain a relationship (Brinker, 2010: 110), and the declaration function, to create a new reality for the recipient (Brinker, 2010: 111). 2.4 Text (Type) competence in the professional field 71 <?page no="72"?> It is important to recognize the function in texts, as the recipient can be stimulated to different actions by a text: according to Austin (1979) to acts such as the “locutive act” (Austin 1979, translated by WKG), through the utterance itself to the “illocutionary act” (ibid.) such as stating or requesting, as well as to the “perlocutionary act” (ibid.) as an effect and the “propositional act” (Searle, 1971) supplemented by Searle (1971). In our working and business lives, we write to achieve something for other people. Our texts are basically conversations and are usually integrated into actual conversations: A sales letter is followed by a phone call, the phone call is followed by an offer, the offer is followed by another phone all ( Jäger, 2007: 82, translated by WKG). This example illustrates that professional writing gives rise to actions such as telephoning but also has legal consequences, which will be discussed in more detail later. In general, there are communication situations in everyday working life that are initiated and kept going orally and in writing. The following “maxims of conversation” (Grice, 1975: 47) must be observed, especially in everyday professional life: • Maxim of quantity: Make your contribution as informative as necessary—no more no less! • Maxim of quality: Don’t say anything you think is wrong! • Maxim of relation: Be relevant with your contributions to the conversation! • Maxim of modality: Speak clearly and understandably! ” (Heinemann, 2008: 122 f., translated by WKG). Professional writing occasions are often “advertising, criticizing, apologizing, contradicting, rejecting, instructing, and lecturing” ( Jäger, 2007: 82, translated by WKG). Jäger (2007) highlights these seven professional writing occasions, which may include prototypical text types (see Fig. 15). Thus, texts are created in specific contexts and discourses: “Contexts control discourse production and comprehension” (Dijk, 2008: 17). 72 2 Theory <?page no="73"?> Fig. 15: Professional writing occasions (after Jäger, 2007: 82 ff., own illustration) The discourse is part of the context. However, it is important to bear in mind that every writer and reader, even if they know the context and belong to a discourse community, have different mental representations: Although in most forms of discourse between members of the same community mental models will be sufficiently similar to guarantee successful communication, it should be stressed that mental models necessarily embody personal elements that make all discourse productions and interpretations unique - and hence misunderstanding possible - even when they have many, socially shared, elements (Dijk, 2008: 60). The reason for the differences and, thus, also possible agreements or disagree‐ ments is the wealth of experience: […] writers and recipients by definition have different models of the same communi‐ cative event; such differences may lead to negotiations about the shared aspects of their context models, but also to misunderstanding and conflict […], in other words, context models are the interface between society, situation and discourse (Dijk, 2008: 72). Every subject approaches the world differently, which also applies to writing: “Mental models are the cognitive representations of our experiences” (Dijk, 2.4 Text (Type) competence in the professional field 73 <?page no="74"?> 2008: 61). Therefore, as previously discussed, the definition of a text type is difficult: “The most straightforward definition might be that genre is a type of text or talk, or, more broadly, of verbal activity or communicative event” (Djik, 2008: 148). A text type is rather the essence of a discourse and context. Both the text type unsolicited offer and the letter of application can be assigned to the behavioral situation of advertising, which can be assigned to the “appeal function” through the demand “[…] to carry out actions or speech acts […]” (Heinemann, 2008: 139, translated by WKG). The three steps of promise, proof, and encouragement are implicitly written down in these two text types through a bundle of different content-related aspects, which will be discussed below by presenting the two text types: unsolicited offer and cover letter. Both text types listed are presented in an official letter. The corresponding text class is the business letter (Fix, U., 2008: 32; Marquardt, 2009: 10). The letter is a communication medium increasingly moving away from the traditional postal service and towards e-mail, but will never completely disappear due to legal requirements. A distinction is made between unsolicited (usually advertising letters, also known as mailings) and solicited (sent on request) offers. Unsolicited offers are also often sent by companies as e-mails. The recipient is not explicitly named. From a legal point of view, this form, with the initial declaration of intent it contains, is not an offer but an advertisement that does not contain a declaration of intent: An offer should name all services, specify the time frame in which they can be provided and clearly describe the contributions of all parties involved. It also sets out the cost framework for the client. Anything that is forgotten in the offer is likely to lead to disputes later on (Baumert, 2008: 172, translated by WKG). The offer is legally binding and can represent a purchase contract’s the first or second declaration of intent. An offer is part of the behavioral step of advertising. The reader should, therefore, be addressed with an attractive promise. The enumeration of benefits and the encouragement to take advantage of the offer are fundamental in an offer. The sales letter, in particular, refers to a company’s product or service and invites the reader to purchase a product or service: In order for the letter to perform its tasks, it must fulfill numerous requirements. Advertising belongs to the category of ‘unloved reading material’ and must therefore deliver the right arguments in a matter of seconds in order to arouse the recipient’s interest (Hirschi, 2011: 51, translated by WKG). 74 2 Theory <?page no="75"?> The success of a mailing is measured by the response rate, i.e., the response from the buyer and the associated second declaration of intent. A response rate of around 5 % indicates that the mailing was successful. “If you don’t reach your customers and lose readers, you lose money” (Baumert, 2008: 1, translated by WKG). According to Hirschi (2011: 52 f.), the following points should be considered as building blocks in a promotional letter (see Fig. 16): the objectives, the target group, the mix of offers, the core idea, the response media, and the content-related mailing components. Hirschi emphasizes the target group once again, as the language formulation must match the offer and therefore also the target group (Hirschi, 2011: 53). This requires a creative use of language and a feel for language, as the letters can be created based on text modules, but must take the target group into account. “Professional authors take the reader’s expectations into account and tie in with them. Even texts that communicate something new to the reader must pick them up in their familiar world” (Baumert, 2008: 7, translated by WKG). Fig. 16: Building blocks of the mailing concept (after Hirschi, 2011: 52 f., own illustration, translated using AI) To increase the attention of the target group, the beginning of the letter should be written using rhetorical techniques, e.g., the use of questions and prompts, the if-then technique, formulation as a message, or the inclusion of the target group (Hirschi, 2011: 56). In the central part of the advertising letter, the AIDA formula has become established as a structuring aid for copywriting. This contains the elements Attention, Interest, Desire, and Action (Lewis, 1903): First of all, attention must be created so that the reader is drawn deeper into the letter. This can be done with a successful headline, for example. Then interest must be aroused, i.e., the reader must be made curious about the offer so that they are keen to find out more. This should happen in the first section of the sales letter. Subsequently, a desire to buy must be generated. To do this, the offer should be described as favorably as possible in the middle section of the letter. This can be done in the salutation of 2.4 Text (Type) competence in the professional field 75 <?page no="76"?> the letter or even in a trailing P. S. (Post Scriptum) (Hirschi, 2011: 57, translated by WKG). The letter should not contain too many difficult or even new words to remain understandable. Furthermore, the necessary information must be included, but superfluous information must be avoided. For this reason, offers are often limited to one page so that the working memory is not overloaded when reading: “With around seven to nine bundles of information - the technical term is cluster - this part of the memory has reached its limits” (Baumert, 2008: 8, translated by WKG). According to Hirschi (2011), concise sentences, few passive formulations, and few auxiliary verbs should also be used. Positive adjectives, structuring punctuation marks such as dashes or colons, impeccable spelling, and an appealing letter design round off a successful advertising letter. The letter can end with the greeting “Yours sincerely”; however, emotionally engaging phrases include “Warm regards,” “Summer greetings from,” or “Kind regards.” The P.S. should highlight the most important benefit. Emphasis is permitted for better readability (Hirschi, 2011: 59 ff., translated by WKG). The application informs the prospective employer about the applicant’s skills and experience. Its aim is to convince the hiring manager to invite the applicant to an interview. An application consists of a cover letter and a cover sheet with a photo, a tabular CV, and certificates. A cover letter is the letter that accompanies the application. The cover letter can be written in response to a job advertisement in the daily newspaper or by searching in job search engines (e.g., at the employment agency). Alternatively, you can apply on your initiative, as not all companies advertise vacancies. The cover letter is part of the advertising step. Therefore, an attractive promise with the naming of advantages and the benefits of the person for the company should be the three sub-steps of an application letter. Professional and personal skills should be presented in the best possible way. The cover letter is, therefore, the ‘entry ticket’ to a new company. Cover letters are subject to the influence of new media. Until 20 years ago, sending application letters with the application folder (CV, copies of certificates, and supporting documents) by post or handing them in person was common practice. Nowadays, an increasing number of online portals, e.g., those provided by employment agencies, allow applicants to create a single CV, upload it, and make their application visible to several companies at once. Alternatively, the uploaded application documents can be sent as a link to several companies. However, even in this case, the cover letter must be adapted to the relevant company and uploaded to the application portal as a single file with the applicant’s signature. This step, in turn, requires an application cover letter to be written following DIN standard 5008. It is also legally relevant, as every 76 2 Theory <?page no="77"?> applicant must submit the application cover letter, and the company must keep the cover letter, even if the applicant is not selected, in case of potential legal disputes (Wisskirchen, 2006). Compliance with the formal requirements for these letters is therefore essential. Both types of text, the offer and the cover letter, are subject to the textual pattern of argumentation, as the addressee must be persuaded either to invite the applicant for a job interview or book a package deal. The text function is the appeal function, as the reader is prompted to take the action of inviting (in the case of the cover letter) or booking (in the case of the package deal). The power of the appeal is asserted through arguments in the respective letters. In the case of the cover letter, this is achieved through textual features such as personal skills and abilities, academic achievements, experience through internships, or the expression of interest in the company. In the unsolicited offer, the content arguments include the price, the offer’s details, and the hotel’s location. Both letters are convincing not only because of the content of the text but also because of the linguistic text features such as correct grammar and spelling or the choice of words. However, the chosen layout, i.e., the formal conditions of the letter, also contribute to its persuasiveness. According to Grice, the quantity of text as a maxim of conversation is limited to one page for both text types. Both types of text just described are written following DIN 5008 (Deutsches Institut für Normung e. V., n.d.; Grün, 2013) in Germany ( Jäger, 2007: 141; Neumayer & Rudolph, 2008: 74): This set of rules has a good intention: it sets clear patterns of recognition, it teaches uniform standards in schools, it facilitates written communication. In the days of the typewriter, when page layout was an ordeal, this standard provided the tools to save the day ( Jäger, 2007: 141, translated by WKG). In the meantime, computers have replaced typewriters, but the standard remains as a guide to how a business letter should be written. This standard specifies, for example, line spacing, paragraphs, placement of the date or signature, and common text modules. Compliance with the DIN 5008 standard is a sign of professionalism, especially in a professional environment: Although first impressions may often be wrong, they are the most important. After all, it is the one that is remembered the longest, and it shapes long-term attitudes towards a person, an organization or a cause. This is particularly true for written documents and letters: one wrong salutation, one clumsy presentation - and the foundation can be laid for a business relationship to fail as soon as it has been initiated (Grün, 2013: Foreword, translated by WKG). 2.4 Text (Type) competence in the professional field 77 <?page no="78"?> The recipient of a business letter expects a certain standard to orient himself and decipher the letter’s content. An example of the relevance of the DIN standard in the professional envi‐ ronment: The personnel manager receives a pile of application letters for a training position in the company. In addition to the usual application letters, there is a handwritten note on which it is written in two sentences that the applicant would like to be offered the training position. Would you give the applicant the chance of an interview? The personnel manager has little time to invite suitable applicants from the pile to an initial interview. Application letters that are not recognizable as such fall through the cracks, because the company wants to be taken seriously and, in addition to the language and content requirements in such a letter, compliance with the standard is an answer to the question of whether the applicant has made an effort for the company. Companies, in turn, expect letters that meet these writing standards also for economic reasons - namely, to reduce the time needed for reading and to make the letter’s content relevant for professional action. This example clarifies that business letters have legal relevance and that postal transmission is a prerequisite for legal compliance in some situations (Baumert, 2008: 47). By sending an e-mail or fax, the sender does not see who receives it. The secrecy of correspondence enshrined in the Basic Law can only be maintained by post or by personal delivery (Neumayer & Rudolph, 2008: 60). The challenge for business letters is adhering to the rules set out in DIN standard 5008 on the one hand, while, on the other hand, adding a touch of creativity, particularly in the case of cover letters and unsolicited offers. Ultimately, the aim is to ensure that the recipient quickly understands the letter and is persuaded by both types of text—to invite the applicant for an interview and to accept the offer as a contract with the second declaration of intent. In both cases, the recipient must be convinced by the advertising function of the text. This is where emotions come into play, because both writing and reading are accompanied by emotions ( Jäger, 2007: 21). If the reader’s interest is aroused by a linguistically emotional bridge, he will continue reading the letter and, in the best case, take action. For this to succeed, the writer must plan this effect in a targeted manner. Professional writing, like academic writing, requires several planning steps. An experienced writer will no longer write these down but clarify them mentally. These are the questions of impact, the scope of the letter, previous knowledge 78 2 Theory <?page no="79"?> of the text type, the question of quality and the degree of creativity, knowledge of the addressee, and ultimately, the answer to the question of who is writing the letter ( Jäger 2007: 29 f.; Fix, 2008: 26). In this context, the writer must write in a highly addressee-oriented way. The lyricist has had numerous communicative experiences, he knows in which situation, under which conditions his actions were communicatively successful (and when they were not). They can therefore call up a large number of patterns from their memory, as patterns of COMMUNICATIVE BASIC TYPES, which mark the social functions of communicative action per se (Heinemann, 2008: 142, translated by WKG). It must also be emphasized that professional writing is often carried out in teams. This includes generating ideas, proofreading, or even writing together on a document, which is underlined by two quotations: Professional authors are rarely lonely poets who write in their attic, isolated from the outside world. They often work in teams under conditions that are familiar elsewhere in the business world. Delivering, processing, refining, ensuring quality, coordinating the process: Many are involved in creating a document (Baumert, 2008: 179, translated by WKG). Writing in the workplace is part of a social network. Writers work in cooperation with others. The writing process is embedded in an overarching project. Writers use other texts (e.g., applications) as a template for writing. Writing is abstracted from the person, the text is an institutional product and speaks the ‘language of the company’ (Lehnen & Schindler, 2010: 154, translated by WKG). For professional writing, sample texts and document templates should also be used to support the generation of ideas and revision (Baumert, 2008: 118). To develop rules for editing, a group can look at older texts or those of competitors and discuss them. Everyone says what they don’t like, then the team forms an opinion on how to proceed with such cases. The result is recorded (Baumert, 2008: 103, translated by WKG). This is where the degree of professionalization becomes apparent. Texts that are important for a company are not planned and written alone. They go through several revision phases. Ultimately, it is not only business letters, such as unsolicited offers or complaints created at work, but also documents such as lists, meeting notes, and delivery bills that characterize everyday working life. Nowadays, employees have to deal with various professional text types receptively and productively. “Written formats are still regarded as the most important means of describing 2.4 Text (Type) competence in the professional field 79 <?page no="80"?> and communicating knowledge” ( Jakobs, 2008: 2, translated by WKG). This is due to the multitude of requirements in the world of work, which are visually represented in Figure 17, in which space has been intentionally left for further pieces of the puzzle, because each company defines its own requirements. Fig. 17: Factors for correspondence (after Jakobs, 2008: 2 f., own illustration) Furthermore, a high degree of work standardization can be observed by estab‐ lishing checklists as a means of quality assurance in the workplace. “The written recording of work processes and products makes work comparable and assessable” (ibid.) and thus represents a degree of professionalization. The use of new media, such as PCs, cell phones, and intranets, is also transforming the communication channel from face-to-face interactions and, thus, primary business agreements to written communication via emails, chats, and text messages. As a result, the industrialization of communication work and the degree of legal protection in the workplace and the business sector 80 2 Theory <?page no="81"?> are increasing. For example, written documentation, such as letters, serves as evidence of this change (ibid.). Writing also means sharing - communicating. The letter in text form emphasizes the sender-receiver principle. The message is communicated to a recipient through the letter (Koch & Pielow, 1984: 163). It should be remembered that text types originally come from an economic context and are oriented towards it. Applicants, for example, even ninth-grade students, now have to enclose an application portfolio and cover letter - this is not prescribed by didactics but is a requirement of the economy. Ultimately, every employee will have to deal with writing their application, not just once or twice in their life, but on an ongoing basis due to the increasingly frequent change of companies (IWD, 2016). At the same time, students are not only career starters but also consumers. They are exposed to advertising almost daily, at the bus stop, on the bus, on the radio, on TV, in emails, or even in the form of letters. Both types of text, the offer and the cover letter, prompt action from the addressee: The applicant seeks to be hired by the company, which, in turn, may accept or reject the application. Nevertheless, the addressee will act. Similarly, the consumer can respond to the offer by submitting his second declaration of intent, or decline and not reply, which is also carried out as part of an action (Koch & Pielow, 1984: 41). Ultimately, the “writing of text types” (Dehn et al., 2011: 77, translated by WKG) is central to the development of text type competence, which leads to explicit text type competence. Even before text types are taught in class, pupils have implicit knowledge of them based on receptive experience in their every‐ day lives. This wealth of experience can be used didactically and methodically for the entire writing process in German lessons. The next subchapter, therefore, presents the writing process and the associated writing models as sub-areas of writing research. 2.5 WRITING PROCESS In this subchapter, cognitive writing models that are important for writing les‐ sons are first presented chronologically. The focus is on answering how writing is done and what influences the writing process. Furthermore, references are made to the researched writing strategies, which should not be considered separately from the writing models but have had and continue to have a synthetic influence on model formation. 2.5 Writing process 81 <?page no="82"?> Feilke (2017: 164, translated by WKG) divides writing process orientation into “process components”, “recursivity”, “writing strategies” and “sociality”, which must be taught in writing lessons. By definition, strategies can have a process-related, cognitive, or metacognitive attribution (Sturm, 2017: 267 f.; see Fig. 18). Fig. 18: Strategy terms according to Sturm (2017: 267 f., own illustration) The process-related concept of strategy focuses on the process components of the writing process. The cognitive writing process focuses on the generation of ideas, reflection, and mental planning of the writing process, and the metacognitive concept of strategy refers to the strategy actions that rely on recursivity through the conscious procedural control of automated strategy forms. The definition and categorization of strategy concepts today can be attributed to relatively recent research on writing processes. This brief flashback intends to shed light on this development and illustrate how the writing process has been researched with the anchored writing strategies. Nystrand (2006: 11 ff.) summarizes the history of writing research as follows: The start of empirical research on writing can be located in North America in the 1970s. Researchers from the fields of philosophy, such as Langer and Cassirer, and psychology, including Bruner, Piaget and Vygotsky, joined forces. The cognitive revolution was brought about by the realization that writing is also reading and learning. This turning point created a new direction of research in psycholinguistics for behaviorists such as Skinner. 82 2 Theory <?page no="83"?> Another driver was attention to the low writing performance of students from disadvantaged backgrounds. In 1975, Harvard University introduced the famous “Tuesday Colloquium” (Nystrand, 2006: 13) for college students to support students, especially those from African and Puerto Rican immigrant backgrounds, in academic writing. Later on, US President Nixon has supported cooperative writing research projects at the National Institute of Education (NIE). The first funded writing research was the work of Hayes and Flower at Carnegie Mellon University. Hayes expanded his understanding of cognitive problem-solving and the role of writing as a problem-solving process through his collaboration with Nobel Prize winning psychologist Herbert A. Simon. Furthermore, the director of City College’s Basic Writing Program, Saughnessy, published the writing mistakes of 4000 college students. She proclaimed the new credo of writing process research: “[…] writing is a social act” (Saughnessy, 1977: 83). Due to the social development in America, this resulted in the research col‐ laboration between Hayes and Flower regarding the cognitive writing process model (Hayes & Flower, 1980: 10, see Fig. 19). In this model, the writer’s world is divided into the areas of “Task Environment” (ibid.), “Writers’ Long-Term Memory” (ibid.) and “Writing Process” (ibid.). The three main processes in writ‐ ing are the sub-activities “Planning” (ibid.), “Translating” (ibid.) and “Reviewing” (ibid.). They saw this model as provisional. This fact is essential to be able to categorize the further development. Fig. 19: Writing process model (Hayes & Flower 1980: 10) 2.5 Writing process 83 <?page no="84"?> The basis for creating this model was observing students who verbalized their thoughts while writing. These think-aloud protocols were fundamental for a new perspective in writing research, because the focus was not on the writing product, but on the writing process: “[…] subjects are asked to say aloud everything they think and everything that occurs to them while performing the task, no matter how trivial it may seem” (Hayes & Flower 1980: 5). The US researchers Emig and Newell and Simon were already working with this method in the 1970s. However, the highlighted sub-actions should not be considered separately; rather, writers were observed performing all three actions in turn. They plan and revise while writing. These sub-actions are mental cognitive actions that cannot always be observed from the outside (Hayes & Flower, 1980: 10 ff.). Longer writing plans are primarily for more complex writing tasks that have not yet been automated (Becker-Mrotzek & Böttcher, 2006: 29). Long-term memory is emphasized in the writing process model. This is utilized to a greater extent during writing planning if the topic is unknown, the addressees are unclear, or, as already explained, there is no writing plan (Hayes & Flower, 1980: 10 ff.). The writing task itself also affects the text planning phase. For example, the topic, the addressees of the text, or even the motivation can negatively or positively influence the writing process (Hayes & Flower, 1980: 11). This model shows that the writer is still observed alone. Long-term memory and the writing task, in particular, are parameters that have a positive or negative effect on the planning of the text. “Memory is the structure that serves the storage and retrieval of information” (Winkel/ Pe‐ termann & Petermann, 2006: 30, translated by WKG). Long-term memory stores all information that is consciously and unconsciously absorbed. All three sub-actions of writing are intertwined and are not separate from each other or even as stages but can be observed in parallel (Hayes & Flower, 1980: 27 ff.). Writing must be seen as a process with many parallel operations. Thus, the writer can also be seen as a “switchboard operator” (Flower & Hayes, 1980: 33). Three factors inhibit the writing process: 1. lack of knowledge, 2. lack of linguistic conventions of the written text, and 3. lack of rhetoric. To overcome these obstacles, the number of tasks to be mastered must be reduced so as not to overload the short-term memory because no writer can pay attention to all three things simultaneously. Instead, the writer has to ignore some things in the process and catch up later in the revision. This is why plans are essential for text quality, especially for struggling writers, if writing is not yet routine or automated. Focusing on the task and 84 2 Theory <?page no="85"?> 7 According to Trappen (2013: 173), there is a five-stage sequence: Inventio, Dispositio, Elocutio, Res and Verba, Memoria and Actio, which the rhetorician Cicero already used. the writing plans are important teaching contents for overcoming writing inhibitions (Flower & Hayes, 1980: 40 ff.). From then on, there were and are new areas of investigation, e.g., the context of writing, the relationship between writing and reading (Bereiter & Scardamalia, 1987), the relationship between writer and discourse community, and writing in the discipline. In addition to the academic writing context, new fields of study were opened, including writing in the workplace and industry. The journal, Written Communication, was first published in 1984 and is still a medium for new results in writing research today (Nystrand, 2006). In the 1980s, researchers Bereiter and Scardamalia (1987) categorized writing according to two strategies: knowledge telling and knowledge transforming. The knowledge telling strategy involves writing down the development of thoughts rather than linguistic skills such as syntax. The application of this strategy is similar to expressive writing. The goals are often not elaborated. According to Sturm, it is used by inexperienced writers and even by authors such as Aldous Huxley. Therefore, this type of strategy is classified more as a cognitive strategy (Sturm, 2017: 269). The knowledge transforming strategy involves the development of thoughts on the one hand and the development of the text on the other. This can only be achieved if the reader’s perspective is adopted, which requires a text representation on a mental level. As a result, this is a metacognitive strategy. The two researchers state that writing strategies are crucial for novice writers. An experienced writer can tactfully use the knowledge telling strategy when revising their text. Novice writers, on the other hand, try to finish their text as quickly as possible and revise it less (Bereiter & Scardamalia, 1987: 3 ff.). In 1981, de Beaugrande and Dressler (1981: 36) also published phases of the writing process in Germany, which, although less well known internationally, are quite comprehensible in their logic and are based on rhetoric: 7 “There is no point at which production is finally completed, but at most a FINAL STAGE” (de Beaugrande & Dressler, 1981: 36). The text quality assessment depends on the text recipient’s image and their level of text processing. Every text can always be revised or further analyzed. The writing process can be divided into several phases, beginning with the planning phase. Following the identification of the objective and choice of text type comes the “phase of IDEATION” (de Beaugrande & Dressler, 1981: 42, 2.5 Writing process 85 <?page no="86"?> translated by WKG) which corresponds to inventio (idea generation) in classical rhetoric: An IDEA is an internally created (i.e., not environmentally forced) design of content that provides CONTROL MEANS (e.g., topics for creative, meaningful behavior including text production) (ibid.). In this phase, writers search through stored “KNOWLEDGE SPACES” (de Beaugrande & Dressler, 1981: 43, translated by WKG), followed by a phase concerned with finding appropriate forms of expression, comparable to the dispositio in rhetoric. This involves retrieving already stored knowledge and the associated expressions. The last phase, the “GRAMMATIC SYNTHESIS” (de Beaugrande & Dressler, 1981: 44 f., translated by WKG) follows on the surface of the text. Experienced writers often require less time while demonstrating a higher processing intensity and greater expertise. “This effect would explain why a skilled text producer can improve the texts of others (and not just his own) without actually having been involved in their mental processes” (ibid.). “Self-corrections often appear to be more difficult for the text producer himself because he already knows the intended ideas and overlooks cases of inefficient expression, even correct errors, which is not the case for external corrections” (ibid.). Referring to the works of Bereiter and Scardamlia, and Hayes and Flower, Kellogg (1994) conducts research in writing psychology. He emphasizes the diversity of factors in the writing process and schemata as a representation of knowledge: Thinking involves a set of mental skills that create, manipulate, and communicate to others the personal symbols of mental life. […] Second, thinking is a process that entails manipulating representations of what one knows about the world […]. Third, what one thinks or knows about a topic is often, though not always, directed at the solution of specific tangible goals (Kellogg, 1994: 10). Scripts are also counted as schemata. They are routinized activities (Kellogg, 1994: 20). As early as 1981, de Beaugrande and Dressler refer to this and define it more precisely: Schemata are global patterns of events and states in ordered sequences, with the main connections being temporal proximity and causality (de Beaugrande & Dressler, 1981: 95, translated by WKG). 86 2 Theory <?page no="87"?> PLANS are global patterns of events and states that lead to an intended GOAL (ibid.). SCRIPTS are stabilized plans that are frequently recalled to determine the roles and expected actions of communication participants (de Beaugrande & Dressler, 1981: 96, translated by WKG). Cognitive operations such as collecting information, planning ideas, writing, and revising are essential to increasing the quality of a text (Kellogg, 1994: 16). If the number of co-writers also increases, e.g., through feedback from classmates, knowledge transforming is generated, as the knowledge about a text changes as a result of the writing stimuli. This exchange leads to new ideas (Kellogg, 1994: 33). This is a condition of the writing environment - the task environment. In this publication, Kellogg gives advice that favors the writing process in students: 1. Fast writing does not guarantee writing quality (ibid.). 2. Writing on a PC is more convenient for revision than writing by hand, which is why people tend to plan and revise much more during the writing process (ibid.). 3. The writing task must not be too easy (tedious) or too difficult (frustrating). At best, the writing task can encourage motivation to write. If the task is important to the writer, they will write well. It is not extrinsic motivation that works here, but intrinsic motivation. Knowledge of a task does not always result in better texts. The effort to complete a task and knowing little about it can be more of a motivational factor, producing better texts (ibid.). 4. Content knowledge about a text type has an impact on writing perform‐ ance. Writing attitude, conceptual knowledge about the world, and meta‐ cognitive knowledge about strategies, tasks, and goals are also parameters of writing success (ibid.). 5. Writing experts are more likely to revise their texts than writing novices, and they allow themselves more writing time due to the revisions to be made (ibid.). If these five pointers are considered, the student’s writing productivity can be boosted. In his research, Kellogg proved that the number of words correlates with text quality, although he points out that the length of a text can also have limits in connection with text quality (Kellogg, 1994: 59 ff.). In 1996, the innovative writing model from 1980 was further developed. Hayes (1996: 1 ff.) no longer focused on the writer alone but also integrated the writing environment more strongly into the “individual-environmental model” (Hayes, 1996; see Fig. 20, translated by WKG). In this writing process model, there are social factors, (e.g., the addressees, co-actors), and physical influencing factors, 2.5 Writing process 87 <?page no="88"?> (e.g., the text draft and the writing tools), and new media. Four internal factors also influence the individual writing process: 1. motivation/ affect, 2. working memory, 3. cognitive processes, and 4. long-term memory (Hayes, 1996: 4). Working memory takes on a new significance that had not yet been consid‐ ered in the classic 1980 model. Although the other factors such as long-term memory, motivation, and task were already integrated in 1980, further research was available in 1996 that confirmed their position more strongly. Fig. 20: Writing model as an ‘individual-environmental model’ (Hayes, 1996) 88 2 Theory <?page no="89"?> In psychology, working memory is assigned to short-term memory, as shown in Figure 21 and explained in the following definitions as an excursus: The three-memory model differentiates between ultra-short-term, short-term and long-term memory. These forms of memory differ according to their storage capacity and duration. The multiple memory model postulates a differentiated structure of the working memory with substructures that are responsible for specific content (verbal vs. pictorial material) and are coordinated by a central executive. The depth of processing model makes the stability of memory content dependent on the intensity or depth of cognitive processing. Depending on the type of information to be processed, a distinction is made between declarative and non-declarative memory. Semantic and episodic content is stored in declarative or explicit memory, while non-declarative memory includes unconscious processes such as priming, conditioning and procedural learning (Winkel/ Petermann & Petermann, 2006: 39, translated by WKG). Fig. 21: Three-memory model (Winkel/ Petermann & Petermann, 2006: 36, translated by WKG) The result of cognitive learning processes is knowledge about the world and about oneself that is accessible to the conscious mind and can be called up directly (semantic or declarative knowledge). Semantic knowledge is represented in the form of concepts and the relationships that link them. The representation of semantic knowledge can be imagined as a network in which the terms (concepts) represent the nodes and the edges the relationships between them (Winkel/ Petermann & Petermann, 2006: 161, translated by WKG). 2.5 Writing process 89 <?page no="90"?> The writing task (task environment) does justice to writing as a social act in this model. The collaborations at school, in the workplace, and references to the research of Hull, Greene, and Haas in the late 1980s and 1990s, confirm the social influence on the individual writing process. The physical influences are, on the one hand, computer programs, but also the medium itself. Revising a text on a computer is easier than with pen and paper. In her research in the 1990s, O’Hara showed that the change in the writing task significantly influences cognitive processes. The individual on the second level of Hayes’ writing model is represented by working memory and motivation as influencing factors on the writing process. Baddeley’s (1986) research on working memory also revealed that working memory has only limited resources for searching for and processing information. Two ways of memorizing would be in the foreground: as a phonological loop, quasi the inner voice, or as a visual spatial sketchpad. In a supplementary study (Kaufer, Hayes & Flower, 1986), Hayes points out that competent writers initially form phrases not complete sentences. The working memory then becomes active as writers revise grammatical or semantic problems after the initial drafting. This process leads to pauses during writing and results in increased writing time. Motivation, as an additional influencing factor, was also accounted for in 1980, but not as prominantly as in the 1996 model. According to Hayes (1996, 14), the study by Dweck (1986) demonstrates that beliefs influence writers’ actions. When individuals view writing as a gift, manifested beliefs can improve. The interplay of individual reading and writing skills is a motivating factor; after all, the text is revised through repeated reading, quasi “read to revise” (Hayes, 1996: 14). In this context, Hayes refers to the research work of Hayes, Flower, Schriver, Stratman and Carey from 1987. Long-term memory was added as a new component. Task schemas are a bun‐ dle of information, including goals and criteria. Knowledge incorporates aspects such as grammar, text type, topic, or target audience, among others. Writers with strong writing abilities typically possess extensive writing experience through intensive practice as well as the competence to revise texts effectively: “Many years of practice may also be required to attain expert performance in any of the genres of writing”, according to Hayes (1996: 26). In this model, the aforementioned factors of the writing process are inter‐ twined. Although they focus on the writer alone, this is because Hayes, as a psychologist, is interested in the individual (Hayes, 1996: 1 ff.). While the 1980 model was provisional, the 1996 model is more scientifically sound due to tested hypotheses, according to Hayes (ibid.). 90 2 Theory <?page no="91"?> Example: If the writer is motivated by the writing task, positive emotions are released. These have a positive influence on the working memory, because the writer is willing to solve the writing task and intensifies their work performance. This sets cognitive processes in motion that repeatedly draw on long-term memory. Knowledge of text types and the associated text function and possible target audience are retrieved and compared to solve the writing task. It does not even have to be a familiar text type. In the same year, the research team Rijlaarsdam & van den Bergh (1996: 207 ff.) published findings on writing process research. The study with ninth graders (n = 22) revealed that revisions tend to be made towards the end of the writing process. Using think-aloud protocols, it was found that writers employ different strategies when writing. Writers who structure extensively in the first part, i.e., generate, read, and revise ideas, maintain this approach in the second part. Conversely, writers who are less structured typically continue in the same manner throughout the writing process. The two researchers observed that cognitive processes cannot always be made visible by speaking quietly. Therefore, observation and verbalization are not always accurate. However, it can be stated that text quality is high if the cognitive activity can be observed by structuring the text. However, the writers are quite different in their writing time. In 2000, Fix (2000: 325) developed the “Writing Didactic Cube” for the “conditions of the school writing process” (ibid., see Fig. 22, translated by WKG). Fig. 22: Writing didactic cube (Fix, 2000: 324 f., translated using AI) 2.5 Writing process 91 <?page no="92"?> He divides the sides of the cube into three dimensions. The first is an “object-re‐ lated dimension” (ibid.), which is influenced by the writing function, whether one writes for oneself or others. The second, the “subject-related dimension” (ibid.), is influenced by the “inner writing situation” (ibid.) with the “[…] cognitive, emotional, motivational, volitional, and situational conditions […]” (ibid.) of the writer. The third dimension is the “instruction and guidelines provided by the teacher” (ibid.). The writer may receive strong guidance, e.g., for writing a CV or in self-directed writing (Fix, 2000: 325). Emphasis should be given to the aspect of self-directed writing that requires less teacher interaction. In 2006, Hayes published research findings that he believes were ground‐ breaking for further developing his writing process model. Research on working memory by Chenoweth and Hayes (2001) and Kellogg (1996) showed that working memory is limited. This means that a writer cannot concentrate on all parts of revising simultaneously. Phonological loops such as reading aloud quietly, speaking aloud, or repeating favor attention, especially in the revision phase of the text (Hayes, 2006: 28 ff.). Free writing as a strategy for generating ideas revealed that most ideas come from unorganized notes and less from completed or unfinished sentences or organized notes. In contrast, outline planning led to a higher word count in the essays and, interestingly, to more ideas than the test subjects who wrote straight away. This sparked a debate among writing psychologists in Europe and the USA about suitable writing strategies for writing didactics that continues to this day. In 2006, Rijlaarsdam and van den Bergh (2006: 41 ff.) published their research results on the writing process. Using thinking-aloud protocols, 15-year-old pupils were observed based on two argumentative essays. They divide the writing process into three temporal writing phases - the beginning, middle, and end of writing. In the first phase, the task tends to be read. Repeatedly rereading the task is an indicator of poor text quality. However, revisiting the task at the end tends to lead to improved writing performance. It was also observed that repeatedly reading the resulting text led to the input of new ideas, and that this was used positively in the second writing phase in particular. Nevertheless, it was challenging to capture the different strategies of the writers, as they were very different and would also change constantly during the process. One conclusion is that repeated reading, evaluating, and revising would improve text quality, especially at the end, i.e., in the final writing phase. Writers with limited knowledge about a topic are more likely to generate their ideas at the beginning of the writing process, whereas competent writers can generate new ideas throughout the writing process. 92 2 Theory <?page no="93"?> Kellogg published notes on writing instruction in 2008. To write a coherent text, the writer must know the text’s content that is anchored in long-term memory. He categorizes the development of writing skills into three stages, which are based on Bereiter’s writing development model and have been further developed (see Fig. 23). The addressee orientation, the final stage, is new. It characterizes writing experts. Stages one to three can continue into adulthood and must be practiced constantly. Planning, writing, and revising require self-regulation of cognition, emotion, and behavior. Referring to Hayes and Bereiter, he emphasizes the significant role of planning and revising. However, even adolescents usually fail to reach this stage, as working memory is limited. Mostly local, but less global, revisions are carried out, which can even be observed among college students: “In knowledge-crafting, the reader’s interpretation of the text must feed back to the way the text reads to the author and to the message the author wishes to convey in the first place” (Kellogg, 2008: 9). In this phase, revisions are necessary at all levels. Writing development does not end with a university degree. Fig. 23: Writing ability development according to Kellogg (2008: 3, own illustration) Since 2012, the latest internationally recognized writing model for mapping the writing process by Hayes (2012: 371) has focused on the process-oriented nature of writing. It is a multi-level model based on pedagogical-psychological studies. The older writing models were thus further developed and reorganized in terms of content (see Fig. 24). This model emphasizes the factors collaborator & critics, transcribing tech‐ nology, and writing schemas. Schemas or so-called frames are “conceptual subsystems of our knowledge” (Fix, 2008: 25, translated by WKG) - patterns that are recalled by the brain as typical (ibid.). Text patterns or text type knowledge are also part of this (Heinemann, 2008: 143). 2.5 Writing process 93 <?page no="94"?> At the resource level, attention was added alongside long-term memory. The process level plays a vital role through the co-actors, now referred to as collaborators. At this level, another aspect comes into view - transcribing technology. Research findings by Berninger et al. (1994) show that motor writing does not happen automatically, as assumed, but is automated as part of the learning process. In this context, Hayes (2012: 372) refers to the work of Jones and Christensen (1999), who improved the text quality of young pupils through writing training. At the third level, the control level, motivation has an even more significant influence than before in the models. Hayes (2012: 372 f.) argues that motivation influences all components of the writing process. The writing time and, therefore, the text quality improve with increased motivation. The term ‘motivation’ is not further differentiated in this model, unlike it was in the 1996 model (see Fig. 20). Fig. 24: Writing process model (Hayes, 2012: 371) Hayes also refers to Bereiter’s knowledge-telling and knowledge-transforming strategy (Bereiter & Scardamalia, 1987). The second strategy involves extensive revisions, e.g., whether the text’s information has been elaborated enough. He also assumes that the telling and transforming strategy changes as classes 94 2 Theory <?page no="95"?> increase. The higher the grade level, the more likely the transforming strategy will increase. This is partly because handwriting is more pronounced, or typing on a PC is automated, allowing the working memory to concentrate on other control points. This model in particular makes it clear that, in addition to writing, several processes take place in parallel at the control, process, and resource level, The writing process is therefore a complex cognitive challenge (ibid.). Many aspects of the writing process are interdependent and must be viewed holistically. In German-speaking countries, the basic model of text production introduced by Bachmann and Becker-Mrotzek (2017: 25 ff.) was published in 2017 based on existing cognitive writing models (see Fig. 25). It is a theoretical model and, in addition to cognitive processes, also emphasizes resources and processes at a linguistic level: The basic conditions, personal characteristics and resources for the respective writing occasion are located in the outer triangle (Bachmann & Becker-Mrotzek, 2017: 43, translated by WKG). The inner triangle shows the four components that are activated and developed through the interaction of the stable outer components: prototypical writing tasks and text patterns on the one hand, adapted writing tasks and text patterns on the other (ibid.). The circle within the two concentric triangles represents the dimension of basal linguistic processes. These convert into text everything that has been recalled, activated, adapted and supplemented in long-term and working memory in terms of prototypical and adapted task and text patterns for processing the specific writing task (ibid.). The rectangle in the center of the diagram stands for the linguistic products towards which all components and processes of text production are oriented: intended text, draft text and text (ibid.). 2.5 Writing process 95 <?page no="96"?> Fig. 25: Basic model of text production (Bachmann & Becker-Mrotzek, 2017: 42, transla‐ ted by WKG) To supplement this basic model, further models are presented to depict the writ‐ ing processes in detail. The representation model (Bachmann & Becker-Mrotzek, 2017: 47 ff., see Fig. 26) should be emphasized, which refers to the long-term and working memory associated with the writing task and the corresponding text patterns. In the second circle, the knowledge of the selected prototypical text pattern is activated in its constitutive parts and the prototypical task is modified to suit the specific communicative situation, which results in an adapted version of the prototypical text pattern. The dominant cognitive actions of the second circle can be assigned to the knowledge activating or knowledge adapting area. The knowledge that already exists in long-term memory for the text pattern is retrieved. This is knowledge that is already pre-linked in terms of content, function and structure (Bachmann & Becker-Mrotzek, 2017: 47, translated by WKG). The links are created in different ways. In this context, great importance is attached to the text procedures, which not only functionally link communicative actions with linguistic expression in a prototypical way, but - especially for experienced writers - are often in pars pro toto relationships with each other. Through such relationships, 96 2 Theory <?page no="97"?> a single procedure opens up further gaps in the text and initiates the activation of further text procedures (Bachmann & Becker-Mrotzek, 2017: 47, translated by WKG). In the representation model, it should be emphasized that working and long-term memory work to different degrees depending on the writer’s knowl‐ edge of the task. If the writer is familiar with the writing task, they can fall back on text patterns stored in long-term memory. If, on the other hand, the writing task is unfamiliar and the working memory is heavily loaded, the writer will adapt familiar text patterns to solve the writing task. Fig. 26: Circle II: Representation of the communicative situation and adaptation of the text pattern (Bachmann & Becker-Mrotzek, 2017: 48, translated using AI) In summary, the writing process models show that multiple operations and external and internal factors influence subjective writing to varying degrees. In particular, the interaction of long-term and working memory with the writer’s knowledge of text types/ patterns plays a crucial role in shaping the writing process. It is also clear that writing models are subject to change due to the increase in new findings and new research methods. New measurement methods in writing research and neurobiological studies will also lead to further developments of current models in the future. 2.5 Writing process 97 <?page no="98"?> 2.6 WRITING TEXTS USING THE SRSD METHOD “Strategy-focused interventions are the most effective category of actions for improving writing skills among varying populations of students in a wide range of international con‐ texts” (Robledo amón & García, 2017: 38). In international research, Robledo-Ramón & García (2017: 39 f.) have compiled a series of strategy-focused writing interventions. Behind each model are instructional sequences that make the writer a monitor of their writing process by analyzing the writing task and encouraging them to set writing goals and apply planning and revision. Each model focuses on different techniques and strategies. These focus on support through “scaffolding” (Gibbons, 2009: 15) and exchange through feedback and collaboration. “In the national and international discourse of empirical writing research, strategies are booming due to their importance for self-regulation” ( Jost, 2017: 183, translated by WKG). This is a didactic learning objective in writing lessons (ibid.). The SRSD model is probably the best-known strategy-focused intervention model for writing lessons in the Anglo-American world. It focuses on six phases, which are explained in more detail below (see Fig. 27): Fig. 27: SRSD process (after Glaser, 2005; own illustration, translated by WKG) ad 1.) In the first phase, the existing background knowledge is activated and developed. The focus here is on reading and discussing the type of text and communicating about writing attitudes and writing strategies. Writing attitude is one of the social and cultural factors that influence writing. These are understood as dispositions “[…] to react in a certain way in specific situations” (Winkel/ Petermann & Petermann, 2006: 102, translated by WKG). 98 2 Theory <?page no="99"?> ad 2.) In the second phase of the discussion, the writing attitude is discussed in more detail. Fears and previously used writing bridges can also be addressed. Initial writing strategies are introduced following the discussion. The advan‐ tages of using these strategies are outlined. Sample texts, which may also contain errors, are analyzed and discussed. Content-related text features of the text types are listed and introduced through mnemonics. Text planning, e.g., using mind maps, is also addressed. The step-by-step approach to text types can also be compared to procedural learning. This second phase corresponds to the cognitive stage, as the acquisition of declarative knowledge is predominant. ad 3.) In the third phase, modeling, the teacher models the writing process alone or together with students. The teacher allows the text to develop in front of the pupils using self-regulation strategies, objectives, mind maps, or mnemonics, illustrating the planning and revision phase visually and verbally. Videos can also be used in this phase. In some cases, peers may also act as models. Two learning mechanisms are used here: learning from the model and learning through observation. The entire modeling phase shows the application of writing strategies and is repeated as needed. A complex task is solved using a solution scheme. In terms of learning psychology, this phase corresponds to the “knowledge compilation stage” (Winkel/ Petermann & Petermann, 2006: 224, translated by WKG). ad 4.) In the fourth phase, memorization, content-related text elements on the one hand, and writing phases, objectives, or self-instructions on the other, are memorized using mnemonics. This step is intended to build declarative knowledge. ad 5.) The fifth phase, support, is characterized by a practice phase conducted with classmates, individually, or with the teacher through scaffolding. Pupils can work together to plan, write, or revise a text. Mnemonics should be used in the process. From a learning psychology perspective, classmates as a social community can reinforce behaviors positively or negatively through various forms of feedback (Winkel/ Petermann & Petermann, 2006: 169). ad 6.) In the final phase, practicing, the student writes independently, applying their strategies and mnemonics (Harris & Graham, 2017: 129 ff.). This transfer of learning is defined in learning psychology as generalization: “The term generalization means that previously learned behaviours are transferred to new, more or less similar situations” (Winkel/ Petermann & Petermann, 2006: 134, translated by WKG). This stage is also referred to as the “autonomous stage” (Winkel/ Petermann & Petermann, 2006: 224, translated by WKG), as writing becomes increasingly routine and automated. 2.6 Writing texts using the SRSD Method 99 <?page no="100"?> Harris and Graham emphasize that every student writes differently and is shaped by their socio-cultural background. Therefore, for some students, it may not be necessary to go through all the phases of the writing process if they already have extensive writing experience. However, these students can play a valuable role by supporting their peers during writing (Harris & Graham, 2017: 139). This means that the SRSD phases can also be used differently. According to Robledo-Ramón and García (2017: 41), the techniques used in this approach include: Mnemonics, discussions, cognitive modeling, thinking aloud, self-instruction, collaborations, peer support, scaffolding, criterion-based evaluation, self-regulatory strategies, and motivation development: Achievement motivation is significantly influenced by control beliefs and causal attribution processes. This results in opportunities for intervention to promote achievement motivation. Finally, the social context of learning can be a decisive mo‐ tivational variable. Motivational and emotional variables are closely linked. Positive feelings such as flow promote motivation to learn and achieve, while feelings of anxiety tend to have a detrimental effect (Winkel/ Petermann & Petermann, 2006: 66, translated by WKG). It is therefore not just one technique that defines this approach, but the application of multiple strategies, as a kind of strategy bundle. According to MacArthur (2017: 136), these strategies relate to the knowledge of concept and procedures. Conceptual knowledge refers to the understanding of the nature of the written text, including text pattern knowledge, and content knowledge. Procedural knowledge, by contrast, refers to the writing process, including planning, writing, and revising. Both forms of knowledge are essential for effective revision. As MacArthur explains: “The procedural knowledge involves reading and applying evaluation criteria to identify problems and opportunities and make actual revisions” (MacArthur, 2017: 136). The combination of sequen‐ ces and techniques shows that they offer ample opportunities for discussion about writing. The interaction is not solely guided by the teacher, but is also supported by the peer group. Graham and Harris emphasize the promotion of metacognitive writing strategies building on Hayes’ writing process model (Hayes & Flower, 1980; Hayes, 1996; Hayes, 2012), and integrating the Writer’s Workshop approach developed by Calkins (1986) and Graves (1983) and link the approach with several theories and research findings (Harris & Graham, 2017: 121). Harris and Graham emphasize four theories: 1. “Meichenbaum’s (1977) cognitive-behavioral intervention model” (Harris & Graham, 2017: 121 f.), 2. “the work of Soviet theorists and researchers (including Vygotsky, Luria, and 100 2 Theory <?page no="101"?> Sokolow […]) that focuses the social origins of self-control, the development of the cognition, and the concept of the zone of proximal development” (ibid.), 3. “the work of Brown, Campione, and their colleagues was ‘informed instruction’, which suggests that students should clearly understand what they are doing and why, […] and the importance of metacognition” (ibid.) and 4. “the work of Deshler, Schumaker, and their colleagues on the validation of acquisition steps for strategies among adolescents with learning disabilities” (ibid.). Since the late 1980s, the approach has been extensively studied, with over 100 research studies conducted to date; it was initially termed “self-control strategy training” (Harris & Graham, 2017: 120) and renamed as “self-regulated strategy development” (SRSD) in 1992 (ibid.) (Graham & Harris, 1987). SRSD encompasses a broad framework of theoretical perspectives incorporating principles from construc‐ tivism, behaviorism, motivation, social cognition, summarized visually in the following graphic (see Fig. 28). Fig. 28: Theory of SRSD (after Harris & Graham, 2017: 123, own illustration) The first studies on SRSD were primarily conducted with students who ex‐ perienced writing difficulties due to learning disabilities (Sawyer, Graham & Harris, 1992; Troia, Graham & Harris, 1999; among others), came from socioeconomically disadvantaged backgrounds (Graham, Harris & Zito, 2005), or were in the early stages of writing socialization in elementary school (Graham, Harris & Troia, 1998; Saddler et al., 2004; Graham, Harris & Zito, 2005; among others). Within the context of strategy-focused instruction such as SRSD, students are encouraged to plan, execute, and revise their writing process through the explicit instruction of the strategies, techniques, and scaffolding (Robledo-Ramón & García, 2017: 38). 2.6 Writing texts using the SRSD Method 101 <?page no="102"?> In the following, a series of selected studies will be presented that examine the implementation of the method “SRSD” method (Graham & Harris, 1996). It will also explore the outcomes of this approach for the students. The general aims of SRSD are: 1. To equip students with the knowledge of writing process (planning to revising) and the corresponding strategies; 2. to enable students to monitor and regulate their writing process consciously; and 3. to foster a more positive attitude towards writing (Graham, Harris & Zito, 2005: 238 f.). The objective is to help students change their associative, unstructured writing styles to more reflective, structured writing practices, and to understand text types better. Consequently, SRSD can be categorised as a self-regulatory writing program that contributes to what von Saldern explains as the “competence of self-education” (von Saldern, 1999: 119, translated by WKG) through independ‐ ent learning: As a rule, such methods essentially fulfill the prerequisites that are essential for self-directed learning: Differentiation and rhythmization according to the individual learning starting point and performance of the pupils (von Saldern, 1999: 120, translated by WKG). SRSD has improved the quality of students’ writing, their knowledge of the text type, and their self-efficacy. In addition, SRSD enhances the structure of the texts and leads to longer texts compared to those written before the intervention (ibid.). SRSD provides even the weakest writers a structured framework to plan and gradually revise texts (Graham, Harris & Troia, 1998: 22). Thus, students gain more knowledge about text types and writing as a cognitive process. SRSD also promotes “self-assessment” (Graham, Harris & Troia, 1998: 23), given that writing, in general, requires a high degree of self-management (Graham, Harris & Troia, 1998: 21). The SRSD approach includes several mnemonics as strategies mentioned in the following studies: The Mnemonic POW organizes the writing process with the prompts: ✓ “Pick my ideas” (Graham/ Harris & Mason, 2005: 217), ✓ “Organize my notes” (ibid.) and ✓ “Who are the main characters? ✓ “Write and say more” (ibid.). The WWW model, on the other hand, supports the writing of a narrative with questions: ✓ “Who are the main characters? ✓ When does the story take place? 102 2 Theory <?page no="103"?> ✓ Where does the story take place? ✓ What do the main characters want to do? ✓ What happens when the main characters try to do it? ✓ How does the story end? ✓ How do the main characters feel? ” (ibid.). The 7-W question strategy, originally developed by Stein and Glenn (1979), was later adapted by Graham and Harris (Glaser, 2005: 88). The Mnemonic TREE helps with argumentative texts (Graham, Harris & Mason, 2005: 217): “Tell what you believe (State your topic sentence), Give 3 or more Reasons (Why do I believe this? ), Examine each reason (Will my reader buy it? ), End it (Wrap it up right! )” (ibid.). The mnemonics presented illustrate writing strategies in short writing form and are designed to be remembered by the pupils. • In the field study conducted by Sawyer, Graham, and Harris (1992), 33 fourthand fifthgraders with learning disabilities were divided into four groups. All participants came from a middle-class neighborhood and lived outside the U.S. capital, Washington, D.C. On average, they were eleven and eleven months old, with a mean IQ of 93.8. Group A received instruction under the SRSD intervention condition; Group B received SRSD without goal setting and self-monitoring; Group C received direct instruction; and Group D served as the control group (Sawyer, Graham & Harris, 1992: 340, 342). The steps and the mnemonics were explained to the test subjects in detail, and the teacher modeled the strategies aloud. The methodological focus for the students was self-instruction, while the trainers emphasized scaffolding (Sawyer, Graham & Harris, 1992: 341). Three research questions were addressed: 1. Are there differences among the four learning conditions? 2. Are the writing products of children with learning disabilities who follow a writing strategy comparable to those of their typically developing peer group? 3. Does the use of writing strategies influence the self-efficacy of the participants? (Sawyer, Graham & Harris, 1992: 342). The students’ handwritten narratives were transcribed and orthographically corrected. Two independent raters, with a rater agreement of 0.88, assessed the texts based on the text elements identified by Stein and Glenn (1979), using a three-point scale (0 = element missing, 1 = element present and 2 = element present in great detail), and the overall quality of 2.6 Writing texts using the SRSD Method 103 <?page no="104"?> the texts on a seven-point scale. The following ten questions were asked to test self-efficacy expectations for writing the narratives: “Can you write a story that-… 1.-…-tells about the main character’s feelings? 2.-…-clearly tells about the setting? 3.-…-has a good beginning? 4.-…-tells who the main character is? 5.-…-tells about several things that happen to the main character? 6.-…-tells when the story takes place? 7.-…-tells where the story takes place? 8.-…-tells what the main character wants to do? 9.-…-has a good ending? 10.-…-Can you write a good, creative (made-up) story? ” (Sawyer/ Graham & Harris, 1992: 343). These questions could be answered on a scale ranging from 10 to 100 points (10 points = not sure, 40 points = maybe, 70 points = reasonably sure, 100 points = absolutely sure). The post-test was conducted one week after the last intervention by the class teacher(s). The writing of the narratives was stimulated by five black-and-white pictures. This resulted in five writing products from the pupils. The testers worked with two to three students at a time in a quiet room, three times a week. They were not allowed to provide assistance or feedback regarding self-efficacy (Sawyer, Graham & Mason, 1992: 344). While Group A received direct instructions on how to use the strategies, they were not allowed to practice them or receive feedback on their pretexts. However, the WWW model was made available to the students. Groups A, B, and C were also introduced to the following model to use and were instructed to memorize the associated writing strategies: 1.-“Look at the picture” (Sawyer/ Graham& Harris, 1992: 345). 2.-“Let your mind be free” (ibid.). 3.-“Write down the story part reminder” (ibid.). 4.-“Write down story-part ideas for each part” (ibid.). 5.-“Write your story, use good parts and make sense” (ibid.). Group B was required to set their targets and check them. They received feedback from the testers, including comments on the number of content-re‐ lated text elements in the texts (story grammar). In addition, the participants in this group were encouraged to verbalise their thoughts on writing aloud 104 2 Theory <?page no="105"?> to themselves (Meichenbaum, 1977). They were also given the following guide, including questions for self-instruction to assist the writing process: ✓ “What is it I have to do? ” (Sawyer/ Graham & Harris, 1992: 345) for the problem definition ✓ “Now I’d better write down my story-parts reminder” (ibid.) for the planning step ✓ “Am I using all my parts so far? ” (ibid.) for the self-evaluation ✓ “Good, I like this part! ” (ibid.) for self-enhancement. The SRSD groups were given another practice session to apply the guide. All three groups were allowed to use their control plan for writing narratives. The control group, on the other hand, received neither assistance nor feedback for writing their three stories. The “SRSD full” group wrote the best narratives regarding content and quality (Sawyer, Graham & Harris, 1992: 346). At the beginning, self-efficacy expectations were similar for all students. Over the course of the interven‐ tion, groups B and C improved the most, with group C having a minimal advantage. The control group, on the other hand, remained the group with the lowest self-efficacy expectations compared to the other three groups (Sawyer, Graham & Harris, 1992: 348). In the post-test, the results of the first three groups were not significantly different. Thus, the three researchers showed that goal setting and self-monitoring during the writing process improved the students’ writing results (Sawyer, Graham & Harris, 1992: 350). • Graham, Harris, and Troia (1998) examined two case studies. In the first, six fifthand sixth-grade students were selected from an elementary school in the Washington, D.C. area, where most families were socially disadvan‐ taged. These six had learning disabilities, low writing motivation, reading difficulties, IQs between 81 and 117, and had previously been instructed using the Writer’s Workshop (Graham, Harris & Troia, 1998: 26). None of these six students planned the text beforehand. The teacher used the six phases explained above to teach the SRSD instruction (see also Fig. 27): 2.6 Writing texts using the SRSD Method 105 <?page no="106"?> 1. “Develop background knowledge” (Graham, Harris & Troia, 1998: 27). 2. “Discuss it” (ibid.). 3. “Model it” (ibid.) 4. “Memorize it” (ibid.) 5. “Support it” (ibid.) 6. “Independent Performance” (ibid.). In the first phase, the teacher asked the students about their prior knowledge of the text type. In the second phase, the students discussed which elements should be included in an essay. This discussion also took place individually by giving them a strategy to learn the elements of an essay. The students then developed a fact sheet based on the TREE model. In the third phase, the teacher modeled how to apply the TREE model by thinking aloud. In this phase, motivational phrases such as “Work hard - write better” (Graham, Harris & Troia 1998: 27 f.) or “Slow down and take my time” (ibid.) were also enunciated by the teacher and observed by the students. The students were free to choose whether or not to use these phrases. In the fourth phase, the students were asked to memorize the TREE model. In the fifth phase, the students wrote an essay using the information sheet and the strategies. The students were asked to work through all the steps, and the teacher supported them if help was needed. Three to four essays were written, after which no more support was provided. Finally, a group conference with all students reflected on how they had dealt with the strategies. The result was better when comparing the pre-test and post-test about the quality and length of the essays. In the second case study with three fifth-grade students, all with a learning disability and an IQ between 98 and 104, the trainer gave all the steps using the “STOP & LIST” strategy. As homework, the students were asked to think about how they could integrate this strategy into their everyday lives and make changes if necessary. All six phases, from Develop background knowledge to Independent Performance, were instructed again (Graham, Harris & Troia, 1998: 31 f.). Each test person found an everyday situation where they could use the STOP & LIST model. After practicing two texts with the teacher, the students were asked to write a text independently as a homework assignment. This self-management increased self-efficacy expectations, making it effective even for weak writers (Graham, Harris & Troia, 1998: 37). 106 2 Theory <?page no="107"?> • The research team Troia, Graham, and Harris (1999) studied only three students in a fifth-grade class with learning disabilities to see if they could set goals, brainstorm ideas, use them for writing, and complete writing assignments independently at home. According to Flower and Hayes (1980), students with learning disabilities minimize the writing schedule, but it is essential to the writing process and the quality of the text. All three students had an IQ between 85 and 115, as analyzed using the Wechsler Intelligence Scale for Children-III test, came from socially disadvantaged families, were male, and were native English speakers. They were between ten and eleven years old. The procedure involved the following six steps: 1. “Pre-instruction” (Troia, Graham & Harris, 1999: 238 f.) - Text elements for a good story and a persuasive essay were identified and two strategy models were presented: “SPACE” - “Setting, Problem, Actions, Conse‐ quence, Emotions” (ibid.) and “DARE” - “Develop a topic sentence, Add supporting details, Reject arguments, End with a conclusion” (ibid.). Each element of these models was discussed with the participants beforehand. 2. “Baseline probes” (ibid.) - The students were asked to write a text for a selected picture. The trainers helped with spelling and typed the students’ handwritten texts at the end. No further help was provided. 3. “Strategy instruction” (ibid.) - The students explored three strategy steps: “goal setting, brainstorming, organizing” (ibid.). They were first practiced orally. In this phase, the students were encouraged to express why they find these strategies helpful. The “STOP & LIST” model - Stop, Think of Purpose, List Ideas, Sequence Them (ibid.) - was introduced and explained. One of the homework assignments was to plan a travel trip. In the final two of the seven lessons, the students single-handedly planned a story and completed it alone at home. 4. “Post-instrucional probes” (Troia, Graham & Harris, 1999: 238 f.) - Three writing samples were prepared by each participant. 5. “Maintenance probes” (ibid.) - The last story has been written. 6. “Task generalization probes” (ibid.) - Students were asked to write a persuasive essay on ‘Should students wear school clothes in schools? ’ The writing time, the length of the texts, and the text elements to be considered were evaluated by raters. Furthermore, the text quality was assessed on a scale from 8 (highest quality) to 1 (lowest rating), as well as vocabulary, grammar, syntax, structure, imagination, and clarity (Troia, Graham & Harris, 1999: 242). 2.6 Writing texts using the SRSD Method 107 <?page no="108"?> The results show that the test subjects did not take the time to work out a writing plan before the intervention. For all three students, the number of text elements, text length, and writing time had increased due to the intervention. In addition, all three students took time to draw up a writing plan. This showed that SRSD can also be effective for students with learning disabilities and that these students are encouraged to self-manage (Troia, Graham & Harris, 1999: 243 ff.). • The researchers Saddler, Moran, Graham, and Harris (2004) initially exam‐ ined two second classes at an elementary school in a socially disadvantaged environment and with students having weak writing skills. They had previously followed the “Writing Workshop,” according to Atwell (1987) (Saddler et al., 2004: 5). Based on the TOWL-3 test (Hammill & Larsen, 1996), eight pupils with considerable writing difficulties were identified. Six of these pupils participated in the study in pairs, which was conducted three times a week for 25 minutes each session, outside the classroom at school with a teacher across twelve sessions (Saddler et al., 2004: 8). The focus of this study was on planning and writing a text using the POW model, the WWW model, and the following writing routines: ✓ “Pick a topic or an idea to write about” (ibid.). ✓ “Brainstorm and organize possible ideas for each part of their story in advance of writing” (ibid.). ✓ “Continue to develop these ideas as they wrote” (ibid.). The teacher took the six SRSD phases into account. First, the students wrote three stories as a pretest. During the intervention, the elements of a story were explained to them, and then three to five stories were written. After three weeks, the students wrote one story as a post-test. One tandem also wrote a repeated post-test as a follow-up after six weeks (Saddler et al., 2004: 6). After all the texts had been typed up, the following points were assessed: 1. the number of words, 2. the number of stories written, 3. the number of elements present in the texts and 4. the general quality of the texts (1 point for low quality and 7 points for highest quality) (Saddler et al., 2004: 7). Before the intervention, the students wrote stories with a score of 2.3, which was in the lower rating range. After the intervention, the students wrote more complex stories, and three even used all seven text elements of a story (Saddler et al., 2004: 11). All students used 108 2 Theory <?page no="109"?> the writing strategies for planning and writing the stories. Thus, it was demonstrated that even students with writing difficulties can successfully work with SRSD in a second-grade class (Saddler et al., 2004: 15). • Researchers Graham, Harris, and Mason (2005) aimed to use SRSD to encourage students to plan, draft, amend, and publish texts in a five-month intervention study. The hypothesis was that students wrote better quality texts, longer texts, and stayed longer in the writing process (ibid.). In Washington, 317 third-graders at four elementary schools were first given the “Test of Written Language - 3” (Hammill & Larsen, 1996) as a pretest. From this, 86 “at-risk” students were identified, 73 of whom, on average, eight years and four months old, took part in the intervention three times a week (20 minutes each). They were divided into three groups. Group A (n = 24) received SRSD instruction, group B (n =24) received SRSD instruction and peer support, and group C (n = 25) was the control group. The male proportion and the proportion with English as their mother tongue predominated in this sample (Graham, Harris & Mason, 2005: 213 f.). Students conducted the study outside the classroom. Group A was encouraged to self-reflect and self-monitor the writing process (Graham, Harris & Mason, 2005: 216). The feedback and instructions from the trainers were individualized so that the students could write narrative and persuasive texts. The models used were “POW” (Graham, Harris & Mason, 2005: 217), “WWW” (ibid.), and “TREE” (ibid.). The trainers first introduced these mnemonics and wrote their texts with their help. The writing products were then compiled in the group. In the support phases, the trainers worked with the following phrases via scaffolding: ✓ “What do I have to do here? ” (Graham, Harris & Mason, 2005: 218) for the problem definition. ✓ “Does it make sense? ” (ibid.) for text planning. ✓ “Continue to develop these ideas as they wrote” (ibid.). for self-reflection. ✓ “I really like that part! ” (ibid.) as a sign of recognition. This method is based on the “Basic Three-Step” strategy by Graham and Harris (1996) and can, in turn, be traced back to the three sub-activities of the writing process by Hayes and Flower (1980) (Glaser, 2005: 87). 2.6 Writing texts using the SRSD Method 109 <?page no="110"?> 8 The correlation of the items is 0.40 (Graham, Harris & Mason, 2005: 225). Group B received additional support from the peer group unlike Group A. The process involved the peer partners first setting a goal for themselves while using the POW model. They then decided how they wanted to help their partner and discussed how the strategies could help. Finally, another model, the TREE model, was introduced for writing an essay (Graham, Harris & Mason, 2005: 218 ff.). All students wrote four texts (one narrative, one persuasive, one personal, and one informative text), which were typed before the assessment. For the narrative and personal texts, one point was awarded for each element taken into account (Graham, Harris & Mason, 2005: 223). In addition, the students were presented with these questions before the pretest and after the post-test: 1. “Suppose you were asked to be the teacher of your class today, and one of the other kids asked you, What is good writing? What would you tell that student about good writing? ” (Graham/ Harris & Mason, 2005: 224). 2. “Why do you think some kids have trouble writing? ” (ibid.). 3. “What do good writers do when they write? ” (ibid.). 4. “When you are asked to write a paper for class or homework, what kind of things can you do to help you plan and write your paper? ” (ibid.). 5. “Tell a friend what kind of things are included in a story! ” (ibid.). 6. “Tell a friend what kind of things are included in a persuasion! ” (ibid.). Furthermore, self-efficacy was tested on a scale of 1 (strongly disagree) to 4 (strongly agree) with the following five items 8 (Graham/ Harris & Mason, 2005: 225): • “When my class is asked to write, my paper is one of the best” (Graham/ Har‐ ris & Mason, 2005: 225). • “When writing a paper, it is hard for me to decide what goes first, second, third, and so on” (ibid.). • “When writing a paper, I have trouble finding the right words for what I want to say” (ibid.). • “When I plan a paper, my plan is one of the best in the class” (ibid.). • “When writing a paper, it is easy for me to keep thinking of things to say” (ibid.). The results of this study show that intervention groups A and B took more time to write and wrote longer and better quality texts than the control group when the pretest and posttest were compared (Graham, Harris & 110 2 Theory <?page no="111"?> Mason, 2005: 227). Group A took longer to write the narrative and persuasive text than Group B. This group also wrote longer texts. The basic elements of a narrative and an argument were considered by both Group A and Group B; there were no statistical differences here. Group A wrote qualitatively better narrative and persuasive texts, while Group B wrote better informative texts. With regard to the questions on writing knowledge, there was no statistical difference between Groups A and B, although both groups were better at describing good and bad writing than group. C. The examination of self-efficacy revealed no statistical difference (Gra‐ ham, Harris & Mason, 2005: 228 ff.). The three researchers in this study concluded that third-graders can improve both their knowledge of the writing process and their writing skills by applying strategies, as the intervention groups improved their writing in terms of length, quality, and completeness of texts (Graham, Harris & Mason, 2005: 238). • The study by Graham, Harris, and Zito (2005) focused on primary school students with writing difficulties (struggling writers) in grades two and three. The hypothesis was that struggling writers’ texts would be better if they referred to their learned strategies (Graham, Harris & Zito, 2005: 238). This study aimed to promote planning more effectively. In addition, peer support was to be integrated into this intervention study (Graham, Harris & Zito, 2005: 244 f). Four experiments were conducted for this purpose: The first sample consisted of 73 third-graders from four schools with a class size of 24 students from the vicinity of Washington, D.C., with more socially disadvantaged families. The distribution of boys and girls was 3: 2 (Graham, Harris & Zito, 2005: 252 f.). The teachers were asked beforehand whether the students were good or bad writers (Graham/ Harris & Zito, 2005: 246 ff.). The test subjects were also asked about their prior writing knowledge and self-efficacy expectations. The question was whether the SRSD model was effective for weak writers (Graham, Harris & Zito, 2005: 252). The subjects were divided into three groups: Group A received SRSD planning instruction, Group B received SRSD and peer support, while Group C was the control group and received the “Writer’s Workshop” (Graham/ Harris & Zito, 2005: 252). Six trainers conducted sessions three times a week for 20 minutes each for five months, working with only one student tandem at a time (Graham, Harris & Zito, 2005: 252 f.). The POW model was used to plan the texts. The WWW model was used for writing the story and the TREE model for the argumentative text (Graham, Harris & Zito, 2005: 249 f.). Their partners supported the test subjects with 2.6 Writing texts using the SRSD Method 111 <?page no="112"?> the sentences “I transferred my strategies/ I helped my partner” (Graham, Harris & Zito, 2005: 251). After instructions on a text type, this text type was written again (post-test). The text types trained were the story, a persuasive text, a personal narrative, and an informative text. The result of the first experiment was that group A wrote qualitatively better and longer texts than the control group, whereas group B could better remember the planning strategies through peer work and wrote better texts than the control group. More trained text elements and longer planning phases could be detected in the texts of Group B than in the other groups (Graham, Harris & Zito, 2005: 255). The second sample was made up of six second-grade pupils from one school. Three tandems were formed to write a story together thrice a week for 20 minutes over five hours. The tandems received SRSD instructions for this. The second experiment showed that students with SRSD wrote better stories qualitatively and quantitatively, and that even second-grade students could handle SRSD (Graham, Harris & Zito, 2005: 256). The third sample consisted of 66 second-grade students over six months. They were divided into three groups: Group A received SRSD only, and Group B received SRSD including peer support, while Group C was the control group that worked with the “Writer’s Workshop” (Graham, Harris & Zito, 2005: 256). Six master’s students each instructed a pair of students in the text types history, argumentative writing, personal narrative, and informative writing. In addition to instruction in pairs and writing these texts, the students were also asked to write a text in class according to SRSD instructions. However, this could not be measured as the students did not work on their texts alone, but stood up and helped other students. The TREE model was again used for the argumentative text, but the first E was omitted as a strategy. For this type of text, a tandem of students was asked to write in front of the class with SRSD instruction so that the strategy became visible (Graham, Harris & Zito, 2005: 257 f.). The results showed no significant difference between Groups A and B. Group A wrote better and longer texts than the control group (Graham, Harris & Zito, 2005: 258 f.). The fourth sample was a composition of eleven teachers who implemented SRSD strategies plus peer support in a classroom setting with 53 weak writers, three times a week for 20 minutes each, for five months. 27 students received the SRSD instruction, while 24 were in the control group and worked with the Writer’s Workshop. Two students dropped out of the program. The comparison between the pre-test and post-test for the trained text types, story and personal narrative, showed that Group A with SRSD 112 2 Theory <?page no="113"?> instruction wrote longer texts and included more content elements in their texts than the control group. The teachers also noticed that the students were more intrinsically motivated (Graham, Harris & Zito, 2005: 260). An initial interim conclusion can be drawn at this point: The integration of the SRSD approach into writing lessons does justice to the required action and student orientation resulting from the communicative turn in German teaching (Eichler & Henze, 1998: 109), because the “competence of self-education” (von Saldern, 1999: 119, translated by WKG) can be built up in students in self-regulatory writing programs (Philipp, 2012: 130, 147). For Germany, the researchers Glaser (2005) and Keßler (2010) can be cited as examples of this, who also deal with the SRSD approach, albeit from a psychological perspective: • In her doctoral thesis, Glaser (2005) promoted the writing skills of primary school pupils (n = 154) in year four for 90 minutes once a week over four weeks by teaching them SRSD writing strategies in the form of self-regulatory essay training (SAT) (Glaser, 2005: 220). This was intended to teach genre-specific writing strategies (Glaser, 2005: 85). Her dissertation project was divided into a preliminary investigation, a main investigation, and individual case descriptions. She used the writing of a picture story and a story with a picture template as a test instrument. Planning, writing, and revising were also fundamentally trained (Glaser, 2005: 218). The mnemonics had to be translated from American and changed. For this purpose, Glaser worked with the “AHA” (Glaser, 2005: 218, translated by WKG) along with the “7 W questions” (ibid.) for writing a story. Vocabulary and syntax were to be improved using a “tension word list” (Glaser, 2005: 119). “In addition, methods of goal setting, self-monitoring, self-assessment, and self-correction were explicitly taught to build up monitoring and control processes” (Glaser, 2005: 219). The tudents were provided with checklists for revising the stories (ibid.). The teachers of the intervention were trained with the help of a manual and worked with the scaffolding method (Glaser, 2005: 219). Furthermore, the phases of the SRSD approach were used, but translated into German and used as stages of instruction. In the preliminary study, the pupils revised their texts in peer group discussions using the criteria lists (Glaser, 2005: 94). The result shows that pupils using the SAT approach wrote longer and higher-quality texts, especially pupils with poor writing skills. The compar‐ ison of the preand post-tests resulted in a corrected effect size of d = 1.82. From the pretest to the follow-up test, this value amounted to d = 2.09. 2.6 Writing texts using the SRSD Method 113 <?page no="114"?> The writing performance and the self-regulatory skills were strengthened between the pre-test and post-test; here, the value was d = 1.44, in the pretest and follow-up test d = 1.89 (Glaser, 2005: 220). Neither self-efficacy expectations nor the attitude towards writing again showed no effect, as determined with the help of the questionnaire survey (Glaser, 2005: 230): The training effect on writing performance proved to be a direct effect. Mediator effects could not be demonstrated either for the process-related measures (planning quality, revisions) or for the writing attitude (self-efficacy expectation) when writing and the text volume. In a path analytical examination of the effects of all investigated influencing variables on writing performance, the direct training effect remained statistically highly significant in addition to the indirect effects for planning quality and text volume (Glaser, 2005: 221, translated by WKG). Glaser concluded by demanding that writing interventions should also relate to other types of text. Furthermore, revision should be examined more closely as part of the writing process. Cooperative learning arrangements with the help of peer groups or student tandems should be further promoted (Glaser, 2005: 232, 235 f.). Keßler (2010) took up the demands of Glaser (2005) and developed an intervention in her doctoral thesis using a classic pre-post follow-up design to promote the revision skills of sixth-graders in ten Giessen secondary school classes with five to eight pupils (n = 273) (Keßler, 2010: 224). The groups of pupils were trained once a week for 90 minutes over five weeks by psychology students and student teachers (Keßler, 2010: 107). Over 90 % of the test subjects spoke German as their mother tongue. The average age was between 11.3 and 11.6 years (Keßler, 2010: 105). The test subjects were divided into two groups. Group A underwent self-regulatory revision training, and Group B underwent revision training only. The self-regulatory essay training by Glaser (2005) and Keßler (2006) was integrated into the new writing project. The hypothesis was that revision skills could be improved through the self-regulatory approach (Keßler, 2010: 223). As Keßler criticized the SRSD studies for focusing more on the completeness of the text product when evaluating text quality, she wanted to introduce further research variables and impulses. Furthermore, Keßler supplemented her study with a questionnaire on self-efficacy expectations (Keßler, 2010: 224). Her areas of investigation were: “Strategy application: content and linguistic revisions, revisions and logical senten‐ ces to improve the coherence of the transcript” (Keßler, 2010: 94, translated by WKG), 114 2 Theory <?page no="115"?> 9 The first value indicates the post-test result and the second reflects the follow-up test result. “Strategy-related writing performance: existing story elements and their linguistic arrangement in the transcript” (ibid.), “Holistic writing performance: narrative quality of the transcript” (ibid.), “Strategy-related transfer: existing story elements and their linguistic form in the draft” (ibid.), “Strategy-related knowledge (questionnaire and open knowledge questions)” (ibid.), “Writing attitude (self-efficacy expectation) for writing and revising texts” (ibid.) and “Revisions carried out on a faulty text template” (ibid.). As in Glaser, the “three-step technique” (Glaser, 2005: 218, translated by WKG) was also used in this study. It is based on the “basic three-step” strategy according to Graham and Harris (1996), which aims to promote the sub-activities of planning, writing, and revising according to Hayes and Flower (1980). The intervention course used the W questions after writing a picture story to assess the students’ prior knowledge of the text type (Keßler, 2010: 115). After the pupils had written their stories about the pictures, they checked their texts using a “login memory card” (Keßler, 2010: 116), on which the text elements of a story were listed. This step led to the first revision. A further revision was stimulated by the focus on language with the “2xA+2xG strategy” (Keßler, 2010: 117, translated by WKG). This resulted in a second revision using formulation aids and text modules. The third revision was aimed at revising the beginnings of sentences and lexis. In addition, the students always received positive feedback on their performance to reinforce the self-regulatory revision training (Keßler, 2010: 118 f.). Group B with pure revision training did not have access to the login memory card and the 2xA+2xG strategy. As a result, both groups showed positive effects (Keßler, 2010: 198 ff.): - Strategy-related writing performance (d = .80/ .38) 9 - holistic writing performance (d = .84/ .85) - strategy-related knowledge (d = .80/ .38) - writing-related self-efficacy expectation (d = .39/ .21). Nevertheless, the results of Group A, which received self-regulatory revi‐ sion training, were higher than those of Group B. Thus, Keßler proved that self-regulatory revision training was more effective through teaching strategies (Keßler, 2010: 224). In the follow-up test, however, there was 2.6 Writing texts using the SRSD Method 115 <?page no="116"?> no significant difference in the self-efficacy expectations of both groups (Keßler, 2010: 225). Keßler concluded that strategy-related knowledge and writing-related self-efficacy expectations were positively influenced by the strategy application (Keßler, 2010: 226). However, she also critically noted that the time-based intervention was unsuitable for weaker pupils and that a larger time window with more practice opportunities would be more effective. She also advocated implementation in the whole class, supported by peer group discussions with classmates (Keßler, 2010: 216 f.). In the meantime, other researchers have also begun examining the SRSD approach in the form of meta-analyses: Philipp (2015a) summarized the studies by Graham and Harris and examined them for effect sizes. He distinguishes “hierarchy-high abilities” (Philipp, 2015a: 158, translated by WKG), e.g., planning or revising, as “[…] mentally costly processes […]” (ibid.), from “low-hierarchy skills” (ibid.), e.g., writing, whereby low-hierarchy is not a pejorative term, but instead refers to automated processes (ibid.). In a comparison of the studies, Philipp concludes that teaching writing strategies is highly significant in the SRSD studies with an effect size of 1.00, which would specifically promote the writing processes. Summarizing in writing (0.82) and training creativity and imagination (0.70) can be listed as further hierarchy-high skills with a high effect size (ibid.). Positive effects were also shown by “activities before writing” (ibid.) (0.54), the “addition of self-regulation in the teaching of writing strategies” (ibid.) (0.50), the teaching of “text structure knowledge” (ibid.) (0.41) and “writing as research” (ibid.) (0.32). In the area of low-hierarchy skills, increasing vocabulary is the most effective (0.78). Combining sentences (0.56) or “practising writing” (ibid.) (0.55) are in the middle range as promoted skills. Training spelling is not significant at 0.19 when writing processes are promoted explicitly in the SRSD studies. In the SRSD studies, Philipp also investigated the effect sizes of the support approaches intended to relieve the writing process (Philipps, 2015a: 162). The setting of product goals with an effect size of 0.80 can e cited as having a powerful effect. Cooperative writing (0.66) and formative feedback (0.61) are also supported approaches with a high effect size. Writing on the computer (0.47) and studying sample texts (0.40) would also show positive effects in the support approaches examined (ibid.). In contrast, the “use of criteria grids for self-assessment” (ibid.) or “procedural support” (ibid.) were not significant. Graham and Harris (2017: 23 f.) conducted a meta-analysis on the effective‐ ness of the use of writing strategies, which indicated that writing lessons with writing strategies have a statistically significant positive influence on text quality, as presented in more detail below: SRSD is a remedial approach with many methods for using writing strategies specifically to improve text 116 2 Theory <?page no="117"?> quality. As already explained, this is done using mnemonics to make it easier for the writer to recall the writing objective, the memory of content-related text features, or writing phases such as planning or revising, as a mnemonic device. Self-regulation is encouraged by monitoring one’s writing process. It is a support approach empirically tested in small groups and entire classes from second grade to upper school, and has proven effective (Graham & Harris, 2017: 22 ff.). Text routines in particular are automated. A meta-analysis of SRSD studies underlines the positive effect on creativity (ES = 0.76), intrinsic motivation, self-efficacy (ES = 1.07), and on content-related text elements/ text features (ES = 0.30). Using sample texts showed a significant positive effect on text quality (ES = 0.30). Syntax (ES = 0.50), grammar, and vocabulary (ES = 0.78) also improved. Writing on a PC also had a statistically significant influence (ES = 0.44), and the addition of grammar and spelling checks had an even greater positive influence (ES = 1.46). Writing support with the help of the peer group (ES = 0.74), setting writing goals such as revising and considering several reasons in an argumentative text (ES = 0.80), teaching text procedures such as planning and revising (ES = 0.52), and activating prior knowledge through mnemonics and brainstorming (ES = 0.48) also had a positive effect on the text quality of the student texts. Concerning feedback, it can be stated that peer feedback (ES = 0.77), adult feedback (ES = 0.87), self-generated feedback (ES = 0.51), and machine feedback (ES = 0.34) have a positive influence on text quality. Overall, the researchers state: “Process writing had a positive and statistically significant impact on improving writing quality (ES = 0.34) […], with 82 % of studies producing a positive effect” (Graham & Harris, 2017: 27). SRSD stands for six characteristics, which are summarized in the following graphic (Harris & Graham, 2017: 126 f.): 2.6 Writing texts using the SRSD Method 117 <?page no="118"?> Fig. 29: SRSD instructions (after Harris & Graham, 2017: 126 f., own illustration) Meta-analyses of the SRSD studies have shown that this support approach leads to improvements in self-regulation, intrinsic motivation, and text quality, specifically through enhanced vocabulary, a greater number of text elements, improved syntax and grammar, increased creativity, and, in some cases, better spelling. Students are particularly supported by writing on a PC, the use of strategies, the use of sample texts, and feedback. In summary, the effectiveness of strategy-focused instruction such as SRSD in improving the quality of pupils’ writing is very high, as numerous studies in the Anglo-American region have shown. Nevertheless, there is a lack of scientific testing in schools in German-speaking countries. 118 2 Theory <?page no="119"?> 2.7 PRELIMINARY INVESTIGATIONS “Before well-planned experiments or surveys are started, it is usually wise to carry out preliminary investigations in order to uncover any sources of error in the instruments and in the planning that have not been thought of ” (Rost, 2007: 169, translated by WKG). The first preliminary study about job-orientated text genres (Giera, 2010), used a systematic document analysis of curricula, training regulations, and written final examinations for 27 training occupations certified by the Chamber of Industry and Commerce. It intended to investigate the types of texts trainees are required to write or read and whether general education schools adequately prepare students for these vocationally relevant text types before they start their training (Giera, 2010: 2). The evaluation confirmed the following hypothesis: “Vocational text types that are taught in the course of vocational training are taught too rarely in general education schools” (Giera, 2010: 43, translated by WKG). The results also highlight the relevance of “transfer text types” (ibid.): The vocational text types legal texts, training contract, offer, collective agreement and regulations are relevant in vocational training. Offers, invoices and order confirmations are the text types in the first three ranks that are more likely to be produced by the trainees, and legal texts, training contract, collective agreement, regulations and the offer are more likely to be received. The general education schools do not work with vocational text types, but with transfer text types that can be converted into vocational text types. The three most important text types are the application letter, the CV and the report, which are taught in all the types of school examined (Giera, 2010: 43, translated by WKG). Another preliminary work evaluated the writing tutorial for first-year students of German at Leuphana University (Giera, Mahler & Neumann, 2013). Student teachers of German were guided through a writing tutorial on the academic writing of a term paper. 124 German students from the 2011 cohort were compared with the 2010 cohort using a quantitative comparison of the grades for the assignments and individual aspects, supplemented qualitatively by feedback forms from the participants and verbal perceptions of the lecturers. The writing intervention, which aimed to consciously break down the writing process into planning, writing and revision phases through, among other things, feedback discussions with the test subjects and the use of criteria checklists, achieved 2.7 Preliminary investigations 119 <?page no="120"?> significant improvements and otherwise positive effects in two promoted sub-areas, so that writing tutorials are still offered and financed at Lüneburg Leuphana University today, which go beyond the subject of German (Giera, Mahler & Neumann, 2013: 38 ff.). Thus, the effectiveness of writing tutoring was demonstrated by utilizing lists of criteria on a formal, linguistic, and content level. In a third preliminary study, Giera (2013) analyzed and evaluated 116 texts produced by 13 students who participated in a writing project lasting several months. This project implemented the Self-Regulated Strategy Development approach developed by Graham and Harris. It included cooperative learning arrangements, such as peer-based revision discussions and the explicit instruc‐ tion of writing strategies (Graham & Harris, 1987, 1996; Sawyer et al., 1992; Giera, 2013). Furthermore, the SRSD approach was aimed at building students’ self-efficacy in writing (Graham, Harris & Mason, 2005: 231; Sawyer et al., 1992: 348; Keßler, 2010: 198 ff.). The following research questions were examined: 1. Which test subjects write higher-quality texts during the project? 2. Which types of text do the test subjects master best? 3. What is the length of the student texts before and after the intervention? 4. What is the relationship between writing attitudes (self-efficacy) and writing performance? The results show that cooperative learning arrangements with the SRSD approach in the form of peer feedback from classmates and the explicit teaching of writing strategies positively affect text quality, text length, and personal attitudes towards writing, provided it is a newly learned text type. The support approach could not improve already learned text types, like writing a fictional story. The participating pupils rated the writing project as positive from their perspective (Giera, 2013: 56 ff.). However, there was no control group in this study. The fourth preliminary study piloted this project from October 2012 to January 2013. Two vocational school classes tested the writing tasks and feedback forms to assess their comprehensibility. This allowed the writing tasks to be improved and the questionnaire items to be adapted for comprehensibility. The feedback from these intervention students at the end of the project was con‐ sistently positive. The participating teachers also considered the project useful, increasing their motivation to support the doctoral project. The implementation of the project was observed, reflected on, and modified collaboratively with the other German teachers at the vocational school. The groups were randomized during the pilot phase. However, it became apparent in practice that when 120 2 Theory <?page no="121"?> students in both groups communicated and compared their experiences, some students in the control group, taught using the conventional teaching method, questioned why they had not been taught the same way as in the intervention group and perceived their lessons as a disadvantage. The control groups also received a crash course after the intervention at the end of the project, as was the case with the intervention group. Nevertheless, it could be concluded that randomization at a school is advantageous from a diagnostic point of view, but is difficult to maintain in practice, especially with older students over an extended period. Vocational school students know that the lesson content is relevant to the examination and are therefore motivated to learn new content. 2.8 SUMMARY OF THE STATE OF RESEARCH Until now, writing strategies such as POW or TREE have always been tested on the respective text types. There are still no evaluated writing strategies in the form of mnemonics for writing letters according to the DIN standard. For this reason, mnemonics across text types are a sustainable option for building up knowledge of text types in terms of content, which can also be transferred to other text types and learning situations. Self-efficacy expectations can be built up positively in writing through school instruction and writing lessons, so pupils also have the confidence to be good writers. This is particularly evident in the studies by Graham and Harris (Sawyer et al., 1992; Graham et al., 2005). Glaser (2005), on the other hand, was unable to measure a positive effect on self-efficacy in her intervention study. The results on self-efficacy expectations and writing attitudes are therefore contrary to the studies by Graham and Harris, which emphasize precisely this positive effect. Performance feedback, whether in the form of feedback, praise, suggestions, or criticism, should be delivered appropriately by the teacher and fellow students. Therefore, feedback should be offered critically and objectively. When given effectively, it has a positive impact on the quality of the text. Working with lists of criteria could also be used as a feedback form when assessing the performance of student texts. In all the studies, peer feedback is repeatedly used as a piece of the writing process puzzle. On the one hand, mnemonics can support this more successfully. It is also recommended to use mnemonics to convey the required text elements, e.g., through checklists (Fix, 2008: 176). It can also be observed that the writing interventions are often implemented with individual pupils or small groups of pupils, but rarely with entire classes, in spite of the fat that whole-class instruction is the everyday teaching routine 2.8 Summary of the state of research 121 <?page no="122"?> for most teachers in Germany. Whether the positive writing effects are due to the intervention program or to the small group of students must be critically questioned. The fact is that text length and quality improved in the writing studies presented. The current research also shows that writing interventions tend to take place on paper rather than on a PC, although even primary school pupils learn to write on a PC. Furthermore, the text types examined so far are not yet vocationally oriented, although most pupils in Germany start vocational training after the 10th grade and, as consumers, are a strong target group for offers. The introduction of digital media in writing lessons opens up a new field of research that has not yet been exhausted and requires research in terms of medium, grade level, school type, and text type (Schneider & Anskeit, 2017: 295), namely research with entire classes to increase external validity (Glaser & Meyer, 2017: 381; Schindler, 2017: 118; Jakobs, 2008: 5). Therefore, there is a need for research into writing interventions on the PC with work-related text types in secondary school classes. Based on SRSD, a writing project should be developed that German teachers can complete with their classes. However, initiating a writing project requires methodological considerations, which were explained in the theory section. In terms of research methodology, writing research must face several challenges. On the one hand, the “[…] authenticity of even didactically produced learning situations” (Feilke, 2017: 166, translated by WKG) should not be neglected. However, this must also include the control of environmental variables in the test situation. Secondly, “The orientation of teaching towards empirical writing development, which has been demanded since the 1990s, is difficult to implement” (ibid.). The assessment of curricular writing competence expectations is empirically complex. The theoretical background shows that the characteristic of writing compe‐ tence is a construct that is measured by several latent variables. With reference to the empirically based writing models of Hayes (1980, 1998, 2012) and Flower (1980), the writing process is influenced by several indicators, including attitude towards writing, knowledge of the text type and the writing task as well as knowledge and application of writing strategies. These indicators combine to form the latent variable of writing competence, which Bachmann and Becker-Mrotzek illustrate in their theory-based writing model (2017). The term writing competence has been formed for this work based on several definitions as follows (Grabowski, 2017; Philipp, 2015; Efing, 2015; Neumann, 2014; Hayes, 2012; Bachmann & Becker-Mrotzek, 2012; Baurmann & Pohl, 2011): 122 2 Theory <?page no="123"?> Writing competence is a construct whose performance can be evoked on the one hand via text quality as a latent variable by means of an associated writing task, and on the other hand can be self-assessed, e.g., by means of a questionnaire, which examines, among other things, the subjects’ knowledge of text types, writing attitudes and use of writing strategies. In addition to the determination of writing competence as a performance test and self-assessment, the observation of writing performance is also included as a third component. The investigation of writing behavior in the process is relevant for this, e.g., with the help of audiography, videography or keylogging programs. Therefore, there is not just one measurable item for determining a student’s writing competence, but a whole bundle. The present nomological network for measuring writing competence requires several writing tasks that are motivated, situational, and testing at several measurement points (see Fig. 30). It calls for questionnaires that cover the parameters of writing competence, from attitudes to text type knowledge and writing strategy usage to students’ background characteristics. Fig. 30: Measuring writing skills (own illustration) A performance test and the survey alone do not allow any conclusions to be drawn about the actual value of writing competence. This is why writing didactics and writing process research demand that the writing process be examined. Writing behavior must therefore be observed in the classroom. In qualitative social research, the observation of student conversations using audio 2.8 Summary of the state of research 123 <?page no="124"?> recordings is well established. Classroom activities can also be videotaped, especially the conversations that lead to a writing process. With newer measure‐ ment techniques, the writing process can also be documented on the computer using keylogging programs. “One methodological suggestion is that the field could benefit from more mixed-method studies that combine experimental demonstration of effects with qualitative analyses of classroom interaction […]” (MacArthur, 2017: 247). Therefore, it is necessary to use several test instruments. The data obtained are “objective and subjective indicators” (Bühner, 2011: 85) of writing competence. To investigate writing processes in an authentic learning context, longitudinal studies in the school setting are a tried and tested method. Since teaching never takes place in the laboratory but in the classroom, the confounding variables must be minimized from a research methodological perspective to measure the construct. This is a major challenge, particularly in surveys with learning groups and classes, as the test requires the quality criteria to be taken into account. “However, studies on the effects of writing training have rarely been carried out to date” (Glaser & Meyer, 2017: 373, translated by WKG). The interaction of predictors of text quality, which are collected simultaneously, also requires further research to revise and develop writing models (Schmitt & Knopp, 2017: 249). There is a lack of empirically tested, further developed writing models from the perspective of didactics (Glaser & Meyer, 2017: 382). “What is needed are statements on domain-specific requirements, writing tasks, strategies, media working environments, and text types (function, form, content, design requirements) […]” ( Jakobs, 2005: 34, translated by WKG). Consequently, school-leavers are required to know text patterns and text type conventions for professional writing (Efing, 2008: 32): “In general, there is a lack of preparation - beyond the text types report, CV, protocol - for writing standardized, normed texts and, in this context, for the productive use of templates and text modules” (Efing, 2013c: 249, translated by WKG). “Developing writing tasks that are closer to everyday life and work seems to me to be another desideratum” (Schindler, 2013: 186, translated by WKG). Therefore, the didactic view of writing should also be directed towards the next higher school level, upper secondary level. To offer holistic (professionally oriented) writing lessons, a linguistic prag‐ matic focus on texts is required in addition to the linguistic system: “Teachers need to change their attitude in the sense of a reorientation away from traditional focal points such as ‘spelling, punctuation’ towards the training of communicative skills and mastery of text types” (Efing, 2008: 31, translated by WKG). As a result, a writing routine can become established that inhibits writer’s block (Efing, 2008: 32). This is not to say that spelling and grammar 124 2 Theory <?page no="125"?> lessons are not important, but according to Efing (ibid.), these two sub-areas can also be developed in a complementary and integrative way during writing. Another challenge is linking didactic research on writing competence with reference to process orientation and text competence with reference to the mastery of linguistic patterns (ibid.). In this context, studies on writing inter‐ ventions reveal a notable gap: “There is also an obvious gap in research on older young people, especially in vocational education and training, but also at upper secondary level” (Philipp, 2017: 198, translated by WKG). Writing support measures such as SRSD are also an approach for pupils in vocational education that has not yet been sufficiently researched (ibid.). SRSD studies are primarily conducted in the Anglo-American region, where they were initially developed. Graham and Harris (2017: 33), however, also call for studies in other countries. There is a lack of research with the SRSD approach and different text types. Thus, there is also a lack of other strategies that need to be adapted to the text type (Foxworth & Mason, 2017: 83). “A meaningful body of research on the development of writing would be very useful in further informing research on writing, including SRSD research” (Harris & Graham, 2017: 145, translated by WKG). MacArthur summarizes well the advantages of writing training with the teaching of writing strategies, also for the upper secondary level: One need is for research that extends strategy instruction into secondary school, college, and adult learning. […] Another direction for instructional development and research is more work on evaluation and revision. […] Evaluation criteria can be used to teach a wide range of genre features and characteristics of good writing that are not captured by organizational elements and that are not normally considered during planning, such as, tone, clarity, vocabulary, coherence, and style. The quality of stories, for example, includes more than just the story grammar elements […] (MacArthur, 2017: 246). The studies, theories, concepts and, preliminary investigations presented so far lead to a research idea presented in the following chapter. 2.8 Summary of the state of research 125 <?page no="126"?> 2.9 RESEARCH IDEA “Writing plays a key role in academic success and in an individual’s professional and social development” (Fidalgo et al., 2017: 3). With this sentence, international researchers Fidalgo, Harris and Braaksma begin the introduction to their chapter on how to teach writing effectively. Examining evidence-based writing studies provides important insights into effective writing instruction (Graham & Harris, 2017: 13), as this is a research gap that has to be further explored. To date, writing studies with SRSD have primarily been conducted at the elementary school. However, the current state of research on communicative requirements in vocational schools clearly indicates that vocational school students also need enhanced writing skills. This refers not only to the text types already taught, but also to the explicit instruction of text-type-specific writing strategies and the transparency of the writing process in all its phases - in line with Kellogg’s call (2008: 15): “[…] training the writer in planning, sentence generating, and reviewing skills”. At lower secondary level, the core curricula and educational standards show that the writing process, especially the planning phase of texts, receive more emphasis in German lessons. However, the SRSD approach, which is the only method to have been empirically tested for effectiveness, has so far been explored in German-speaking countries only by the psychologists Keßler and Glaser and through a meta-analysis of several studies by Philipp. This is despite the growing inclusivity of schools, where students with writing difficulties in particular require greater self-regulatory support, which often has to be provided by a teacher in a full classroom: Nevertheless, the complex and global nature of these interventions entails two drawbacks: from an educational point of view, teachers may not feel sufficiently competent to apply this kind of instruction to their own curriculum without the external support of researchers, whilst from a theoretical point of view, it is difficult to grasp the underlying mechanisms or features of these kinds of interventions that may explain their effectiveness (Robledo-Ramón & García, 2017: 39). Writing on the PC has so far only been anchored in the secondary level I core curriculum through the application letter text type. However, the responsibility for teaching professional text types on the PC—essential for participation in society as both a consumer and employee— cannot rest solely with vocational schools through general German or communication classes. As a rule, general 126 2 Theory <?page no="127"?> education schools do not prepare students for professional text types, but rather for transfer text types such as letters of application (ibid.). These results of the preliminary research indicate that the training of professional writing skills is necessary for pupils in general education schools, particularly during the transition to and within commercial vocational training. However, there is currently a lack of intervention studies in line with “Ger‐ man lessons in preparation for training” (Efing, 2013a: 22, translated by WKG). The following project idea emerges from these variables: In an intervention, both ninth grade classes at general education schools and training classes at vocational schools are to undergo a pre-vocational writing project in equal measure (see Chapter 3). In addition, control groups will carry out the same preand post-tests as the intervention groups, before and after the intervention. The application letter, based on DIN standard 5008, will serve as the foundation for teaching a more complex professional text type, the unsolicited offer (advertising letter), in the next phase. The students will compose these letters on the PC and learn how they are structured and formulated. Revising on the PC is easier than on paper, as entire text passages can be moved, modified, or supplemented more efficiently. Furthermore, students are taught writing strategies for planning, writing and revising texts as part of a step-by-step cooperative writing process. These skills are not only beneficial for training occupations in the hotel and catering industry, but are relevant for a wide range of IHK training occupations (Giera, 2010: 67). The writing process is thus designed independently with the help of methods and working techniques in accordance with the educational standards for the subject of German (KMK, 2004: 12). The texts produced by the students (“text forms”, Pohl & Steinhoff, 2010: 6) should also be understood as “forms of learning” (ibid.) and be subjected to student reflection. Through this project, students not only learn subject-specific text types, but also acquire technical vocabulary and discourse rules in concrete contexts, which are prerequisites for developing technical language skills from novice to expert ( Janich, 2012: 11). Multiple data collection points can provide an insight into the text quality and how it changes as a result of the intervention. For example, audio recordings during the feedback discussions provide trans‐ parency into students’ conversations, and can be compared with the revisions made to the pre-text, thus highlighting revision steps influenced by peer group recommendations. The use of evaluation tools such as checklists can be particularly beneficial for writers in communicating both the content and formal features of the text. Psychologically, this is based on the principle of reciprocity: when individuals receive extensive feedback, they are more likely to provide 2.9 Research idea 127 <?page no="128"?> similarly detailed feedback in return. However, the feedback criteria should be objective (Goldin & Ashley, 2012: 217). Feedback processes among classmates have the advantage that differing prior knowledge and experiences can lead to a variety of solutions, thereby enriching the writing process. The writing task can thus be perceived as a shared problem that needs to be solved collaboratively. Compared to peer interaction, student-teacher feedback tends to be less authentic. (Goldin et al., 2012: 112 f.). Peer feedback is a frequently used method in writing lessons. In this form of interaction in particular, students also learn from each other through observation (Couzijn & Rijaaarsdam, 1996, 2004; Graham & Perin, 2007). German didactics calls for interventions that can be implemented in the classroom and, in the best case, are realistic German lessons: “With a view to research, the question should increasingly be asked under which conditions the findings of German didactics can be made fruitful for the development of lessons” (Pieper, 2018: 8, translated by WKG). In summary, there is a clear need for research on intervention studies using the SRSD approach with entire classes in Germany (Keßler, 2010; Graham & Harris, 2017; Philipp, 2017) with professionally relevant text types for the transition from school to training (Efing, 2018; Hoefele & Konstantinidou, 2018; Neumann & Giera, 2018; Steinhoff, Grabowski & Becker-Mrotzek, 2017; Schindler, 2017; Efing, 2012; Giera, 2010; Jakobs, 2005, 2008; Wyss-Kolb, 1995) and incorporate new media to depict and analyze the writing process (Lehnen, 2017; Linnemann, 2017; Steinhoff, Grabowski & Becker-Mrotzek, 2017; Philipp, 2017; Weinzierl & Wrobel, 2017). The empirically verified implementation of professionally relevant text types in German lessons is a necessary step in preparing students for professional activities and everyday life. 2.10 RESEARCH QUESTIONS AND HYPOTHESES The initial situation and the current state of research indicate a clear need for intervention studies aimed at developing career-oriented writing skills. This doctoral project addresses that research gap by conducting an intervention study. The structured intervention explores a fundamental question at the interface between school and work, namely: 128 2 Theory <?page no="129"?> 10 These four areas refer to Hayes’ writing process models (1996, 2012; see chapters 2.5 and 2.10 and Fig. 20). How do students from general and vocational schools develop their writing skills over the course of a writing project using the “Self-Regulated Strategy Development” approach and the two professional text types “cover letter” and “unsolicited offer”? To answer this question, four more specific research questions are formulated from the four research perspectives of Affect, Behavior, Content Learning and Metacognition (see Figure 31). 10 This perspective is grounded in the observation by writing researchers that writing success is dependent on a variety of factors and that affect, cognition and behavior in particular influence the writing process of writers (Fidalgo et al., 2017: 5). The Affect dimension refers to Hayes’ 1996 model of writing and focuses on students’ attitude towards writing (Beliefs & Attitudes, see Fig. 20). The Cognition perspective was subdivided into Content Learning and Metacognition. Content learning refers to the latent knowledge about a text type. Metacognition concerns the use of writing strategies, such as planning and revising. Metacognition thus involves self-reflective thinking about one’s own writing processes (Kellogg, 1994: 17). For each research question, a corresponding null hypothesis and alternative hypothesis are formulated, which are relevant for answering the research questions. These are presented in alphabetical order for clarity. It is important to emphasize that the following formulated hypotheses are not causal hypotheses intended to test the SRSD training method, but rather a framework primarily for conducting empirical research within a teaching context that anticipates a positive learning effect. The primary objective of the writing intervention is to improve text quality, measured in terms of content, linguistic and formal criteria, across both text types (see Chapter 3.1 Method). As previously discussed, the effectiveness of the SRSD approach has already been empirically validated. Within this school-based writing project, students’ learning development will be observed and analyzed through the lens of four specific research questions. 2.10 Research questions and hypotheses 129 <?page no="130"?> Fig. 31: The four perspectives of the study (own illustration) The corresponding null hypotheses (NH) and the alternative hypothesis (AH) are formulated for each research question (abbreviated to F). The null hypothesis represents a “zero effect” or “no difference” (Rost, 2007: 61). As Rost (2007: 81) notes, “Null hypothesis significance testing in quantitative-empirical research has been increasingly used since the early 1950s […] and has now established itself as a standard method.” The following question arises for the Affect research area: • Q 1 : What is the relationship between the students’ writing attitude and their writing performance during the writing project? Hayes (2012) already highlighted the attitude towards writing as an important variable for writing success in his writing process model. The intervention studies using the SRSD approach have shown a connection between self-efficacy in writing and writing performance (Bandura, 1977; Sawyer et al., 1992: 348; Graham et al., 2005: 231; Pajares & Valiante, 2006: 158 ff.; Keßler, 2010: 198 ff.; Giera, 2013: 61). Therefore, it can be expected that the writing performance of the test subjects in terms of text quality will be higher if the writing attitude (self-efficacy expectation) is equally high. However, it is still unclear at this point whether the students’ self-efficacy in writing can be improved during the course of the intervention. Nevertheless, the better the writing attitude and the associated self-efficacy, the better the expected writing performance (Sawyer et al., 1992: 348; Graham et al., 2005; Keßler, 2010: 198 ff.). The following hypotheses are thus proposed: 130 2 Theory <?page no="131"?> • AH 1 : If students have a positive attitude towards writing business letters, they also write “application letters” and “unsolicited offers” at a medium to high level (overall impression) compared to those with a negative attitude towards writing (Bandura, 1977; Sawyer et al., 1992: 348; Graham et al., 2005: 231; Pajares & Valiante, 2006: 158 ff.; Keßler, 2010: 198 ff.; Giera, 2013: 61). • NH 1 : Students who have a negative attitude towards writing business letters write these letters at a low level (overall impression). The following question arises for the Behavior area of investigation: • Q 2 : How well do students perform in writing “cover letters” and “unsolicited proposals” during the writing project using the “Self-Regulated Strategy Development” approach? Intervention studies by Graham and Harris (e.g., Graham, Harris & Troia, 1998; Troia, Graham & Harris, 1999; Graham, Harris & Zito, 2005) and the first intervention results in Germany by Glaser (2005) and Keßler (2006, 2010) demonstrate improved text quality through the SRSD approach. It is therefore expected that, with the help of the SRSD approach, the participants in the project will produce better-quality application letters and job offers after the intervention than before. With regard to the students’ competence in text type, it is also anticipated that they will perform better in writing cover letters in the pretest than in writing offers, as cover letters have already been taught in the curriculum and the students can draw on prior knowledge of this text type (Hayes, 2012: 371): “Writing about topics that students know well provides a scaffold to support the writers and to allow them to devote a higher degree of executive attention to the juggling of planning, generating, and reviewing” (Kellogg, 2008: 15). Whether this prior knowledge in the area of application letters forms the basis for stronger intervention effects, and thus for higher text quality in the posttest, remains an open research question. Furthermore, it is also necessary to investigate the level (Kessler & Ziener, 2004) at which the students write and revise their texts during the process. In the intervention group, increases are expected in the following two areas: 1. the overall impression in the global assessment and 2. the number of words. The overall impression is highly informative with regard to text quality. The number of words is a predictor for the assessment of text quality, as demonstrated by several studies (Neumann & Matthiesen, 2011). A high word count predicts higher text quality in short to medium-length school texts 2.10 Research questions and hypotheses 131 <?page no="132"?> (Neumann, 2010: 22). It can therefore be hypothesized that the word count, the number of content features considered, and the level of global assessment— as variables of text quality in the application letters and unsolicited offers— will be lower at the beginning of the writing project (i.e., in the pretest), than at the end (i.e., in the posttest). It can also be assumed that, in the pretest at the beginning of the intervention, students will demonstrate higher text quality in the already familiar application letter text type than in the unsolicited offers. The following hypotheses can be derived from this: • AH 2 : At the beginning of the writing project, students participating in the intervention write “application letters” and “unsolicited offers” predominantly at an inter‐ mediate level with regard to the variables overall impression, language pragmatics, language system, and number of words. They maintain or slightly improve this level over the course of the writing project (Graham, Harris & Troia, 1998; Troia, Graham & Harris, 1999; Graham, Harris & Zito, 2005; Kellogg, 2008: 15). • NH 2 : Pupils who participate in the intervention write “letters of application” and “unsolicited proposals” predominantly at an intermediate level, which deteriorates over the course of the writing project. The following question arises in the area of Content Learning: • Q 3 : What do students know about the content of the text types “cover letter” and “unsolicited offer” at the beginning and end of the writing project (intervention)? The assumption of this research question is that, at the beginning of the writing project, students from general education schools will have fewer writing skills in writing professional text types, especially the unsolicited offer, than students from vocational schools. This may be attributed to a lower level of familiarity with text samples (Hayes, 2012: 371). Both the prior knowledge of the respective text types and the amount of practice with them are likely to be lower among students in general education schools (Hayes, 2012: 371). However, it is expected that the number of content features of the respective text type will increase as a result of the intervention. Furthermore, pupils with a higher level of text type knowledge are expected to produce higher-quality letters than those with little or no such knowledge (Schindler, 2005: 224). The following hypotheses can be formulated from this: 132 2 Theory <?page no="133"?> • AH 3 : Compared to the control group, students in the writing project ex‐ hibit greater text type knowledge about “cover letters” and “unsolicited offers” at the end of the writing project than at the beginning (Schindler, 2005: 224; Hayes, 2012: 371). • NH 3 : Compared to the control group, students in the writing project show stagnant or lower text type knowledge of “application letters” and “unsolicited offers” at the end of the writing project than at the beginning. The following question arises in the area of metacognition: • Q 4 : What letter-writing strategies do students use in the planning and revision stages? The willingness to incorporate text-type-specific content and explicitly taught writing strategies for planning and revising into the usual writing process may be more difficult for students who are already accustomed to “text routines” (Feilke, 2010: 4) (Hayes, 2012: 371; Giera, 2013a: 60). Nevertheless, accepting and giving feedback is expected to be essential in improving text quality during the revision process. It can be assumed that students who receive feedback from peers are more likely to revise and enhance the quality of their texts (Xiong et al., 2012; Mirzaii & Aliabadi, 2013). The following hypotheses can therefore be formulated: • AH 4 : Text quality and length improve as a result of writing tasks in the intervention, when students plan and revise their “cover letters” and “unsolicited offers” using the writing strategies promoted in the eleven-week writing project (90 minutes once a week) (Feilke, 2010: 4; Mirzaii & Aliabadi, 2013; Giera, 2013: 60). • NH 4 : The quality and length of the texts produced by the writing tasks set in the project improve less when students make limited use of the writing strategies promoted in the eleven-week writing project (90 minutes once a week). The questions and the associated hypotheses can therefore be categorized into the two writing process models as follows (see Figs. 32 and 33). The abbreviations represent the four areas of investigation: Affect, Behavior, Content Learning and Metacognition. 2.10 Research questions and hypotheses 133 <?page no="134"?> Fig. 32: Examination perspectives in the basic model of text production by Bachmann & Becker-Mrotzek (2017: 42, translated by WKG) Fig. 33: Research perspectives in the writing process model by Hayes (2012) 134 2 Theory <?page no="135"?> 11 The position of the examination areas in the models could also be added at other points. The aim is to empirically investigate the relationship between these four areas and text quality. 11 Based on both models, it becomes evident that writing competence comprises a large number of variables; however, this study focuses exclusively on four areas of investigation. Following the discussion of the theoretical and scientific foundations in Chapter 2, the research design with the associated methods will be presented in the third chapter. 2.10 Research questions and hypotheses 135 <?page no="137"?> 3 RESEARCH DESIGN & METHODS To provide a clear overview of the research design, Chapter 3 begins by explaining the methods in the first section, followed by a description of the sample in the second, and the test instruments in the third. In the fourth section the timing and the content of the writing project are outlined. This is followed by a tripartite discussion of data collection, processing and evaluation. 3.1 METHOD “The changes associated with learning must be observable or otherwise demonstrable, whereby behavior and movements, but also changes in physiological, cognitive and emo‐ tional reactions are suitable as indications of learning processes” (Winkel/ Petermann & Petermann, 2006: 17, translated by WKG). Constructs such as writing competence cannot be observed directly; they must be examined through tests (Ingenkamp & Lissmann, 2008: 59). Tests in educational settings can examine a wide range of areas, including learning processes and school performance (Bühner, 2011: 19). A test is defined as follows: Tests are methods of pedagogical diagnostics with the help of which a behavioral sample, which is intended to represent the prerequisites for or results of learning processes, can be measured as comparably, objectively, reliably and validly as possible and can be evaluated, interpreted and made usable by teachers or educators for their pedagogical actions (Ingenkamp & Lissmann, 2008: 105, translated by WKG). In this study, the status quo of writing competence and its progression in the form of an intervention are to be observed. The intervention also aims to teach text patterns, which are also addressed by Hayes under “Writing Schemas” (Hayes, 2012: 371). The school writing project is therefore set up as an eleven-week long-term intervention and as a field study with five exemplary classes in Lower Saxony, which are chosen unsystematically. The resulting lon‐ gitudinal diagnosis examines “changes in characteristics over time (progression profile) for individuals or groups” (Bühner, 2011: 28, translated by WKG). This <?page no="138"?> study is thus carried out as a long-term intervention in a pre-posttest control group design: Typically, both groups are tested before and after; in between, the experimental group receives a treatment whose effect is to be proven and the control group does not (Grabowski, 2017: 328, translated by WKG). Strictly speaking, when the groups are compared with each other, a cross-sec‐ tional analysis is also methodologically involved, as it examines “differences in the expression of characteristics between individuals or groups” (Bühner, 2011: 27, translated by WKG) are examined. Therefore, both longitudinal and cross-sectional diagnostics can be applied within the context of this study. „Field research [is] a branch of social research based primarily on studies by Kurt Lewin, which aims to observe people under natural conditions (rather than in laboratory settings) and to intervene as little as possible through the research method” (Mayring, 2010: 33; [modified by W.-K. G.]). Field studies have the advantage of being conducted in authentic classroom environments. In such settings, interventions become an “[…] integral part of the lesson” (Marx & Steinhoff, 2017: 253, translated by WKG). As a result, they demonstrate external validity, by yielding real-life outcomes (Rost, 2007: 67). Uncontrolled variables are assumed to have random effect on the test subjects (Rost, 2007: 67). If the lessons with the intervention are delivered by a researcher rather than the regular classroom teacher, the study reflects a hybrid of field and laboratory research, as the usual classroom dynamics are altered (Marx & Steinhoff, 2017: 253 f.). In this project, the regular German teacher is present on a random basis to monitor attendance and the teaching process. The well-known “Hawthorne effect” (Grabowski, 2017: 320) should be minimized, as the intervention is designed to span several weeks and not for a short period of time. Assessing the learning status and growth in writing competence through a lesson-related intervention, using a pre-test and a post-test in the form of a writing task, is a prerequisite for measuring the construct of writing competence with the latent variable of text quality. Measuring writing competence using writing tasks as input and the corresponding student texts as output is a common approach in writing didactics research (Grabowski, 2017: 318). On the other hand, repeated assessment is especially necessary for longer intervention periods (Steinhoff & Marx, 2017: 261), because the construction of the true value is defined as the “mean value over an infinite number of repeated independent measurements (t) of the observed values (x) of a person (v)” (Bühner, 2011: 43, translated by WKG). This approach seeks to minimize errors identified in classical test theory as “measurement error theory” (Bühner, 2011: 42, translated 138 3 Research design & methods <?page no="139"?> by WKG). On the one hand, repeated measurements aim to approximate true values, the results may still be distorted by “practice and transfer effects” (ibid.), “unsystematic external influences” (ibid.) or by “unsystematic internal influences” (ibid.). This test theory therefore has certain limitations. The student texts produced during the writing project are subjected to content analysis. This method is intended to examine a “fixed communication” (Mayring, 2010: 13, translated by WKG), and the researcher is expected to “proceed systematically” (ibid.), in a “rule-based” (ibid.), “theory-based” (ibid.) manner to “draw conclusions about certain aspects of communication” (ibid.). The student texts are interpreted qualitatively within their communication context, even if the coding and subsequent calculations are quantitative (Mayring, 2010: 48). Thus, the writing task evokes writing competence at a specific point in time with a specific sample. The text is assessed as a product with the help of a coding system, where each code corresponds to an item. Several items, in turn, correlate with the superordinate latent variable in the sense of “reflective and formative indicators” (Bühner, 2011: 32 f., translated by WKG). The content analysis procedure follows the approach recommended by Rost (2007: 76). In addition to the use of writing tasks, content, formal and linguistic indica‐ tors in a text must also be examined to determine text quality and writing competence. These indicators are also items that contribute to the construct of writing competence and the latent variable of text quality. Along with this product-related methodology, a process-oriented approach must also be supplemented. Therefore, the evaluation of conversations, both between pupils and between pupils and teachers, is enriching. Surveys administered using questionnaires at various points during the intervention can document the status, progress and effect of an intervention through students’ self-assessments. When pupils write using computers, the use of keylogging software is also established to map writing processes at different moments. The aim of the test is to distinguish groups with below-average writing skills (struggling writers) from those with average to above-average writing skills. The test is thus based on an “external test construction” (Bühner, 2011: 93): Usually, a large number of items that could potentially differentiate between the groups are initially collected during test construction. […] This can manifest itself, for example, in large differences in the mean values of the items between the groups (Bühner, 2011: 94, translated by WKG). The student texts are evaluated quantitatively by at least two independent raters using a coding manual grounded on the DIN 5008 standard and the “Indicator Model of Scholastic Writing” (Neumann & Matthiesen, 2011, translated by 3.1 Method 139 <?page no="140"?> WKG), the IMOSS coding method (see Section 3.3). Rating procedures are used to assess writing products, such as student texts, to measure text quality as a complex construct. Text quality is conceived as follows: “Quality refers to judgments about how well a document communicates or achieves its purpose with intended audience” (Kellogg, 1994: 55). The combination of holistic and analytical ratings provides an overall assessment of the student text on the one hand and evaluates partial aspects of the text on the other (Marx & Steinhoff, 2017: 262). Although analytical scoring, which relies on a set of specified evaluation criteria, is more time-consuming than holistic assessment, it is generally more reliable (Kellogg, 1994: 57). Neumann (2014: 63), referring to Weigle (2002: 121, translated by WKG), differentiates between the two rating procedures based on the criteria of “authenticity”, “practicability”, “reliability” and “impact”. The integration of both rating procedures, together with a coding manual, benchmark texts from a preliminary study, and rater training, results in a reliable and differentiated assessment of writing performance (Neumann, 2014: 64). The rationale for the relevance of this approach is outlined as follows: It should be possible to find every target dimension in the evaluation. Data that has not been recorded cannot be reconstructed afterwards, but summarizing individual characteristics always works. […] After defining the variables and ‘labeling’ them, the data can be collected on certain graded values that present the degree of the respective implementation. Dichotomous judgments as low-inference decisions, as just noted, usually prove to be more reliable between the assessors than multi-level, high-inference ones (Neumann, 2014: 63, translated by WKG). The majority of the questionnaire responses are automatically converted into data using the EvaSys software (Electric Paper Evaluationssysteme, 2018). For the close-ended questions, which are mainly based on a four-point rating scale (ranging from 1 = “strongly disagree” to 4 = “strongly agree”), the mean value of thematically related items is calculated during the evaluation and compared across other groups and measurement times. The open-ended questionnaire responses must first be coded and then analyzed. In addition to frequency analyses in the form of a descriptive data analysis, correlations between the questionnaire data and the student texts will also be considered. Furthermore, factor analyses and significance tests will be conducted. To compare selected classes from vocational schools with those from general education schools, a cross-sectional analysis is conducted across different groups in the institutions of general education and vocational schools (see Section 3.2). As self-reports by test subjects are not always accurate and is 140 3 Research design & methods <?page no="141"?> subject to the “risk of reactivity” (Grabowski, 2017: 319, translated by WKG), it is necessary to use several test instruments. The Inputlog software (Leijten & Van Waes, 2013) provides an objective tool for making individual students’ writing processes visible. Pauses and revision steps are recorded with precise timing. These time measures can then be ana‐ lysed with the qualitative data from the audio recordings. Thus, the qualitative data from the feedback discussions and writing processes are correlated with the quantitative data obtained from the student texts and questionnaires. To investigate writing processes effectively, it is necessary to code the student texts on the one hand and to analyze the related conversations and processes on the other, a methodological approach that has gained increasing attention since “[…] the mid-2000s” (Neumann, 2014: 54; [modified by W.-K. G.]). In particular, conversations about the writing process offer significant potential for evaluation: The information-transmitting, cognitive process-controlling and instructional activ‐ ities of the teacher and peers in the classroom are also to be regarded as an institution-specific form of communication that functions as an acquisition context for institutionally appropriate language use (Quasthoff & Heller, 2014: 7, translated by WKG). Therefore, both peer conversations among students and classroom discussions between teachers and students are observed and recorded during the modeling phase to appraise the “process quality” (Neumann, 2014: 54, translated by WKG). Students’ peer group conversations are audiographed using recording sticks, as these help to minimize observer effects (Krelle, 2014: 79). To analyze the modeling phase, video recordings are made in the ninth grade during the instruction on application letters and unsolicited offers. This provides secondary data for further analysis. In this phase, the teacher composes a sample letter while verbalizing his thoughts, ideas, strategies, and sentence structures, akin to the think-aloud method, as the text gradually takes shape: Think-aloud is a research method which offers access to cognitive processes without affecting participants’ performance when solving a task if participants are just asked to verbalize what they think. Thus, the procedure to collect think-aloud protocols must be carefully designed with respect to the specific research question. The think-aloud method has been used to study different phenomena: problem solving; learning (reading, listening to stories, reading-to-write); thinking; use of memory; decision making; to mention a few examples (Castells et al., 2014: 221). 3.1 Method 141 <?page no="142"?> Pupils actively participate and give tips or suggestions for improvement. This form of oral participation “[…] is a fundamental condition for successful learning in all subjects” (Quasthoff & Heller, 2014: 6, translated by WKG). To minimize observer effects, the camera is not pointed directly at the students so that they are not afraid to verbalize their thoughts on the text. All student contributions during the discussion are anonymized; students are not addressed by name. This procedure is explained to them in advance. “With regard to German lessons, this means that there is significantly less influence if you videograph forms in which there is as large a class audience as possible” (Krelle, 2014: 79, translated by WKG). Listening skills are required both for the student discussions and for the teacher-led modeling phase. However, these cannot be directly observed: “This means that listening and listening skills are initially a latent construct, i.e., a fact whose existence can only be inferred from other, directly observable facts (indicators)” (Behrens, 2014: 43, translated by WKG). This notion underscores the necessity of using multiple methods, as highlighted in the following quote: To summarize […], it can be stated that both oral and written language competence are complex skills of productive language use. The underlying communicative tasks are shaped by corresponding practices (in the above sense), so that the cognitive processes that become visible on the surface in performance cannot lead to diametrically different assessments of competence. This places the demand for valid, reliable, and objective statements, which should also be fulfilled via common methodological requirements, at the center of further discussion (Neumann, 2014: 57, translated by WKG). The selection of methods includes both quantitative and qualitative procedures, which are integrated into a mixed method design. When assessing writing competence, it is common practice to observe both writing products and writing processes within such a design. This approach overcomes the dichotomy be‐ tween quantitative and qualitative studies and instead emphasizes the enriching potential of methodological diversity (Neumann, 2014: 72): The dispute between qualitative and quantitative analysis suggests that the synthesis lies where the qualitative analytical steps necessarily inherent in every analysis are made explicit and the points in the analytical process at which quantitative steps can be meaningfully incorporated are then identified (Mayring, 2010: 123, translated by WKG). The quality criteria of a test are taken into account and discussed in detail in Section 5.2. The requirements for test procedures are strictly regulated in 142 3 Research design & methods <?page no="143"?> science. Established test quality criteria classify each study with a critical focus. “The main quality criteria include objectivity, reliability, validity, and the scalability of a test” (Bühner, 2011: 58, translated by WKG). These main quality criteria can be further precisley defined and expanded to include additional secondary quality criteria. The following refers to a selected set of these criteria: “The implementation of a test must not vary from examination to examina‐ tion” (Bühner, 2011: 59, translated by WKG). The SRSD phase model structures the intervention within the lessons. The pre-test and post-test are conducted using both writing tasks and questionnaires. Although the content of the writing tasks varies, they are structured consistently to allow comparisons through repeated measurement. The questionnaires are employed repeatedly. As the data is collected by one person, inconsistencies caused by different teacher personalities are avoided. No assistance is provided to the students during the test phase. However, questions related to technical terms in the writing tasks are answered before the test begins. Furthermore, the writing task is always read aloud slowly to all test groups to ease the strain on working memory. These measures ensure the objectivity of the test. “Every evaluator must determine the same score or performance values for a test person. This requires precise evaluation instructions” (ibid., translated by WKG). Therefore, a coding manual is provided, which the raters who are not involved in the intervention strictly follow. All respondent data is anonymized for this purpose to do justice to the objectivity of the evaluation. Objectivity of interpretation applies to the ratings that are carried out twice, specifically for the student texts and student interviews. It should be emphasized that an assessment is always inherently subjective. However, using coding manuals with benchmark texts and the training of raters, subjectivity can be transformed into objectivity. Nevertheless, rating outcomes may still vary (see interrater reliability): Every evaluator should come to the same assessment or interpretation of the test results, such as whether the respondent’s score should be classified as average or above or below average compared to the norm group in a particular test (Bühner, 2011: 60, translated by WKG). However, the second part of the quote is crucial here. While, individual ratings may deviate slightly, marginal ratings for average, above-average, and below-average performance should be consistent. The study must address this by means of previously defined borderline ratings (ibid.). The objectivity of interpretation can certainly be increased in further studies. 3.1 Method 143 <?page no="144"?> In the study, the construct of writing competence is examined from four perspectives, using a large number of items and a diverse selection of test instruments. Rather than relying on a single measurement, multiple assessments are conducted. As empirical writing research continues to evolve, further findings and innovative research methods may emerge. The integration of productand process-related data material aims to ensure content validity: Content validity is when a test (or its test items as a whole) and each individual item really or sufficiently accurately captures the characteristic to be measured. Precise here does not mean the aspect of measurement accuracy, but rather precise refers to the representation of the construct by the item (Bühner, 2011: 61, translated by WKG). The students’ self-assessments using the questionnaires are compared with direct behavioral observations and writing performance tests. This study, therefore, aims to maintain content validity as far as possible. “Predictions are made on the basis of the results and their occurrence is investigated” (Mayring, 2010: 117, translated by WKG). This type of validity is examined by means of a hypothesis test using correlation analyses (see Chapter 4), which also serves to test the predictive validity. “The results are checked for plausibility using established theories. The appropriateness of the operational definitions is considered on the basis of the theoretical background” (Mayring, 2010: 117). The hypotheses are formulated on the basis of the theories and models previously outlined. Therefore, the results of this study should represent construct validity: Research results that are closely related to one’s own research question and the object of investigation and whose validity one is convinced of are used as a benchmark (Mayring, 2010: 117, translated by WKG). The external criterion is also a component of validity. The results of the SRSD study (see Section 2.6) show a high degree of comparability, as the number of studies conducted and the consolidation of their results in the meta-analysis are highly representative. For the text types themselves, on the other hand, there are only a few studies suitable for comparison, as a writing task would have to be tested several times in a similar format. The studies by Baurmann (2014) and Neumann (2011) offer only pointed comparisons in the context of letter writing. The external criterion is, therefore, only partially available and must be further examined for its suitability in subsequent studies. Ecological validity refers to the relevance of the study to real life contexts (Rost, 2007: 68). This is especially important here, as the study is conducted as a quasi-experimental field study within German lessons.Although lump samples can increase the risk of α-error, the results of the pretests should demonstrate 144 3 Research design & methods <?page no="145"?> that the subgroups within the age groups are homogeneous (Rost, 2007: 96). This criterion leads to the next one - namely, reliability: “The reliability of measurements is defined as the variance ratio. It is the proportion of the variance of the true values in the variance of observed values” (Bühner, 2011: 147, translated by WKG). The control group is examined at two different measurement times. This is to ensure the reliability of the initial survey, which is equivalent to a “re-test” (Mayring, 2010: 116, translated by WKG). The estimate is made “[…] by correlating the raw score pairs between the first test and the test repetition with the same test” (Bühner, 2011: 159, translated by WKG). The questionnaires are sometimes used multiple times. The same results are not expected for the intervention group. For this reason, the measurement of the construct writing competence is conducted using several test variants to check the consistency of the results, following the logic of a “parallel test” (ibid.). This involves calculating “maximum likelihood estimates of reliability” (Bühner, 2011: 158, translated by WKG). Furthermore, correlations (= r) of the ratings of student texts are calculated as the degree of internal consistency, following the evaluation of interrater reliability with Cronbachs-α (ibid.). The interrater reliability determines the degree of agreement between the raters. The more complex the rating process, the lower the likelihood of exact agreement (Mayring, 2010: 117). “It should be noted at this point that it is also unrealistic to measure without measurement error” (Bühner, 2011: 70, translated by WKG). Furthermore, scaling is necessary: “The quality criterion of scaling means that the formation of a test value is carried out using a valid calculation rule” (Bühner, 2011: 67, translated by WKG). The analytical ratings according to the IMOSS coding procedure are considered both seperately and in summary across two scales. These scales were tested for validity by Neumann and Matthiesen (2011), and (Neumann, 2007). The scales for the analytical-content ratings, derived from the content checklists for each specific text type, are summed based on the existing items. These checklists are grounded in expert interviews and comprehensive literature research. The sum of the items to be determined for the letter content is intended to reflect the extent to which the respondents addressed the relevant content elements, thereby indicating their competence in letter writing. The correlation of this scale with the content item (C001) is intended to double-check the justification for the scale’s application and the respondents’ writing ability. This content scale is also used for the evaluation of student interviews and during the modeling phase. Thus, the scaling is uniform for several test instruments per text type (Bühner, 2011: 67 f.). 3.1 Method 145 <?page no="146"?> The test subjects are not subjected to unnecessary physical and psychological stress during the test procedures (Ingenkamp & Lissmann, 2008: 60). In the preliminary study, most of the test subjects completed the writing task within approximately 20 minutes and required no more than an additional 20 minutes to complete the accompanying questionnaires. As the total duration did not exceed a standard 45-minute lesson, the criterion of reasonableness is considered fulfilled (Bühner, 2011: 73). The fairness of a test refers to the extent to which individuals are not disadvantaged on the basis of gender, origin, or other demographic factors (Ingenkamp & Lissmann, 2008: 61; Bühner, 2011: 73). This can be assessed through the calculation of item difficulty: “High item difficulty means that many people agree with the item (for questionnaires) or many people solve it correctly (for performance tests)” (Bühner, 2011: 87, translated by WKG). The results of the preliminary study showed that the mean values for the evaluation of the student texts were generally good. Test fairness and average item difficulty can therefore be attested, although both should be re-examined based on the main study results. Mayring (2010: 118) supplements the classic quality criteria with further quality criteria by including procedural documentation as a vital component. This shows how the exact sequence of the investigation and the procedures used for data collection, processing and evaluation are carried out. These steps will be thoroughly documented in this study to ensure transparency and reproducibility. As a result, a retest can be conducted to compare current results with subsequent studies and establish “correlative validity” (Mayring, 2010: 119, translated by WKG). The established IMOSS coding manual will be used for the holistic and analytical coding of the student texts. This will be adapted to the text types used in the study, updated with new anchor examples/ benchmarks from the piloting phase and made available to raters in the form of a coding manual. For the analytical dichotomous coding according to content, formal, and linguistic criteria, expert interviews were conducted with five experts per text type, each possessing many years of relevant professional experience (see appendix in Giera, 2020). Semantic validity refers to the accuracy of the meaning reconstruction of the material. It is expressed in the appropriateness of the category definitions (definitions, anchor examples and coding rules). A review can be carried out by expert judgment (Mayring, 2010: 119, translated by WKG). The experts confirmed the correctness of these content, although in practice, many of these criteria are often applied implicitly rather than explicitly cata‐ 146 3 Research design & methods <?page no="147"?> logued. An excerpt from this survey is included in the appendix. During the standardization process, the test value determined for individual test subjects are compared to those of a broader sample to determine whether the performance is average, above average or below average. This comparative analysis establishes a critical frame of reference (Bühner, 2011: 71 f.). This is also applied in the present study, particularly for categorising the average and below-average performance. Pre-defined cut-off values are used for classifying individual test scores, which are based either on the median (for fixed scales) or on the mean value of all test subjects (for non-fixed scales). As testing is conducted within the participants’ regular learning groups and at a single point in time, strict adherence to standardized administration procedures is ensured to maintain the comparability of the results. The writing tasks and the questionnaires are identical for all test subjects and are available to the test participants in print form (Bühner, 2011: 72). The development of new test formats and corresponding studies is especially valuable when no existing procedures are available or when theoretical frameworks require further refinement (Bühner, 2011: 73). In this study, both the writing tasks and the questionnaires are redesigned because existing instruments from SRSD-related studies, largely in English, have not been adapted for use in German-speaking contexts. While some items for the SRSD and the IMOSS studies may be integrated into the questionnaire, the items related to specific text types, application letter and unsolicited offer, require a context-specific redesign. These new instruments are expected to offer substantial didactic value for German language instruction and make a significant contribution to empirical writing process research. 3.2 SAMPLE The selection of the sample is based on the underlying research questions and the hypotheses formulated in the Chapter 2. It takes into account factors such as age, level of education, “spectrum of experience and behavior” (Bühner, 2011: 87, translated by WKG), “language proficiency of the target group” (ibid.) and “item/ test fairness” (ibid.). In total, 136 students initially participated in the survey. During the survey period, a cohort of 16 vocational students from the catering sector, who held a secondary school leaving certificate, was excluded from the final evaluation. This decision was based on high fluctuation in attendance and a significant dropout rate during the survey period (April-June 2013). Many students in this subgroup discontinued their vocational training, and a general 3.2 Sample 147 <?page no="148"?> demotivation towards schooling was evident across subjects. The dropout rate for this specific subgroup was too high to generate reliable information. As a result, the final sample comprised 103 students (see figure below). Among them, 48 subjects were male (46.6%) and 55 female (53.4%). Approximately 31 % of participants had a migration background, aligning with the national average reported in the Microcensus (Federal Statistical Office, 2017b). Fig. 34: Composition of the sample (own illustration) The study primarily focused on ninth-grade Realschule students, as they are statistically more likely than their peers from Gymnasium or Hauptschule to pursue commercial apprenticeships through the Chamber of Industry and Commerce. The average age of the Realschule pupils in the ninth grade was 15.05 years, which corresponds with the national demographic data. Of the 103 test subjects, 68 students participated in the writing project and thus received the intervention, while 35 subjects fromed the control group, which received no treatment during the intervention period. The group served as a waiting control group, receiving the treatment at a later stage. The uneven number of control and intervention groups, resulting from organzational constraints, prevented randomization. Instead, entire classes served as intervention and control groups: It is often not possible to select a random sample, or it is not possible to randomly assign the individual participants to different groups (e.g., treatment group vs. 148 3 Research design & methods <?page no="149"?> comparison group […]) because they are so-called lump samples […] or because it is not possible for organizational or ethical reasons (Rost, 2007: 68, translated by WKG). The sample was divided into several subgroups to enable differentiated analysis according to the research questions and hypotheses. Comparisons were made between: the intervention group vs. the control group, the intervention group in Year 9 vs. the control group in Year 9 and the ninth-grade group vs. the group of vocational trainees. Due to the relatively small sample per subgroup, special attention was given to maintaining a minimum number of 35 participants: “Thus, at α = 0.05 and β = 0.20, at least large effects can be verified as statistically significant” (Rost, 2007: 94, translated by WKG). The primary focus of this study was the implementation of the writing intervension in secondary schools. The vocational school students were primarily used to compare how the writing requirements of professional contexts differ from those in Year 9 schooling. Their real-life experience with writing, particularly typing on a PC using DIN standards, is integral to their professional training. The average age of the vocational students was 20.00 years, which is in line with the national average. This subgroup consisted of three male and 23 female hotel management trainees. According to vocational school teachers, the uneven gender distribution is typical for the profession, as male applicants are often discouraged by the emphasis on housekeeping tasks such as room maintenance. The intermediate level hotel trainees (n = 26) had already successfully transi‐ tioned from school to the workforce and were more practically experienced than the other two intervention groups, although they had not yet received formal training on the “offer” text type at school. The test subjects were divided into control groups and intervention groups based on school structures. As the project focused on promoting the writing of secondary school students, this group was the most strongly represented. The hotel management trainee group was included to investigate how writing tasks and processes manifest in professional settings, such as reception and banquet departments, where writing unsolicited offers is part of the final examination. In this group, all of them have already completed at least one real application process at the start of their training. They are also anticipating another application process in in near future, as they will complete their training and possibly change companies, which is often common practice in the hotel industry. The group sizes varied considerably between vocational and secondary school students. It should be borne in mind that entire classes were tested during German lessons. Vocational classes always tend to be smaller than general schools, a structural phenomenon commonly observed in other professions as 3.2 Sample 149 <?page no="150"?> well. Therefore, the groups in this study represent unsystematically selected, exemplary groups at general and vocational schools. To ensure experimental control, the control groups continued with standard German instruction and did not receive any explicit writing instruction during the eleven-week study period, as previously agreed with the respective German teachers. All participating schools were located in the IHK administrative district of Lüneburg-Wolfsburg in Lower Saxony. The profession of hotel specialist is part of the hospitality industry and belongs to the commercial training professions of the IHK. This occupation is also among the top ten training occupations chosen by school leavers across Germany (BiBB, 2014). The background of the ninth graders and vocational school students must also be considered when planning the content and timing of the writing project: Ninth graders, typically around 14 to 15 years old, are in the midst of puberty, a developmental phase that can begin at the end of fifth grade and extend to the beginning of eleventh grade. Adolescents are primarily in search of values and are often in conflict with existing rules. For ninth graders, peer relationships frequently take precedence over academic performance. Building a good relationship with pupils is particularly important during this stage (Calmbach et al., 2017; Cleveland, 2013). Psychologist Claus Koch adds in an interview (GEW, 2017: 8 ff.) that distanc‐ ing oneself from parents and teachers is a normal phase during adolescence, as young people seek to gain their own experiences.Young people are more likely to make emotional decisions during puberty, as the neurotransmitter dopamine is released as a reward system. The so-called tests of courage are typical, especially for male adolescents. Cliques and peer groups are very important for young people to discover new environments. According to Koch (ibid.), the search for values and norms has not diminished with the rise of digital media, but rather intensified. Young people are not indifferent to school, as long as the subject matter is interesting and motivating for them. To prevent disruptions to lessons, it is important that the class teacher or German teacher review the existing rules in the learning group, based on the school’s house rules, before the project starts. Furthermore, the pupils should reflect on their learning in a diary at the end of the lesson. The teacher assesses the pupils after each student’s social and work behavior after every session and immediately informs the class teacher of any concerns. The German teacher’s unexpected and random presence during the lessons is carried out from time to time in the writing project to allow for natural observation of classroom dynamics. A variety of methods and social forms are used throughout the project, which is divided into stages. These proactive strategies, in particular, 150 3 Research design & methods <?page no="151"?> have been carefully integrated in the writing project. It can be expected that young people are generally open to using PCs, although writing texts may initially be less motivating. However, this highlights that expected behavior of the test subjects can influence the effectiveness of the intervention (Glaser & Meyer, 2017: 373). Therefore, positive work behavior must be supported through pedagogical measures such as efficient classroom management. Preventive planning before the lessons is essential. The vocational school students will likely have already gained personal experience with job applications and package arrangements, even if not all of them have written job offers as part of their curriculum or in the company. Nev‐ ertheless, they are expected to have situational and technical prior knowledge. The ninth graders, in contrast, are familiar with the job application situation primarily though internships they have already completed. Writing job offers, however, is probably a new experience to the students. Their familiarity with advertising, comes mostly from everyday life or vacationing experience with parents, but few students are likely to have sufficient prior knowledge of writing offers. Writing is a multisensory activity. German didactics must take this into account (Koch & Pielow, 1984: 41). Therefore, the writing project addresses the heterogeneity of the student body and its diverse learning paths. Even if a specific teaching method has proven effective for a certain text type, this does not mean that all pupils can be equally reached with it. One possible approach is to activate all the senses - seeing, touching, hearing and speaking. Given the number and composition of the test subjects, initial and necessary research findings related to writing professional text types at the interface between school and vocational training should be presented as trends rather than definite results. The participants’ learning backgrounds as described above also show that vocational school students possess more prior knowledge than ninth graders due to their practical experience in training companies and their involvement in various stages of application procedures. Writing unsolicited offers was not yet taught in German lessons at either school type. Neverthless, students’ engagement and behavior may improve if the topic of the writing project is interesting to them, which in turn can positively influence their social behavior. 3.2 Sample 151 <?page no="152"?> 3.3 TEST INSTRUMENTS The selection of the following test instruments can be traced back to the theoretical framework of the construct of writing competence (see Section 2.1). The ‘suitcase of test instruments’ used in this study was well-equipped, comprising a total of five test methods and twelve test instruments. These are explained below and presented in full in the appendix (see Giera, 2020). All test instruments were developed in-house (except for certain items in the questionnaire), due to the absence of existing tools that addressed the study’s specific research questions. Therefore, these instruments were tested for their suitability during a pilot phase and were further developed throughout this period, thus ensuring the collecion of primary data. The data were collected using a wide array of methods including questionnaires, student texts, audio and video recordings, and individual writing processes, most of which were documented using the “Inputlog” software (Leijten & Van Waes, 2013). The following figure illustrates the chronological deployment of the test instruments across the intervention and control groups. To measure text quality, four situated tasks, referred to as “tasks with a profile” (Bachmann & Becker-Mrotzek, 2010: 194, translated by WKG), were used in the writing project, which focused on two text types, cover letters and unsolicited offers (advertising letter) (see appendix in Giera, 2020). Writing tasks are a particularly effective method for determining text quality. All four writing tasks can be classified under the highest requirement level, Level C, according to Kessler and Ziener (2004). This requirement area demands a change of perspective in both the writing situation and the intended addressee. Furthermore, the DIN 5008 standard for composing formal letters on a PC had to be observed and applied. In both the cover letter and the unsolicited offer tasks, not all necessary information was provided. Therefore, students were required to generate their own content and ideas to construct persuasive texts. These four tasks thus demanded a transfer of knowledge. 152 3 Research design & methods <?page no="153"?> Fig. 35: Chronology & test instruments (own illustration) The advantage of choosing this test instrument lies in the fact that topic-related knowledge can be reproduced freely. However, the objectivity of the evaluation may be compromised due to the complexity of evaluation criteria. “It is therefore very time-consuming to carry out content analyses” (Bühner, 2011: 131, transla‐ ted by WKG). The content focus of the writing tasks centered on correspondence between consumers and fictitious recipients (merchants/ addressees) from the catering, hotel, and tourism industries. These sectors were deliberately chosen, as travel and hospitality are familiar domains in students’ everyday lives and 3.3 Test instruments 153 <?page no="154"?> are likely to be significant in their futures. For example, students are often familiar with travel through school trips, family vacations, or club outings. The participating school classes are located in tourist-attractive regions and are surrounded by hotels. The catering establishments referrenced in the tasks were selected based on the young people’s interest in gastronomy. For this reason, a fast-food restaurant was chosen as the context for writing an application letter task. For legal reasons, all company names and addresses used in the writing tasks were fictitious. To ensure better comparability, the identity of the letter writer was also fictionalized. This decision was intended to help students identify more easily with the application process or a professional environment. Each writing task is introduced with a detailed description of a fictional situation and person. This is intended to ensure the “prerequisite structures (conditiones)” (Isenberg, 1976, translated by WKG), as illustrated in the task examples in the appendix in Giera (2020). The task then references these conditiones through “reference structures” (ibid.) and includes “intention structures” (ibid.), prompting students to write a letter on the PC using the specific criteria. These situated writing tasks require social interaction through written communication between the student, who must empathize with a particular role and context, and the defined and fictitious recipient (Heinemann, 2008: 132). The intention to write a letter of application or an offer is expressed through a clearly defined goal and a direct prompt, formulated as an imperative or request sentence, functioning as an illocutionary act (Heinemann, 2008: 135). In the intervention groups, the students will first write a pre-text for each text type before the writing training to determine their prior knowledge of this text type. During the intervention phase, the students revise their pre-text (practice text) and compose a post-text upon completion of the intervention. The comparison of the text quality at several measurement points is intended to show whether the text quality has improved as a result of the intervention. If the writing task is regarded as an item to measure text quality, the students’ written output represents an open response format. The decision to use two different measurement methods for assessing student texts is based on the advantage of combining holistic and analytical assessments, which enables the evaluation of texts from two distinct perspectives - holisti‐ cally from above, like a bird’s-eye view, and analytically from below, like a frog’s perspective. The holistic assessment procedure may harbour measurement er‐ rors, which can be mitigated by employing multiple raters per text: “In principle, however, the same judgment given twice also stabilizes a measured value; 154 3 Research design & methods <?page no="155"?> average judgments are more reliable than individual judgments” (Neumann, 2017: 210). This mixed approach allows test results to be used for “everyday teach‐ ing/ learning of writers” (Neumann, 2017: 211, translated by WKG). This proce‐ dure is based on established studies including the NAEP studies in the USA, the Hamburg Essay Study (Lehmann, 1992), and the DESI and VERA studies from 2007, 2009, 2011 and 2015 (IQB Berlin, 2015). This evaluation procedure forms part of the qualitative content analysis and directly relates to the previously formulated research questions and hypotheses. In all survey classes, pupils complete questionnaires at various measure‐ ment points. These questionnaires primarily consist of closed-ended items and open-ended supplementary tasks. The questionnaires were created by the students themselves, as no suitable pre-existing instruments were available for this topic. Questionnaire 1 (“Your profile”) is based in part on the student questionnaires by Neumann and Matthiesen (2011: 73 ff.) and Graham, Harris, and Mason (2005: 224). Questionnaires 2 (“What do you know about the cover letter? ”) and 3 (“What do you know about the unsolicited offer/ advertising letter? ”) are adapted from the items on narrative texts by Sawyer, Graham, and Harris (1992: 343). Questionnaire 4 (“Your view of writing”) draws on the items from Graham, Harris, and Mason (2005: 225). All questionnaires are self-report instruments, providing insight from the student’s perspective. During the piloting phase, all questionnaires were tested for comprehensibility, feasibility of evaluation and completeness. To avoid the tendency to answer in the middle, most items predominantly use rating scales with four values. The response options are primarily semantic, ranging from “strongly disagree” to “strongly agree” were selected as rating scales (Bühner, 2011: 113). Only a few responses are dichotomous with “yes” or “no” or include numerical ranges and open response fields (Grabowski, 2011). • Questionnaire 1: “Your profile” is completed once by all students before the first pre-text. It contains questions about general background characteristics of the test subjects, such as age, gender, and place of birth, as well as questions regarding their speaking, writing, and text type skills. The item formats are partly bound (checkbox tasks) or partly open-ended (see appen‐ dix in Giera, 2020). This questionnaire is administered only once, before the intervention phase, to all students. Due to the predominant number of open-ended items, it spans three pages and includes questions about general personal information (six items), language skills (six items), writing habits (three items), knowledge of text types (two items) and career aspirations (one item), amounting to a total of 18 items. 3.3 Test instruments 155 <?page no="156"?> • Questionnaires 2 and 3: “What do you know about the cover letter? ” and “What do you know about the unsolicited offer (advertising letter)? ” are designed to assess students’ prior practical knowledge and self-assessment of respective text types. For the students, these questionnaires serve as a tool for self-reflection on their understanding of the application letter and offer letter. In the intervention groups, both questionnaires are used again after the post-test, to help students gain more practical knowledge through writing during the intervention and to increase their writing attitude (self-efficacy expectation) towards the text types taught. Each questionnaire consists of one A4 page and is administered twice, once in the pretest phase and once in the posttest phase, for all participants. In the first part, students respond to questions on prior practical knowledge, which include four linked item formats. The second part includes eight self-assessment items related to the specific text type, each using a four-point rating scale. Each questionnaire contains a total of 12 items. • Questionnaire 4: “Your view of writing” is designed to encourage all students to assess their personal writing behavior. The intervention groups complete this questionnaire before and after the intervention. Since the intervention involves training in computer-based writing, text planning, and text revision, it is assumed that students will rate these three areas more positively after the training. It also contains questions related to writing self-efficacy and is used in the pretest and posttest phase for all test subjects. The intervention groups also complete this test in the middle of the writing project. The questionnaire contains 21 question items with fixed response formats using a four-point rating scale. It is approxiamtely 1.5 A4 pages long and covers the following areas: general self-assessment of writing (three items), choice of writing media (three items), writing process planning (six items), and writing process revision (nine items). • Questionnaire 5: “Feedback” is used to check the quality of the writing project. At the end of the intervention, students in the intervention groups can give feedback on the writing project. This feedback deals primarily with the selection of text types, methods, and media during the intervention. It serves as a tool for critically reflecting on the design and implementation of the project from the learners’ perspective and contributes to quality assurance. The questionnaire is used only once by participants of the intervention. It is approxiamtely 1.5 pages long and includes items on the following areas: text type feedback (three items), feedback methods (four items), feedback media (three items), overall feedback (three items) and open 156 3 Research design & methods <?page no="157"?> feedback (one item). Except for the last item, the other response questions use a four-point rating scale. • Questionnaire 6: “Questionnaire on the writing task” is used immediately after the pre-test and post-test writing tasks. It asks students whether the writing task motivated them or whether they enjoyed working on it. It contains two items, both using a four-point rating scale. The sheet is just under half an A4 page long. There are 14 items. In the intervention classes, the feedback discussions, conducted once during each intervention phase, are audio-recorded using a dictation machine, and distinctive paraphrases are then coded and evaluated. It intends to better understand the students’ negotiation of the revision steps that are not directly visible in the finished text. The analyzed data can explain performance effects, where applicable. The students work in small peer groups consisting of three to four students. The discussion is initiated by the pupil who presents their text to the peer group and reads it aloud (Heinemann, 2008: 127). Meanwhile, other classmates receive a copy of this text to read along (see Section 3.4). Next, the checklists covering formal, content-related, and linguistic aspects are distributed among the group (see appendix in Giera, 2020). Each group member give feedback to the reader in the form of a “reaction” (ibid.), assessing specific aspects (e.g., whether a quoted price is present, partially present, or missing). This is not only communicated to the writer verbally, but also recorded on checklists, allowing the writer to decide which feedback to accept in order to improve their text in preparation for the pretest task as an exercise. The peer group may respond to each other’s “ratification” (ibid.) as long as the discussion does not turn into a prolonged debate that interrupts the feedback procedure. The primary goal should be the oral and written assessment of the student’s text by their peers. During data preparation and evaluation, the following aspects are recorded, the length of the conversation, the number of participants, and the content of the conversation. The listener must reconstruct the intention of a student’s feedback, possible by using the three checklists. However, the acceptance of feedback by the writer is not expected or recorded (Wunderlich, 1976: 13). The feedback is provided in terms of factual content through “representativa (e.g., statements, assertions, descriptions)” (Wunderlich, 1976: 77, translated by WKG), reflecting the peer groups’ point of view. However, the writer may perceive the feedback on a relational or appeal level negatively as “directives (e.g., orders, requests, instructions)” (ibid.). This perception can influence how, and whether, the writer revises their text. It is assumed that students should be used to feedback discussions from other subjects beyond the formal institutional 3.3 Test instruments 157 <?page no="158"?> interaction at school, which is also required in the curriculum. Furthermore, the students are expected to have prior knowledge of feedback procedures, including the rules and values that should be taken into account during feedback discussions (Parsons, 1951; Heinemann, 2008: 133). In addition, individual writing processes of pupils in the respective classes are recorded using the “Inputlog” software (Leijten & Van Waes, 2013). This makes it possible to identify revision strategies that are expedient in the writing project. Ultimately, the students should be encouraged to revise their texts in terms of language, structure, and content. Exemplary video recordings of the teacher’s modeling phases are also produced to document this process. Linnemann (2017: 335 ff.) systematizes the use of test instruments for record‐ ing the writing process according to the categorization by Janssen, Van Waes, and van den Bergh (1996) as illustrated in Figure 36. In summary, the writing products, the student texts, are recorded asynchronously and indirectly. They provide less detailed insights because the writing process, i.e., how the student arrived at the text, is not mapped. The questionnaires, on the other hand, are asynchronous and direct test instruments, as they inquire explicit reflections on the writing process after the text is completed. The input log recordings, however, are synchronous and indirect, as the writer has no verbal influence on this data. Nevertheless, the level of detail is high, as every keystroke, and therefore the entire writing process is recorded. 158 3 Research design & methods <?page no="159"?> Fig. 36: Methods of writing process research as a function of synchronicity, directness and detail (after Linnemann, 2017: 337, translated by WKG) The video recordings during the modeling phase conducted by the teacher are also synchronous with the writing process and are direct and highly detailed. The text is created step by step in real time and the students can perceive it visually on the projector. Additionally, the teacher uses think-aloud strategy, incorporating student comments into the resulting text. This leads to a comprehensive process protocol that captures both instructional modeling and collaborative input. The audio recordings of peer feedback discussions, although created after the first version of the student text has been produced, also form an important part of the writing process, as they provide recommendations for revision that can be incorporated by the student in the upcoming writing tasks. They are not part of dialogic product creation, as the text is not changed in the feedback discussion. Therefore, the classification should be asynchronous. Although the writer is present during the feedback session, the information comes from the classmates and should be classified as indirect. This is extended product data that can, but do not have to, lead to a revision and is highly detailed due to the amount of information from the feedback session. The diverse selection of test instruments in this research project, varying in synchronicity, directness and level of detail, allows 3.3 Test instruments 159 <?page no="160"?> 12 As part of a student project, formal, content-related and linguistic criteria lists were developed as checklists for writing term papers in German, which led to a successful revision of the term papers (Giera, Mahler & Neumann, 2013). The form of these lists was retained for the writing project, but adapted to the new lesson content. for a valid mapping of the writing process, as relying on a single method would be too one-sided (Glaser & Meyer, 2017: 377). 3.4 PROCEDURE OF THE WRITING PROJECT When investigating the effectiveness of writing lessons in schools, it is essential to provide a precise description of the instructional phases. This not only allows for a scientifically valid test-retest reliability analysis but also enables the pedagogical replication of the teaching approach in classroom practice (Fidalgo et al., 2017: 5). The following subchapter outlines the structure of these instructional phases. The background knowledge should be activated by the pretest, the question‐ naire “What do I know about the offer/ application letter? ”, a short information text, and sample letters. Sample letters are placed unsorted on the classroom floor, prompting students to collaboratively sort and rank them in small groups. In this phase, the students are encouraged to develop criteria and the corresponding criteria lists, 12 which represent the content, linguistic, and formal requirements of the respective text type. Knowledge of text types and text patterns ultimately results from linguistic, formal and, content-related text experience. Heinemann (2008: 143) adds that these text patterns also require text formulations, lexical specifications, and syntactic patterns as conventions and norms. Writer can use these conventions as a framework or choose to deviate from them in specific contexts. The students also recognize that every sample letter has strengths, yet that none of them is without flaws. The aim is not to identify a “perfect” model, but to develop an awareness of what constitutes a strong text through ranking the samples and peer feedback discussions. Only in the discussion phase, where students deliberate about the content, formal, and linguistic criteria/ aspects, can decisions be made as to which text is considered better or worse by the student teams. For example, in the case of a cover letter, relevant content criteria include: How did you hear about the company (reference to the job advertisement)? , What are the reasons/ motivation behind the application? (motivation), or What is the purpose of the letter? (all checklists can be found in the appendix in Giera, 2020). The formal and linguistic checklists apply equally to both the cover letter and offer letter. Examples of the formal aspects include the overall impression of 160 3 Research design & methods <?page no="161"?> the letter or the recipient’s address. The formal criteria are based on the spec‐ ifications provided by the German Institute for Standardization (Grün, 2013). The DIN Standards, particularly those governing “Letterhead lettering” (Grün, 2013: 7 ff., translated by WKG), and business letters, offer precise guidelines, e.g., the use of italics, bold fonts, paragraphs, and punctuation that extend beyond the comma and punctuation marks. For the linguistic aspect, the checklist includes criteria such as ‘polite forms of address’ and an ‘appropriate style of language (factual, formal)’. While DIN 5008 does not specify any linguistic standards for this, Grün’s (2013) DIN publication provides numerous sample texts and usage examples. The criteria lists play a significant role not only in planning and revising student texts but also in peer feedback sessions. Therefore, these lists are then discussed collectively by the learning group for clarity, comprehensibility and acceptance (ibid.). During the discussion phase, the personal attitude towards writing, collected before the intervention lessons using the questionnaires Deine Sicht des Schrei‐ bens and Dein Steckbrief, is also addressed in the form of a class discussion. To deepen student interest in the criteria lists, two students are assigned one criterion each, printed on individual slips of paper. Working in tandems, they are tasked with looking for a suitable sentence in the sample letters and writing it down legibly for the class on the piece of paper. The enlarged and cut apart criteria lists are then reassembled in class through this “criteria puzzle” activity, where each tandem contributes their example to the reconstructed list. These lists remain displayed in the classroom for continued reference, practice and memorization. The modeling phase is carried out by the teacher, who plans, writes, and revises a text collectively with the students. The teacher’s thinking aloud method gives students insights into how writing can be exemplified using the criteria lists, and the integration of planning and revision phases (Philipp, 2012: 153). To achieve this, the teacher sits in front of the class with a projector and laptop and composes both an offer and an application letter on-site, making the process visible and participatory. The mnemonic techniques such as the “KaWa” (Birkenbihl, 2010: 54, translated by WKG) and the “ABC list” (ibid.) are used beforehand for the respective text type to recollect prior knowledge and highlight key aspects of the respective text type. First, the teacher plans their writing process out loud and with the help of key points on the PC. They then pick up the task and try to weave many of the specified details into the letter. This is followed by writing in sentences and, at the same time, formatting and making the first linguistic revisions. The students can make recommendations or additions and are thus involved in the modeling process, even if the teacher 3.4 Procedure of the writing project 161 <?page no="162"?> 13 In this work, KaWas are equated with the terms acronym, mnemonics and mnemos. remains in charge. The writingprocess is deliberately interrupted and continued a week later to show that delayed revisions are useful, and that the teacher also makes mistakes, makes corrections, and revises linguistic, content-related, and formative aspects. Although the teacher is considered an expert in the eyes of the students, the phases of planning, writing, and revising should be made transparent so that they can use this for their own writing process. The strategies and (text type) specific criteria are memorized using the mnemonics “KaWas” 13 (ibid.) and “ABC lists” (Birkenbihl, 2003: 12). KaWas are word pictures created using the ABC list method, where alphabetical terms are written down associatively for a topic (Birkenbihl, 2010: 266). Birkenbihl developed these two learning techniques on the basis of Ellen J. Langer’s research findings. In this context, facts are not presented rigidly, which would lead to forgetting, but are reinforced by the new category formation. The discussion about the criteria, in particular, challenges existing norms and conventions and while simultaneously fostering learning (Langer, 1989: 63, 120). The use of mnemonics has proven effective not only in SRSD studies, but also in other writing interventions (Fidalgo et al., 2012: 77). Both mnemonics have the advantage that they can be applied to a wide range of learning contexts and thus to any genre of text (see Fig. 37). Fig. 37: KaWa example (own illustration) 162 3 Research design & methods <?page no="163"?> The entire process of revising the text is classified as support and practice and is guided by the teacher in the writing project through the concept of scaffolding. The term scaffolding was originally coined by the research team Wood, Bruner, and Ross (1976), who investigated the problem-solving abilities of three-to five-year-old children, as they constructed a pyramid during an experiment. In the context of classroom teaching, the pupil receives a framework of step-by-step support that is hierarchically aligned with the content being taught (Fürstenau, 2009: 70; Wellenreuther, 2009: 77; Gibbons, 2009). Furthermore, in this phase, the peer groups also read aloud their own texts, which were written in the pretest, and thus present them for the first time. The students act as lay reviewers, guided by their intuitive and prototypical understanding. During the discussions, the peer group marks, while the authors are present, whether the individual characteristics under consideration were “fully,” “partially,” or “not at all” achieved. It is also possible to include brief comments justifying these decisions in a fifth column. The aim is not to fulfill every criterion or aspect, but to use feedback as a guide to ultimately write an improved version of the letter. Reading one’s own text aloud is also a situation of presentation and appreciation, as the written letter encounters a circle of addressees for the first time. Using the criteria lists during the feedback discussions has the advantage of ensuring that feedback is given on a factual level. This helps students understand that it is not they as individuals but their written text that is being assessed. The students are made aware of this as part of the writing project, where each student receives feedback twice and gives feedback on at least four other texts, since a peer group should consist of at least three members. By reading and evaluating their peers’ texts, students gain ideas for improving their own texts. Furthermore, they implicitly internalize the criteria from the three checklists for each text type. Feedback discussions within peer groups can also form the basis for writing motivation. Students may recognize that some criteria are not completely fulfilled, but they will also receive positive feedback on aspects that do not require revision. This should also strengthen their overall attitude towards writing. The other members in the peer group then provide feedback on the text using the three checklists. This text revision procedure closely resembles existing, evaluated techniques, such as the writing conference by Spitta (1992: 13) or the models by Böttcher and Wagner (1992: 26). The formal, content-related, and linguistic checklists are completed in writing in the presence of the speaker on a three-point scale, “fully achieved”, “partially achieved”, and “not achieved”. Comments can be added either in writing or orally. 3.4 Procedure of the writing project 163 <?page no="164"?> The reason for this structure is to give the writer an opportunity to review the completed checklists and independently decide which aspects they would like to revise. The goal is to improve the overall quality of the text with the help of content, linguistic and formal criteria. By giving and receiving feedback, students become familiar with revision strategies and procedures, thereby developing their abiltiy to revise in the sense of Hayes and Flower’s (1980) writing process model. Even if writing letters presents a challenge for many students, they will also receive positive feedback. In this way, they can access what Vygotsky described as the “zone of proximal development” (Vygotsky, 1987, translated by WKG). In the post-test, a new writing task is assigned for the text type. After each intervention lesson, the students also write an entry in their learning diary. The aim of the learning diary is to encourage students to reflect on themselves, review the lessons and also give feedback to the teacher. It has been proven that learning diaries are relevant for academic success, as self-reflection on the learning steps taken so far supports progress toward the next ones. When discussing the second type of text, the offer, the steps are repeated in the same order as described earlier. Through the writing of application letters and unsolicited offers required in the pretest and posttest, the students are required to plan, write, and revise their writing process independently (Kessler & Ziener, 2004: 16 ff.). 3.5 DATA COLLECTION The data was collected during the writing project in the German lessons of the respective schools and classes. Written permission for the data collection was obtained from the school authorities in Lüneburg, as well as from the head teachers and all parents or guardians. Pupils who were of legal age at the time of the study were allowed to decide independently whether they wanted to participate. The figure Overview of the use of the test instruments (see Section 3.3) demonstrates when and how often the individual test instruments were used and by whom. Responsibility for conducting the survey lay with the project implementers. All test instruments are included in the appendix (see Giera, 2020). A total of four writing tasks were used to collect the student texts. The cover letter and job offer pretests were collected before the intervention, and the posttests afterward, with the job application posttest being written by the test subjects in the middle of the project, following the first part of the intervention. 164 3 Research design & methods <?page no="165"?> Only then was the offer content addressed in the German lessons. Apart from the PC, the students had no other aids when writing the texts in these survey moments. The questionnaires (n = 6) were generally administered after the writing tasks. The unsolicited offer followed the survey of the application letter and the associated questionnaires 1 and 2. After writing the first unsolicited offer, the students were asked to complete questionnaires 3 and 4. This sequence was chosen because the questionnaires enabled students to reflect on the experience of writing the respective tasks, while the writing behavior was still fresh in their minds. This form of questionnaire-based inquiry constitutes “asynchronous data collection” (Weinzierl & Wrobel, 2017: 221, translated by WKG), as it was not conducted during text writing and represents an “indirect observation of objectifiable process characteristics” (ibid.). In total, the six questionnaires were used several times, so that there were eleven questionnaire surveys, leading to the evaluation of 1269 questionnaires. All text products were written by the students on the computer in the Windows Word program. The collected questionnaires were assigned the same code as the text products. A total of 548 student texts were collected for the evaluation. Audio recordings were made during the feedback discussions in the German lessons. Each group was provided with a dictation machine in the form of a USB recorder. The groups sometimes changed during the intervention phases, as the pupils worked at different paces or preferred working with different peers. These conversations were then transferred to an external hard disk and encoded. In total, over nine hours of audio recordings were examined (09: 12: 13 h). The videos of the teacher’s writing processes were recorded by a pupil and subsequently saved on an external hard disk by the project manager. Although this recording had been planned, it was initially intended for documentation purposes rather than for data analysis. However, during the course of the data analysis, it became clear that the videos demonstrated writing processes across the various school types that were useful for interpreting the results. There are 76 minutes of video material available for the analysis. The intervention groups wrote their texts using the “Inputlog” software (Leijten & Van Waes, 2013), which ran in the background on the PC. Students were required to log in at the beginning of the writing process and log out again at the end. At the end of the writing project, the software data was copied from the school servers to an external hard disk and deleted from the local systems. The keylogging program thus serves as an instrument for indirect observation of objectifiable process characteristics and represents a form of synchronous data 3.5 Data collection 165 <?page no="166"?> 14 In studies, the definition of pause length varies between one second, two and five seconds, depending on the research question. See the studies by Linnemann (2014: 121 f.) and Van Waes/ Leijten & Weijen (2009: 44). collection during text production. The variables collected by Inputlog include pause length, number of pauses, general and active writing time, total number of words, number of written words and bursts. In this study, a pause is defined as an interruption in writing process lasting five seconds or more. 14 This definition excludes short pauses within individual word formation and focuses on longer interruptions (from five seconds), such as those occurring between sentences and paragraphs, that may indicate (meta-)cognitive planning. Pauses are triggered by different cognitive processes: One possible interpretation of this pattern of findings indicates that more complex planning processes (e.g., at the level of the discourse target) are associated with longer pauses between larger linguistic units and that local, less complex processes (e.g., checking spelling) are indexed by shorter pauses within words (Weinzierl & Wrobel, 2017: 227 f., translated by WKG). “Bursts” (Chenoweth & Hayes, 2001) indicate writing fluency: A burst is defined as the number of words that are written down between two breaks in writing. This refers to phases of the writing process in which words are written down one after the other, as if intermittently over time, as coherently planned units (Weinzierl & Wrobel, 2017: 228, translated by WKG). Novice writers produce shorter burst lengths than expert writers (Chenoweth & Hayes, 2001). However, in addition to the bursts, the number of words can also serve as an indicator and benchmark for writing fluency, especially when the writing time is limited (Weinzierl & Wrobel, 2017: 228). The software data for each test subject was sorted using the same code as for the texts and questionnaires. A total of six writing processes and their resulting writing products were recorded. It is not possible to provide with a precise figure for the amount of data collected in this way, as Inputlog also automatically stores other variables. In addition, the project leader documented the lessons in a research diary. Even though these impressions are subjective, they intended to support the interpretation of the results. Subjects who became ill or were absent were completely excluded from the study. In longitudinal studies, it is common that not all subjects are present from the beginning to the end, due to illness, unexcused absences, or dropouts. If 166 3 Research design & methods <?page no="167"?> individual items were missing, the data were marked and handled as missing data (Rost, 2007: 177). 3.6 DATA PREPARATION All primary data, such as the interviews, student texts, and questionnaires, were first converted into secondary data: The student texts were printed out, scanned, and then coded by two independent raters. The modeling phases with the teacher and the students were video-recorded and further processed and coded as tertiary data in the form of transcripts. The questionnaires were scanned and the responses were coded. The feedback interviews with the classmates were recorded and coded using an audio stick. The input log data was provided by the software and transferred to SPSS. The raters were provided with a coding manual for all coding procedures, excerpts of which can be found in the appendix (see Giera, 2020). The rater training procedure is explained in the following subchapter. 3.6.1 COURSE OF THE TEXT RATING TRAINING Rating procedures are particularly common in larger studies to measure pupils’ writing skills. The training of the raters is essential for the transparency of the study and for ensuring reliability (Mayring, 2010: 61). A distinction is made between holistic and analytical ratings (see Chapter 3.1): “The judgment of several raters is in any case, although still subject to measurement error, more stable than an individual judgment” (Neumann, 2014: 65, translated by WKG). The raters were a group of twelve students at Leuphana University studying German, whom I taught as a lecturer. They attended a two-semester research preparatory course as part of the “Writing for and in the workplace” module. Student texts from the pilot study were used for the first coding exercises. In the pilot project, the writing tasks, the questionnaires, and the content of the writing project were tested at a vocational school in the Lüneburg-Wolfsburg district of the Chamber of Industry and Commerce in a vocational school class of hotel management specialists in their second year of training and in a pre-vocational class aiming to obtain a secondary school leaving certificate. A coding manual based on the “Indicator Model of Scholastic Writing (IMOSS)” (Neumann & Matthiesen, 2011: 19 ff., translated by WKG) served as the basis for coding. The individual assessment variables were rated on a five-point scale, 1 = “lowest level” and 5 = “highest level”, by two independent raters. 3.6 Data preparation 167 <?page no="168"?> 1. First, the overall impression of the letters is assessed holistically. Holistic text encodings are normative assessments (Goldin & Ashley, 2012: 211). Further analytical assessments follow. 2. The content design means the following: “Does the text contain the content aspects required by the task? ” (Neumann & Matthiesen, 2011: 19, translated by WKG). This criterion should be fulfilled to motivate recipients to take action. For this to happen, however, the reader must be informed through the knowledge conveyed in the text in the sense of “informativity” (de Beaugrande & Dressler, 1981: 3 f., translated by WKG). The situation in which the reader finds themselves must be considered in the sense of “situationality” (ibid.), and ultimately the “intentionality” (ibid.), i.e., the function of the text for the reader, must become clear. The following content aspects are assessed for the cover letter: reason for the letter, justification of the application position (training position, intern‐ ship, job offer), how and why was the company chosen, self-presentation (e.g., soft skills, strengths, qualities, professional achievements), current activities and desired starting date as well as a request for an interview and a compelling conclusion. For the unsolicited offer, however, the following content aspects are taken into account: reason for the letter, ability to arouse the customer’s/ reader’s interest in buying, alignment with the target group’s needs, presentation of the benefits/ advantages of the product/ service, specific details of the offer including pricing, and an encouraging closing statement that prompts the customer to buy. 3. In the text structure category, the external structure of the letter is assessed in terms of an introduction, main body, and conclusion. 4. The category theme development/ coherence assesses how the theme introduced in the task unfolds in the letter. Coherence reflects the seman‐ tic-thematic quality of a text in its deeper structure and in the relationship between producers and recipients (Fix, 2008: 22), and is defined as follows: Coherence concerns the functions by which the components of the textual world, i.e., the constellation of concepts (terms) and relations (relationships) underlying the surface text, are mutually accessible and relevant to each other. A concept can be defined as a constellation of knowledge (cognitive content) which can be activated or recalled to consciousness with more or less uniformity and consistency. Relations are the links between concepts that occur together in the textual world; each link should bear a designation of the concept with which it makes a connection […] (de Beaugrande & Dressler, 1981: 5, translated by WKG). 168 3 Research design & methods <?page no="169"?> 5. The category style/ addressee orientation considers the linguistic style of the letter and its orientation towards the addressee. “In defining ‘style’, we assume that the use of certain patterns suggests certain realizations in certain constellations” (Rehbein, 1983: 23). Style is never a single utterance or sentence, but rather a systematic sequence of utterances. There can also be “style conflicts” (Rehbein, 1983: 24). Without a style oriented toward the text as a whole, the function of the text cannot be recognized by the addressee (Fix, 2008: 28). Style therefore conveys the communicative purpose of the text and signals to the addressee what the action and situation are, and who is expected to act (Sandig, 2006: 2 f.). 6. The category word choice/ vocabulary considers “[…] the choice of words and phrases in the text, i.e., how flexibly and variedly the writer can use the German vocabulary” (Neumann & Matthiesen, 2011: 23, translated by WKG). 7. Spelling errors are assessed under the spelling category. In this case, a scale value of 5, the highest rating, indicates that there are no spelling errors. 8. The grammar category soecifically refers to word-level grammar. The following types of errors are taken into account: “conjugation errors, declension errors, tense errors, case errors, mode errors, errors in the prepositional structure” (Neumann & Matthiesen, 2011: 25, translated by WKG). Scale value 1 stands for “meaning-distorting” (ibid.), scale value 2 for “communication-burdening” (ibid.), scale value 3 for “complexity problems” (ibid.), scale value 4 “occasional inconsistencies” (ibid.), and scale value 5 indicates “no grammatical errors” (ibid.). On the surface level of the text, grammatical forms such as the choice of tense can establish the cohesion of a text, according to de Beaugrande and Dressler (1981: 3). This criterion should be applied as a communicative-pragmatic condition following the text-grammatical approach of Gansel and Jürgens (Gansel & Jürgens, 2008: 55): “Texts are not treated as isolated, static objects, but as communicative entities” (Gansel & Jürgens, 2008: 65, translated by WKG). 9. In the final category, sentence construction, the syntactic structure of the sentences in the letters is examined. This criterion is treated separately from the grammar category, as punctuation emphasizes sentence segments and their segment boundaries (Gansel & Jürgens, 2008: 66), which are crucial for the addressees’ understanding of the text. 10. The handwriting category is not applied in this study, as all texts are written on a PC. The exact number of words/ text length is determined precisely for all pupils during the evaluation of the “Inputlog” data (Leijten & Van Waes, 2013), or counted additionally if data is missing. 3.6 Data preparation 169 <?page no="170"?> The raters were initially given all the student texts from the pilot exercise. These texts were coded according to the IMOSS procedure described earlier. The raters then disclosed and reflected on their individual assessments of each student text. Any discrepancies in ratings during the practice phase were quickly recognized and discussed collectively. This procedure led to the creation of benchmark texts for the coding manual, as required for rater training by Kellogg (1994: 56). In the next step, the checklist “Category I: Content” was introduced and discussed, which was relevant for the dichotomous evaluation of the two text types. The raters were trained on each text feature/ aspect so that they could carry out dichotomous evaluations as an exercise using the pilot texts. The number 1 stands for “present” and the number 0 for “absent” with no further gradations. All evaluations of the exercise texts were again reflected upon together in the seminar to ensure consistency in coding. The students also evaluated the feedback discussions with the peer groups. They received the audio recordings for this purpose. They coded, identically to the checklist “Category I: Content”, how the feedback providers assessed the content features in a text of the feedback recipient. The available rating scale ranged from “fully achieved”, and “moderately achieved” to “not achieved”. The checklists “Category II: Formal structure” and “Category III: Language” were also included (see appendix in Giera, 2020). These documented whether the student in question had received help from their peer group and to what extent content, linguistic, and formal text features were fully, partially, or not at all addressed in the letters. In this way, the feedback recipients received verbal and written feedback, which they could use for text revision. For the final coding of the students’ texts and feedback discussions of the writing project, the students were divided into one of three groups: Group A, with four students, coded the students’ texts using the IMOSS method on a five-point scale. Each text was coded twice. The raters received the student texts anonymously and were not permitted to vote. Both pre-, post-, and practice texts were randomly mixed. Group B consisted of three students who double-coded the content, linguistic, and formal features in all student letters anonymously. Group C consisted of two students, responsible for coding the feedback interviews. All groups carried out the coding during the second half of 2014. Counseling sessions were held every two weeks throughout the coding period. Although the rater training was very time-consuming, spanning two semes‐ ters, it was necessary to maintain the objectivity of both text evaluations and analyses of student discussions. While it would have been possible to rate only 170 3 Research design & methods <?page no="171"?> 15 The dichotomous ratings in particular should have been much higher. After closer analysis, the reason for this was the very strict rating behavior of one rater, which was not corrected. A correction would have led to even higher and better results in the area of content learning. a percentage of the student texts double-blind, this approach was not adopted as the sample was not too large at just under 500 texts. The reliability of the ratings was calculated using Cronbach’s reliability coefficient α. A satisfactory value of α = .77 was determined for the IMOSS ratings (exact agreement of the five-point scale) across all scales and α = .67 was calculated for the dichotomous ratings of the content-related text elements. 15 As illustrated, the category system for analyzing the texts and student interviews preceded the data preparation of the test instruments. 3.6.2 DATA PREPARATION OF THE INDIVIDUAL TEST INSTRUMENTS The data from the writing products, including the variables of word count/ text length, text quality, absolute number of content features of the respective text type, feedback sheets, the keylog data, video and audio recordings, and the research diary containing subjective perceptions, were prepared for the calculation as follows: ad) Writing products To determine the word count, the “Inputlog” software (Leijten & Van Waes, 2013) was used, as it provided precise measurements. All test subjects were instructed to limit their texts to one page, in line with common conventions for application letters and unsolicited offers. To enable comparability of text lengths, the mean value (M) was first calculated across all test subjects for each text type. This mean value was then categorized as Level B, corresponding to the “regular standard” according to Kessler and Ziener (2004). Text lengths below the mean value range were classified as level A (minimum standard). This approach ensured that the assessment of text length is objective. All student texts were then rated globally on a five-point rating scale by two independent raters. The time of the survey, grade level and pupils were concealed from the raters to prevent rating bias. Although this assessment was interpretative in nature, it followed a coding manual that included benchmark texts derived from educational policy documents. Level 1 represented a very low level of writing, and level 5 the highest level of writing. This scale was used in several studies, e.g., including the IMOSS study (Neumann & Matthiesen, 3.6 Data preparation 171 <?page no="172"?> 16 The coding variables of the Formal, Language and Content (cover letter) checklists were checked by six experts in January 2018 (follow-up survey) to verify that they were up to date (see appendix, Giera, 2020). 2011), and is based on predefined criterion-referenced assessment norms. The competence levels were defined as follows (Giera, 2014: 7): Levels 1 and 2 represent Level A, minimum standard not met (for Level 1) and just met (for Level 2) respectively. Levels 3 and 4 represent the regular standard and competence level B. As level C, the expert standard, is rarely achieved, only level 5 was used in the overall assessment for the texts. In this coding procedure, the overall impression of a text is influenced by factors such as spelling, grammar, syntax, and style, which is precisely why these linguistic-pragmatic and systematic areas were assessed separately using the same five-point scale by two independent raters. This assessment required the raters to read through the texts repeatedly to ensure accuracy. The checklists used in class functioned both as peer group assessment tools and as a basis for rater evaluation. 16 The content-related text features of the respective text type were coded dichotomously with 0 for “not present” and 1 for “present” by two independent raters. The absolute frequency of each content feature was determined for every student text. This procedure can be used to determine which content features, and instructional elements from the writing project, were particularly taken into account by the students when writing and which were not. ad) Feedback form The questionnaire data were collected through a combination of open-ended and closed-response items. The feedback forms contained both types of questions, with a predominant use of multi-level ordinal scales for closed responses. A four-point rating scale was used for the ordinal measurement level (1 = strongly disagree to 4 = strongly agree). One level more or less would have led to a response behavior with a ‘tendency towards the middle’, so a four-level scale was used. Average (mean) values, including half-steps, were generated by combining the ratings from two evaluators. In the case of closed questions, the answers could be recorded dichotomously. If the questions were open-ended, the answers first required to be reviewed and then coded in terms of their content. It was important to consider that while some answers did not occur with high frequency, they often displayed semantic similarities. The questionnaires were coded accordingly by an external rater, following a coding manual. Only open-ended responses required to be categorized by the coding process and integrated into the coding manual. 172 3 Research design & methods <?page no="173"?> ad) Keylog data The inputlog data had to be retrieved for each individual student and corre‐ sponding written text. The program only outputs a general summary logging file. After narrowing down the values, only selected variables of the software were used for data preparation, which are listed below: • Part I Process values (“Process Information”) - Number of words (“Words Total”) - Number of characters, incl. spaces (“Characters Total typed incl. spaces”) - Number of characters, incl. spaces per minute (“Characters Per Minute incl. Spaces”) • Part II Product values (“Product Information”) - Number of words (“Words Total”) - Number of characters, incl. spaces (“Characters Total typed incl. spaces”) - Number of characters, incl. spaces per minute (“Characters Per Minute incl. Spaces”) • Part III Product-Process Relationship - Product/ process produced ratio incl. spaces (“Product/ Process Pro‐ duced Ratio (incl. - Spaces”) • Part IV Process Time - General time (“General”) - Number of P-bursts (“Number of P-bursts”) - Total pause time (“Total Pause Time”) - Number of breaks (“Number of Pauses”) - Average pause time (“Mean Pause time”) - Active writing time (“Active Writing Time”) ad) Video recordings of model writing with the teacher and the class The transcripts of the teaching videos were coded dichotomously. Each inter‐ view segment was first assigned to the interview phase (introduction, main phase, or conclusion). This was followed by categorizing structural features into gap, overlap, or interruption. Finally, the content was coded as questions, answers, suggestions, or acceptance, and each segment was assigned to different speakers. The general duration of the conversation was also recorded. The content, formal, and linguistic aspects that were considered in the respective conversation segments were then coded. The number above indicates the quantity of content-related text elements during the modeling. 3.6 Data preparation 173 <?page no="174"?> ad) Audio recordings during the feedback sessions Two raters listened to the students’ feedback interviews and coded each interview using a standardized content checklist. All features mentioned in the conversation were coded with 1 = “present”. If features were not addressed during the conversation, they were rated as 0 = “not present”. ad) Subjective perceptions of the teacher The notes on the individual lessons were first sorted chronologically and according to the respective phase of the writing project and compared with all learning groups. This was followed by a compilation as in a research diary, whereby special moments in the respective classes were highlighted. 3.7 DATA EVALUATION IN THE SUB-AREAS The data was predominantly prepared using a “deductive category definition” (Mayring, 2010: 83, translated by WKG). Only the open response options in the questionnaire required to follow an “inductive category definition” (ibid.). It should be emphasized that by using the same evaluation categories with data from several test instruments, ensures both comparability and cross-referenc‐ ing. The checklists for content, formal, and linguistic criteria were used for the evaluation of the student texts, student interviews, and the transcripts of the modeling phases. This approach follows a “structuring content analysis” (Mayring, 2010: 93, translated by WKG), as the categories were defined before‐ hand, as Mayring explains: Formal structuring: The basic idea here is that certain components are filtered out of the text according to a formal, previously defined criterion. Codes are now assigned according to a syntactic, thematic, semantic or dialogic criterion, as required by the main question. The coding rules for this are defined in the code commentary or memo (Mayring, 2010: 114, translated by WKG). Descriptive evaluations, as the first step of evaluation, form the basis for “the extent to which statements can be made at all with the collected and coded information” (Neumann, 2014: 65). Depending on the research question and hypothesis formation, the data was further calculated in addition to the descriptive statistics, as briefly explained below (Rost, 2007): • ‘Standard deviation’ (also called ‘dispersion’, abbreviated ‘S’ or ‘s’, in journals also ‘SD’) is a statistical parameter that provides information about the variability of measured values” (Rost, 2007: 48, translated by WKG). 174 3 Research design & methods <?page no="175"?> • The ANOVA analysis (analysis of variance) is used to analyze the mean values of the different subject groups (subgroups). The variance measures the dispersion around the mean value. The analysis of variance thus shows which groups differ from each other and how large the difference is in the population. It also tests whether the mean values also differ significantly in the population. The analysis of variance (ANOVA) is probably the most frequently used statistical test method in experimental research, not least because it reacts surprisingly insensitively to certain violations of prerequisites such as deviation from the normal distribution and variance heterogeneity, even in combination, as long as the same cell frequencies are present (Rost, 2007: 188, translated by WKG). This calculation was carried out for all mean value comparisons. The T-test, which was also used, can be applied to independent samples, two groups of cases, or dependent samples: Statistical tests are described as robust if some of these requirements can be neglected without the validity of the results having to be seriously questioned. The widely used T-test (used to examine a difference in means between two [independent or dependent] groups for statistical significance […]) and the analysis of variance […The T-test is a special case of the analysis of variance) are, as simulation studies […] have shown, extremely robust methods in which not too dramatic violations of some prerequisites (normal distribution, variance homogeneity) hardly affect the α-error (Rost, 2007: 188, translated by WKG). Depending on how the hypothesis was formulated, a T-test was carried out either with dependent or independent samples. • A correlation indicates whether two variables behave in the same direction. The correlation coefficient is measured, which should ideally be close to 1, and it examines whether a correlation also exists in the population. The test, therefore, investigates the hypothesis that the variables are related within the population. “The correlation is a measure for quantifying the closeness of the relationship between two characteristics” (Rost, 2007: 19, translated by WKG). The chi-square and Kendall’s tau calculation belong to the group of non-parametric methods (Rost, 2007: 188). They are suitable statistical methods for analyzing the freedom and independence of distribution, especially for smaller samples (Rost, 2007: 189). Primarily, correlation calculations were carried out with Kendall’s tau, which examine the lower performance level of some results in more detail. These refer to below-average performance compared to the mean value of the test subjects. 3.7 Data evaluation in the sub-areas 175 <?page no="176"?> The Pearson chi-square calculation was used for the cross-tabulations with four fields when more than 60 cases were tested. • Regression calculations are carried out to test the quality of the model and the significance of the results. The aim was to examine several variables simultaneously in order to calculate an estimated value for the dependent variable. The overall impression of the texts was often chosen as the dependent variable. This type of calculation provides an explanation of the differences in variance due to dependent variables as R ² . An R ² value of 0 shows no correlation; 1 indicates a perfect correlation. Values below 0.3 are considered weak, while values above 0.8 are considered very strong. Regression analyses were repeatedly carried out in all four areas to examine how the dependent variable, overall impression (B001) or content (C001), was influenced by other variables. Regressions are useful for making predictions and for further developing the writing model. • Significance levels are used to discuss “cost-benefit considerations” (Rost, 2007: 210, translated by WKG). Significance tests for statistical hypothesis testing were carried out for all four areas of investigation. “Due to its dependence on the number of test subjects, statistical significance provides no information about the size of the relevance of an effect” (Rost, 2007: 212, translated by WKG). Significance tests, which are two-sided, mean that the two variables in the population do not correlate with each other. The value 0.000, for example, indicates that a correlation does exist in the population: If the researcher concludes on the basis of his statistical analyses that the observed differences between his investigated groups have only a low probability of chance (the calculated p-value is p = 0.048 or perhaps p = 0.009, i.e. it is less than 5 % or perhaps less than 1 %), he has sufficient reason to reject his null hypothesis H0 […] The smaller the sample is chosen, the more the probability of falling into the β-error trap increases (Rost, 2007: 83, translated by WKG). The β-error is the second type of error and indicates that the null hypothesis is incorrectly not rejected (Rost, 2007: 83). In the area of Affect, the variable Attitude towards the task was first calculated as a mean value based on two items from questionnaire 6 (see Section 3.3). This was followed by ANOVA, significance, and T-test calculations for this area. As additional variables, items from questionnaire 4 “I write good letters” and “I like writing on the PC” were examined in greater detail and calculated similarly to the first variable. Correlation calculations were carried out to determine correlations with the overall impression (B001) and the writing attitude. To do this, however, it was necessary to determine the values that were below 176 3 Research design & methods <?page no="177"?> the calculated mean value, which constituted the low-level range. Finally, a regression analysis was carried out to test whether the overall impression also predicts the writing attitude. In the Behavior section of the study, the mean value of the overall impression (B001) was first determined and compared using T-tests, significance tests and ANOVA calculations. This was followed by the language system and language pragmatics scales as the second and third variables. The fourth variable was the number of words, a product-based measure. The mean value was calculated for all four variables. Values falling below the average were classified as the low level group. These values were then tested in relation to each other using Kendall’s tau correlation calculation. Finally, the regression calculation was carried out for the overall impression (B001) as the dependent variable and other calculated values such as the number of words as inclusion variables. In the area of Content Learning, the mean value for content (C001) was first compared across survey dates and subgroups. The values in the low level range were also identified for this variable. Scales were derived from questionnaires 2 and 3, which depict the mean value for knowledge of the respective text type. Low level ranges were also analyzed from these mean values. The fourth variable included the values from the dichotomous ratings of the content aspects contained in the letters. These ratings were used to calculate a mean value And the corresponding low level ranges. The variables presented were tested for correlation in the form of significance tests, T-tests, ANOVA, and Kendall’s tau. The dependent variable overall impression (B001) and the number of words were included in this calculation. A regression analysis concluded this area of investigation. The feedback interviews were only analyzed descriptively. In the area of Metacognition, the relevant items from questionnaire 4 were first calculated to produce a mean value. These items addressed the sub-activities of revising and planning, which are relevant variables for this area of investigation. Correlations with the variables number of words and overall impression were then checked using cross-tabulations. A regression analysis was also carried out here. The modeling videos were only evaluated descriptively as a supplement. The keylog data, such as the number of words/ process, word proportion, general writing time, active writing time, number of pauses, and pause length, were determined as mean values and tested for significance. A T-test was used to determine the differences in the mean values between the pre-test and post-test. Furthermore, the average number of pauses and the length of the pauses for the writing processes examined were also evaluated across five writing phases (Linnemann, 2014: 138), aiming to 3.7 Data evaluation in the sub-areas 177 <?page no="178"?> provide a descriptive overview of all writing phases of the four writing tasks, complementing the questionnaire survey for the last area of investigation. 178 3 Research design & methods <?page no="179"?> 4 RESULTS “The routine adherence to traditional levels of significance cannot be justified by anything, but by nothing at all, certainly not by the stupid remark ‘everyone does it that way’, and is evidence of wonderfully consummate thoughtlessness” (Rost, 2007: 210, translated by WKG). In this chapter, the results of the study areas Affect, Behavior, Content Learning, and Metacognition, are first presented separately; in addition, significance levels above the 5-% threshold are pointed out. 4.1 KEY RESULTS FROM THE AFFECT RESEARCH AREA The Table 1 presents the mean values for the variable attitude towards the writing task (MIN 1, MAX 4) at four measurement points - pretest (FB1) and posttest (FB3) for the cover letter and offer letter writing tasks respectively. The results of the intervention group and the control group are considered separately (see Table 1). At first glance, it is apparent that both groups report an above-average attitude toward the tasks, as they are above the median. They are almost at level 3 on the four-point scale. Further details are: Over the course of the writing project, attitudes towards the task remained stable in the intervention group from M = 2.6 (SD = 0.73) to M = 2.6 (SD = 0.70) for the applications and increased slightly from M = 2.6 (SD = 0.81) to M = 2.7 (SD = 0.98) for the offers when the pretest is compared with the posttest for the respective text type. Although the improvement is modest, it indicates a slight positive shift in attitude. In contrast, the control group showed a decline in attitude: from M = 2.5 (SD = 0.81) to M = 2.4 (SD = 0.86) for the applications and from M = 2.8 (SD = 0.93) to M = 2.2 (SD = 0.98) for the offers. The result shows that there are different mean values between the intervention and control groups, with a 15 % error in the posttest offer. The T-test confirms that the control group exhibited a significantly lower positive attitude toward the writing task. Overall, while the intervention group’s attitude remained stable or improved slightly, the <?page no="180"?> 17 Survey results at the end of the intervention on attitudes towards the task posttest offer (MIN 1/ MAX 4): Intervention group 9th grade M = 2.6 (SD = 0.90; n = 52); control group 9th grade M = 2.2 (SD = 0.94; n = 25); intervention group 2nd year of training M = 2.9 (SD = 0.66; n = 16); control group 2nd year of training M = 2.2 (SD = 1.11; n = 10). control group’s attitude clearly declined over the course of the writing project (see Table 1). - - Pretest - - Postt‐ est - - - M (SD) - - M (SD) - - IG CG Total IG CG Total - (n = 68) (n = 35) (n = 103) (n = 68) (n = 35) (n = 103) - - - - - - - Attitude (Task Ap‐ plication) 2.58 (0.73) 2.54 (0.81) 2.57 (0.75)- 2.63 (0.70) 2.36 (0.86) 2.53 (0.77) Attitude (Task Of‐ fers) 2.61 (0.81) 2.77 (0.93) 2.67 (0.85) 2.72 (0.86) 2.23 (0.98) 2.56 (0.92) Enjoys writing let‐ ters on PC 2.85 (0.98) 3.23 (0.84) 2.98 (0.95) 2.90 (0.95) 2.77 (1.14) 2.85 (1.01) High Quality-letters 2.63 (0.71) 2.60 (0.85) 2.62 (0.76) 2.81 (0.79) 2.66 (0.87) 2.76 (0.82) Tab. 1: Mean values and standard deviations of the items for measuring writing attitude (Affect) Note: MIN 1/ MAX 4 (student survey), IG = intervention group, CG = control group, MIN 1/ MAX 4 (student survey) Further analysis by subgroups (intervention group in ninth year, control group in ninth year, control group of trainees and intervention group of trainees), confirms these findings. In the application pretest, no significant differences in task attitude were discovered between the four subgroups, with an overall average level of M = 2.5 (SD = 0.77). Between the pre-test and post-test for application tasks, the mean scores of the groups were more or less the same, with no significant differences. However, the mean score of the intervention groups in the post-test is slightly higher than that of the control groups. In the offer post-test, slight differences between the intervention groups and the control groups can be observed. 17 It was found that the intervention groups showed a 180 4 Results <?page no="181"?> more positive attitude than the control groups. This result is significant (p < .015) when the mean value of the intervention group is compared with that of the control group. There is also a significant correlation between the variables low level overall impression and low level attitude toward writing task in the pretest application survey, i.e., before the intervention (χ ² (N = 103) = p < .004*). This means that students with a low attitude towards the task also received a low rating for their text (overall impression B001). In the cover letter post-test, however, no correlation was found between the variables Low writing level and attitude towards the task. In contrast, a significant correlation was again confirmed in the offer posttest at the end of the intervention (χ ² (N = 103) = p < .001**). First conclusion: The attitude towards the writing task thus has an influ‐ ence on the overall evaluation of the text, as can be seen from the pretest application and posttest offer. Table 1 shows further results of the survey items, namely I think that I write good quality letters and I like writing letters on the PC, from the View of Writing questionnaire (MIN=1, MAX=4) at two survey points (at the beginning and at the end of the intervention). During the investigation of these two items, the intervention group ultimately shows slight improvements, while the control group develops an increasingly negative attitude towards writing for one item: In the pretest (before the intervention), there is a predominantly above-aver‐ age positive attitude towards writing, which is almost at level 3 out of a possible four. This result improved in the intervention group: for the item I think that I write good quality letters from M = 2.6 (SD = 0.71; see Table 1) to M = 2.8 (SD = 0.79). However, for the item I like writing letters on the PC, the result remained constant at M = 2.9 (SD = 0.98) to M = 2.9 (SD = 0.95). The control group, on the other hand, also showed a slight increase for the item I think I write good quality letters from M = 2.6 (SD = 0.85) to M = 2.7 (SD = 0.87), but showed a decrease for the item I like writing letters on the PC M = 3.2 (SD = 0.84) to M = 2.8 (SD = 1.14) (see Table 1). Second conclusion: Thus, these descriptive calculations show only very small effects in favor of the intervention group in the area of Affect/ Attitude towards letter writing, with no significant differences. 4.1 Key results from the affect research area 181 <?page no="182"?> Table 2 below first presents the results of the holistic rating for the overall impression of the applications and offers in the intervention and control groups (MIN=1, MAX=5): - Pretest Posttest - M (SD) M (SD) - IG CG IG CG - (n = 68) (n = 35) (n = 68) (n = 35) Overall impression (Applica‐ tion) 2.9 (0.80) 2.9 (0.69) 2.8 (0.76) 2.9 (0.64) Overall impression (Offer) 2.6 (0.87) 2.5 (0.83) 2.9 (0.82)** 2.0 (0.87) Tab. 2: Mean values and standard deviations of the items for measuring text quality (Behavior) Note: MIN 1/ MAX 5 (holistic coding of student texts), ** (p < .000), IG = intervention group, CG = control group Overall, the assessments are in the middle range. The results for the control group’s applications remained the same across the two measurement points (from M = 2.9/ SD = 0.69 to M = 2.9/ SD = 0.64), but a clear decrease is observed in the offers (from M = 2.5/ SD = 0.83 to M = 2.0/ SD = 0.87). The posttest of the offers showed that just under 60 % of the control group write offers at a lower level. The intervention group initially achieved the same level of writing performance in the applications (from M = 2.8/ SD = 0.80 to M = 2.8/ SD = 0.76), but there was a clear improvement in the holistic text assessment in the offers (from M = 2.6/ SD = 0.87 to M = 2.9/ SD = 0.82), as shown in Table 2. Third conclusion: The writing performance of the intervention and control group falls within the middle range for all writing tasks, except for the posttest offer, where the control group received a below-average overall assessment for this text type. The correlation between the variables Overall impression and Attitude towards the task is illustrated in the following correlation table using a correlation calculation (Kendall’s tau-b), initially based on the pretest application letter (n = 103; see Table 3): 182 4 Results <?page no="183"?> 18 In addition, there is a significant correlation between the variables overall impression and I like writing letters (p-2-sided = .005*), which is confirmed using the Pearson chi-square test. - r - (n = 103) - Overall Impression 1 + Attitude (Application Task) 0.29* Overall Impression + Good Quality Letters 2 0.18 Overall Impression + Enjoy Letters on the PC 2 0.23 Good Quality Letters + Attitude (Application Task) 0.15 Enjoy Letters on the PC + Good Quality Letters 0.26 Enjoy Letters on the PC + Attitude (Application Task) 0.11 Tab. 3: Correlations (Kendall’s tau-b) for the application (pretest) Note: 1 Low level 1.0 to 2.0 (holistic coding of student texts; MIN 1/ MAX 5), 2 Low level 1.0 to 2.0 (student survey; MIN 1/ MAX 4), * (p < .005) The results indicate a positive correlation (r = 0.29, p < .004*) between the holistic assessment Overall impression and the item Attitude towards the writing task (low level). Similar positive correlations are also observed for the items I like writing letters on the PC and I write good letters (r = 0.26, p < .009). 18 The correlations for the posttest application and the pretest offer, overall impression with the variables I like writing letters on my PC and I write good letters, are also positive, but not statistically significant. However, additional correlations appear at the end of the intervention in the posttest offer (see Table 4). - - r - - (n = 103) - Overall Impression 1 + Attitude (Task Offer) 0.33* Overall Impression + Good Quality Letters 2 0.13 Overall Impression + Enjoy Letters on the PC 2 0.02 Good Quality Letters + Attitude (Task Offer) 0.32** 4.1 Key results from the affect research area 183 <?page no="184"?> 19 The results of the regression calculations for the other survey dates are lower (pretest offer R ² = .12/ p < .006; posttest application R ² = .04; posttest offer R ² = .09). Enjoy Letters on the PC + Good Quality Letters 0.38** Enjoy Letters on the PC + Attitude (Task Offer) 0.12 Tab. 4: Correlations (Kendall’s tau-b) for the offer (post-test) Note: 1 Low level 1.0 to 2.0 (holistic coding of student texts; MIN 1/ MAX 5), 2 Low level 1.0 to 2.0 (student survey; MIN 1/ MAX 4), ** (p < .001) There are positive and significant correlations for the post-test offer (see Tab. 4). The cross-tabulation calculations mathematically examined the correlation between the variables Low level overall impression and Low level attitude. The following results were obtained for all test subjects (n = 103): At the end of the writing project, there is a highly significant correlation between the Attitude towards the task and the overall impression (r = 0.33; p-2 sided = 0.001**). Similarly, a highly significant correlation is also observed between the item I write good letters and the item Attitude towards the task (r = 0.32; p-2-sided = 0.001**). Fourth conclusion: It is thus clear that subjects with a lower attitude towards writing can hardly overcome the average middle level of the rating scale. In addition, it was further examined whether the regression analysis for the attitude to writing also calculated an explanatory relationship between the variables overall impression of the application letter in the pretest (before the intervention) and the items positive attitude to writing, I like writing letters on the PC, and I write good letters. The model summary shows the following result for the application letter pretest: The dependent variable Overall impression has a positive correlation with the three included items (I like writing letters on the PC, I think I write good quality letters, and Positive attitude towards the task) and can therefore be explained by these (R² = .18). This calculated regression has a significant correlation (p < .000**). 19 The three included items therefore explain the variance in writing performance. 184 4 Results <?page no="185"?> Fifth conclusion: 18 % of the differences in the writing results in the pretest application can be attributed to the students’ attitude towards writing (affect). This indicates a clear correlation between the attitude towards writing and the text quality at the beginning of the writing intervention. The following question was addressed at the beginning: • Q 1 : What is the relationship between the students’ writing attitude and their writing performance during the writing project? The results presented allow the conclusion that there is a positive correlation between writing attitude (Affect) and writing performance. In summary, the following hypothesis can be stated for the first area of investigation, Affect: ✓ AH 1 : If students have a positive attitude towards writing business letters, they also write “application letters” and “unsolicited offers” at a medium to high level (overall impression), compared to those with a negative attitude towards writing. 4.2 KEY RESULTS FROM THE BEHAVIOR STUDY AREA In terms of job applications, the intervention group achieved almost constant results in the middle range in the pre-test (M = 2.9/ SD = 0.80; MIN 1, MAX 5) and post-test (M = 2.8/ SD = 0.76) compared to the overall impression variable (B001) (see Table 5). The control group also demonstrated average-level writing in both the pre-test and the post-test. A T-test comparison between the control and intervention groups revealed no significant differences in mean values. However, there is a significant correlation with the variable Low level overall impression in the pre-test application and in the post-test application across all test subjects (p < .000**). 4.2 Key results from the behavior study area 185 <?page no="186"?> - Pretest Posttest - M (SD) M (SD) - IG CG IG CG - (n = 68) (n = 35) (n = 68) (n = 35) Overall Impression of Application (B001) 1 2.9 (0.80) 2.9 (0.69) 2.8 (0.76) 2.9 (0.64) Overall Impression of Offer (B001) 1 2.6 (0.87) 2.5 (0.83) 2.9 (0.82)** 2.0 (0.87) Language Pragmatics (Application) 2 3.0 (0.63) 3.0 (0.56) 2.9 (0.86) 3.1 (0.53) Language Pragmatics (Offer) 2 2.6 (0.78) 2.7 (0.93) 3.0 (0.94) 2.0 (1.01) Language Classification (Applica‐ tion) 3 3.0 (0.74) 3.3 (0.64) 3.0 (1.06) 3.5 (0.70) Language Classification (Offer) 3 2.8 (0.77) 3.1 (0.76) 3.2 (0.77) 2.1 (1.08) Word Count (Application) 126 (36) 127 (30) 164 (34)** 114 (31) Word Count (Offer) 124 (37) 124 (44) 174 (30)** 117 (54) Tab. 5: Mean values and standard deviations of the items for measuring text quality (Behavior) Note: 1 MIN 1 / MAX 5 (holistic coding of student texts), 2 MIN 1/ MAX 5 (IMOSS coding of student texts with 5 items), 3 MIN 1/ MAX 5 (IMOSS coding of student texts with 3 items), ** (p < .000), IG = intervention group, CG = control group First conclusion: Students in the intervention and control groups demon‐ strated average writing levels in their application texts, with no significant differences between the groups. Students who achieved a low overall assess‐ ment in the pretest continued to perform at a similarly low level in the post-test. The examination of the offers shows an improvement in the overall impression of the intervention group (from M = 2.6/ SD = 0.87 to M = 2.9/ SD = 0.82). This improvement is significant (p < .000**), in contrast to the control group, which showed a decline in overall impression (from M = 2.5/ SD = 0.83 to M = 2.0/ SD = 0.87). The number of pupils in the intervention group with a low assessment of the overall impression was halved from 41 % (n = 28) to 21 % (n = 14) over the 186 4 Results <?page no="187"?> course of the intervention. This means that the struggling writers, in particular, benefited from the intervention. In the pretest, the control group predominantly wrote offers at a medium level, but a shift was observed in the posttest: 63 % (n = 22) of the control group wrote offers at a lower level. This indicates a clear increase in the number of control students receiving low ratings in overall impression over the course of the intervention. The T-test with independent samples confirms that the intervention group and control group did not differ significantly from each other in the pretest. In the post-test, however, a significant difference in writing performance can be observed in the overall impression (p < .000**; see Table 5). Second conclusion: Like the intervention group, the control group also started in the middle of the assessment range, but its performance deteriora‐ ted, while half of the intervention students with an initially low writing level improved in the post-test. The language system and language pragmatics scales (MIN 1, MAX 5) showed the following results after the analytical rating (see Table 5): The intervention group achieved average performance in the cover letter for the language-sys‐ tematic area (M = 3.0/ SD = 0.74 to M = 3.0/ SD = 1.06) and in the language-prag‐ matic area (from M = 3.0/ SD = 0.63 to M = 2.9/ SD = 0.86). The control group showed average performance and slight increases in the language-systematic area (M = 3.3/ SD = 0.64 to M = 3.5/ SD = 0.70) and language-pragmatic area (from M = 3.1/ SD = 0.56 to M = 3.1 / SD = 0.53 ). The intervention group improved in both language pragmatics (M = 2.6/ SD = 0.78 to M = 3.0/ SD = 0.94) and language systematics (M = 2.8/ SD = 0.77 to M = 3.2/ SD = 0.77). The control group, on the other hand, showed a decline in both language pragmatics (M = 2.7/ SD = 0.93 to M = 2.0/ SD = 1.01) and language systematics (M = 3.1/ SD = 0.76 to M = 2.1/ SD = 1.08). Before the intervention, both the control group and the intervention group had similar language-systematic and language-pragmatic writing performance in the medium assessment range. This is also reflected across all IMOSS coding variables. Similar to the overall impression, the comparison between the pre-test and post-test of the applications shows consistent outcomes, regardless of the intervention. However, the offers stand out from the pretest: while the control group’s performance declined, the intervention group showed improvement. 4.2 Key results from the behavior study area 187 <?page no="188"?> In addition, the regression analysis for the job application pretest shows that the language system and language pragmatics scales have a significant effect on the overall impression (R² = 0.59; p < .000**). The correlation analysis further indicates that the language pragmatics sub-area (p < .893) has a significantly stronger influence on the overall impression (p < .000**) than the language system sub-area (p < .117). In the posttest application, both the language pragmatics and language systematics scales significantly predict the overall impression (R² = 0.79; p < .000**). The regression analysis of the pretest offer also shows that the language pragmatics and language system scales predict the overall impression (R² = 0.51; p < .000**). In the posttest offer, both the language pragmatics and language system scales also predict the overall impression (R² = 0.65; p < .000**). Third conclusion: The test subjects showed average performance in the language pragmatic and systematics across the applications. However, in the post-test, the offers revealed a notable difference: while the control group exhibited low performance, the intervention group demonstrated clear improvement. The language pragmatics and language systematics scales reliably predict the overall impression of a text. The study variable number of words produced the following results when comparing the applications (see Tab. 5): The intervention group improved continuously from the pretest (M = 126/ SD = 36) to the posttest (M = 164/ SD = 34). Even during the intervention (M = 150/ SD = 33), an increase in word count compared to the pretest was already observable. Only 10 % (n = 7) of the test subjects wrote fewer words in the post-test than in the pre-test. In contrast, the control group wrote fewer words in the posttest application (M = 114 / SD = 31) than in the pretest (M = 127 / SD = 30). A T-test conducted on the job application post-test revealed a significant difference in mean values between the intervention and control groups (p < .001**). The intervention group also demonstrated an increase in word count from the pretest (M = 124/ SD = 37) to the posttest (M = 174/ SD = 30). If the intervention group is divided into the subgroup of ninth graders (n = 52; from M = 125/ SD = 41.67 to M = 172/ SD = 30.71) and trainees (n = 16; from M = 121/ SD = 21.26 to M = 179/ SD = 298.57), continuous improvement of both groups is evident in both groups, although the difference between them is not statistically significant. The control group, by contrast, wrote fewer words between the pretest offer (M = 124/ SD = 44) and the posttest offer (M = 117/ SD = 54). It is also evident 188 4 Results <?page no="189"?> that there is a significant difference in the number of words in the posttest offer between the intervention group and the control group (p < .000**). Fourth conclusion: The effect of the intervention is recognizable from the number of words. In the post-tests, the intervention group wrote significantly more words than the control group. The guiding question for the Behavior study area was: • Q 2 : How well do students perform in writing cover letters and unsolicited proposals during the writing project using the Self-Regulated Strategy Development approach? The results show that students in the intervention group could write texts at an intermediate level or higher. Furthermore, unlike the control group, they did not decline in the variables overall impression, language pragmatics, language system and text length, but rather improved in the post-test. Therefore, the null hypothesis can be rejected and the alternative hypothesis accepted. ✓ AH 2 : At the beginning of the writing project, students participating in the project write “application letters” and “unsolicited offers” predominantly at an intermediate level with regard to the variables overall impression, language pragmatics, linguistic system, and number of words, which they maintain or slightly increase in the course of the writing project. 4.3 KEY FINDINGS FROM THE CONTENT LEARNING STUDY AREA By responding to the questionnaire items (MIN 1, MAX 4), the intervention group indicated that they write applications with above-average attention to content (see Table 6). There is hardly any difference between the pre-test (M = 3.4/ SD = 0.54) and post-test (M = 3.4/ SD = 0.64), suggesting a consistent self-as‐ sessment by students regarding their content-related text type competence. 4.3 Key findings from the content learning study area 189 <?page no="190"?> - Pretest Posttest - M (SD) M (SD) - IG CG IG CG - (n = 68) (n = 35) (n = 68) (n = 35) Overall Impression of Application (B001) 1 2.9 (0.80) 2.9 (0.69) 2.8 (0.76) 2.9 (0.64) Overall Impression of Offer (B001) 1 2.6 (0.87) 2.5 (0.83) 2.9 (0.82)** 2.0 (0.87) Language Pragmatics (Application) 2 3.0 (0.63) 3.0 (0.56) 2.9 (0.86) 3.1 (0.53) Language Pragmatics (Offer) 2 2.6 (0.78) 2.7 (0.93) 3.0 (0.94) 2.0 (1.01) Language Classification (Applica‐ tion) 3 3.0 (0.74) 3.3 (0.64) 3.0 (1.06) 3.5 (0.70) Language Classification (Offer) 3 2.8 (0.77) 3.1 (0.76) 3.2 (0.77) 2.1 (1.08) Word Count (Application) 126 (36) 127 (30) 164 (34)** 114 (31) Word Count (Offer) 124 (37) 124 (44) 174 (30)** 117 (54) Tab. 6: Mean values and standard deviations of the items for measuring content learning Note: 1 MIN 1/ MAX 4 (student survey with 12 items), 2 MIN 1/ MAX 5 (IMOSS coding of student texts), 3 MIN 1/ MAX 10 (analytical coding of student texts), 4 MIN 1/ MAX 12 (analytical coding of student texts), ** (p < .000), IG = intervention group, CG = control group The intervention group thus ascribed themselves a higher level of content competence in the application text type, as reflected in both the pretest (M = 3.0/ SD = 0.65) and posttest (M = 3.1/ SD = 0.67). In the pretest, only 3 % of the intervention students initially rated their content knowledge in job applications as low, compared to 6 % after the intervention. For the offers, below-average ratings were higher: 18 % of intervention students gave low ratings in the pretest and 15-% in the posttest. The control group reported writing application texts that took content into account more than average as well. However, there was a slight decline from the pre-test (M = 3.3/ SD = 0.66) and post-test (M = 3.0/ SD = 0.71). The number of control students with low self-assessments increased significantly, from 6 % in the pretest to 17 % in the posttest. When asked whether they could 190 4 Results <?page no="191"?> 20 Coding results for content design pretest application (MIN 1/ MAX 5): Intervention group 9th grade M = 3.2 (SD = 0.65; n = 52); control group 9th grade M = 3.3 (SD = 0.68; n = 25); intervention group 2nd year of training M = 3.6 (SD = 0.77; n = 16); control group 2nd year of training M = 3.2 (SD = 0.71; n = 10). write and consider the contents of the offers, the control group showed almost consistent results (from M = 2.6/ SD = 0.82 to M = 2.7/ SD = 0.80). However, their assessment remained lower than those of the intervention group. Among the control students who initially rated their content-related text competence for offers as low, 34 % still gave low assessments in the posttest, indicating stagnation. First conclusion: The test subjects rated themselves as having a high level of content knowledge in the area of job applications. A shift is noticeable in the area of job offers: while the control group rated their content knowledge as average, a comparatively high proportion, 34 %, rated themselves below average, in contrast to the intervention group. The pretest application results show that all students (n = 103) performed at an average to good level in terms of content (M = 3.3/ SD = 0.72) based on the analytical evaluation of the student texts in the content area (C001). Differences among subgroups were minimal 20 , with the trainee intervention group standing out slightly (M = 3.6/ SD = 0.77). However, no significant differences were observed before the start of the intervention. The posttest application results showed that all test subjects achieved results in the medium range (M = 3.1/ SD = 0.75) for content evaluation, with significant differences between subgroups. The trainee intervention group, in particular, stood out again with a higher mean score (M = 3.7/ SD = 0.75). However, when comparing the preand post-test results for the intervention group (from M = 3.3/ SD = 0.69 to M = 3.1/ SD = 0.80) and the control group (from M = 3.3/ SD = 0.68 to M = 3.1/ SD = 0.63), the combined results in the area of content design (C001) for the applications remained largely stable, showing no substantial change following the intervention. A noticeable change is evident in the offers. The intervention group improved in terms of content (from M = 2.7/ SD = 0.85 to M = 3.1/ SD = 0.82), while the control group showed a decline (from M = 2.7/ SD = 0.90 to M = 2.2/ SD = 1.3). Thus, in the post-test offer, the content evaluation between the intervention group and the control group differs by a whole evaluation level, which also shows significantly different mean values confirmed by the T-test (p < .000**). 4.3 Key findings from the content learning study area 191 <?page no="192"?> 21 Coding results for the sum of the aspects pretest application (MIN 1/ MAX 10): Intervention group 9th grade M = 6 (SD = 1.82; n = 52); control group 9th grade M = 6 (SD = 1.25; n = 25); intervention group 2nd year of training M = 7 (SD = 1.70; n = 16); control group 2nd year of training M = 6 (SD = 1.30; n = 10) Coding results for the sum of the posttest application aspects (MIN 1/ MAX 10): Intervention group 9th grade M = 8 (SD = 1.49; n = 52); control group 9th grade M = 5 (SD = 1.78; n = 25); intervention group 2nd year of training M = 9 (SD = 0.93; n = 16); control group 2nd year of training M = 6 (SD = 1.41; n = 10) If attention is focused solely on the intervention students who write below average in terms of content in the offers, it is noticeable that the number is decreasing. While this figure was 38 % in the pretest offer, it was only 21 % during the intervention and 16 % in the posttest. In contrast, the number of control students with a low level of content organization (C001) increased from 13 students (37.1%) to 23 students (65.7%) between the pre-test and post-test. Across all four writing tasks, the Kendall tau b calculation also shows that there is a significant correlation between the content of the letter (C001) and the overall rating of the letter (B001) (p-2-sided = 0.000**). Second conclusion: The test subjects write applications with good content. However, it is particularly evident in the offers that the intervention group benefits from the intervention and stands out significantly in comparison to the control group. The content of the letter has a significant effect on the overall assessment. The results of the analytical dichotomous ratings for the content aspects/ ele‐ ments in the letters of application (scale 1-10) and offer (scale 1-12; see Table 6) reveal the following: In terms of content, the intervention group improved significantly in the application letters, far surpassing the mean value of five aspects. In the pretest, the intervention students initially achieved a mean score of 6 aspects (SD = 1.80), which increased to 8 aspects (SD = 1.41) after the intervention. This indicates that, on average, 8 out of 10 possible aspects of content were considered in the post-test applications. The control group initially had similar rating in the pretest with a mean value of 6 aspects (SD = 1.27), but this value remained unchanged in the posttest either. 21 A closer examination of the individual content elements in the application letters shows that approximately 80 % of all students have a particularly good command of the following aspects: stating the Application objective in the subject line (B29), introducing themselves (B26), and explaining how they 192 4 Results <?page no="193"?> found the company (B23_2) particularly well. In contrast, improvements in other content areas, such as repeating the subject in the introduction (B23_1), including a request for an interview (B27), stating a potential start date (BZ1), listing attachments (BZ2), and including an appropriate closing sentence (B28), were observed only among students in the intervention group. There were also isolated improvements in the aspects like the description of current activities (B24) and reasons for motivation (B25). The number of students who considered less than 50 % of the aspects rose from 11 % to 29 % in the control group, whereas this figure dropped sharply from 18 % to 3 % in the intervention group. These results indicate that the control group performed worse overall in terms of the content of the applica‐ tion assessments, particularly in the lower assessment range. No significant difference was observed between the control and intervention groups in the pre-test, indicating comparable starting conditions. However, the post-test results revealed significant differences in mean values between the two groups (p-2-sided = 0.000**). The offers were also analyzed using dichotomous coding of content elements (see Tab. 6). A comparison of the subjects shows the following results: For the offers, the intervention group was able to increase the number of content aspects/ elements from M = 5 (SD = 2.13) to M = 7 (SD = 2.19) (MAX 12) when comparing pretest and posttest. In contrast, the control group showed a significant decline (p-2-sided = 0.000**) from the pretest (M = 5/ SD = 1.99) to the posttest (M = 4/ SD = 2.03). The T-test and the ANOVA calculation for the posttest offer also confirm a significant difference in the mean values between the control and intervention groups (p < .000**), which was not present in the pretest. Further analysis of the subgroups for the number of content elements in the offers shows: Both the ninth-grade intervention group (from M = 5/ SD = 2.12 to M = 6/ SD = 1.90 ) and the trainee intervention group (from M = 5/ SD =2.07 to M = 9/ SD = 1.87 ) improved significantly. Notably, both subgroups started with the same number of content-related elements, but the trainees showed a greater degree of improvement than the ninth graders. In contrast, both control groups either stagnated or declined: the ninth-grade control group remained constant (from M = 4, SD = 1.64 to M = 4, SD = 1.94), while the trainee control group decreased slightly (from M = 6, SD = 1.96 to M = 5, SD = 2.04). A detailed analysis of the individual content aspects of the offers shows that the aspects including P.S. (AZ1), discount (AZ2), subject line (A22), introduction (A23_2), advantages of the offer (A26), benefits of the offer (A27), invitation to purchase (A28), and attachments (AZ3) are well managed and mastered by over 4.3 Key findings from the content learning study area 193 <?page no="194"?> 80 % of the students. Additonally, the intervention group also improved the aspects such as concrete description of the product/ service (A24) and time limit of the offer (AZ4). Minor improvements were observed in aspects like repetition of the subject (A23_1) and consideration of the target group (A25). In the control group, 26 students (74.3%) produced offers with a below-median number of content elements in the pre-test. This figure increased to 29 students (82.9%) in the post-test. In contrast, in the intervention group, 44 students (65 %) were below average in the pre-test. However, this number dropped significantly to 20 students (29 %) in the post-test. Third conclusion: In terms of the content of the offers and applications, the intervention group demonstrated significant learning gains in terms of the number of text elements/ aspects, while the control group’s performance declined. When examining whether a relationship exists between the overall impression (B001), the content design (C001), the number of words, and the number of aspects across all test subjects, it becomes apparent that there are correlations: The correlations in all paired samples are highly significant (p < .000**). For example, in both the application pretest (first value) and the application posttest (second value), the content design correlates with word count (r = 0.62 / r = 0.51) and with the overall impression (r = 0.54 / r = 0.75). This result can also be observed for the pretest and the posttest offers: the content design correlates with word count (r = 0.54 / r = 0.58) and with the overall impression (r = 0.64 / r = 0.76). There are also significant correlations between the number of words and the sum of the aspects (r = 0.49/ r = 0.69; p-2-sided < .000**) in the pre-test and post-test offers. The correlation calculations with Kendall-Tau-b for below-average results indicated significant correlations in the application pretest and offer posttest among word count, overall impression, and content design (p < .000**). In the pretest application (before the intervention), the correlations are very strong: students with a below-average writing level (overall impression B001) tend to have a low word count (r = 0.41; p < .000**), a low number of content aspects (r = 0.46; p < .000**), and a low content rating (C001) (r = 0.63; p-2-sided < .000**). Across all subjects (n = 103), the word count also correlates significantly with the number of content aspects (r = 0.56; p-2-sided < .000**). At the end of the intervention, correlations with Kendall-Tau (see Tab. 7) show that show that a low number of content aspects significantly correlates 194 4 Results <?page no="195"?> with fewer words in the posttest offer (r = 0.40; p-2-sided < .000**), between the low overall impression (B001) and the low impression of content (C001) (r = 0.70; p-2-sided < .000**), and between the low level of content (C001) and the word count (r = 0.34; p-2-sided < .001**). Furthermore, low content design scores also correlate with reduced word count (C001) (r = 0.49; p-2-sided < .000**). - - r - - (n = 103) Word count + Overall impression (B001) 1 0.35** Word count + content design (C001) 1 0.34** Number of words + number of aspects 2 0.40** Word count + content assessment 3 0.1 Overall impression (B001) + Content design (C001) 0.70** Overall impression (B001) + number of aspects 0.57** Overall impression + assessment of content 0.12 Content design (C001) + number of aspects 0.49** Content design (C001) + content assessment 0.19 Tab. 7: Correlations (Kendall-Tau-b) with content design, overall impression, word count & student assessment of content (offer posttest) Note: ** (p < .000), 1 MIN 1/ MAX 5 (IMOSS coding of student texts), 2 MIN 1/ MAX 12 (analytical coding of student texts), 3 MIN 1/ MAX 4 (student survey with 12 items) There is a significant correlation between the number of content aspects in the application post-test and the overall impression (B001). This indicates that a lower number of content aspects is significantly associated with lower writing quality, as reflected in the overall impression rating. Interestingly, however, there is no significant correlation between these scores and the students’ self-assessed knowledge of text types. A regression analysis further examines the relationship between the variables Low overall impression, evaluation of the aspects, the content rating (C001), and the response item Content. The following observations were made: The regres‐ sion analysis shows that in the pretest application, before the intervention, a low number of content aspects is significantly associated with a lower overall impression (R² = .41; p < .000**). Similarly, at the end of the intervention (in the 4.3 Key findings from the content learning study area 195 <?page no="196"?> post-test offer), there is also a significant correlation between the number of content aspects and the overall impression (B001) (R² = .56; p-2-sided < .000**). Fourth conclusion: The content design (C001) shows a significant correla‐ tion with the overall impression (B001), across all four texts. It is also highly significantly linked to both the word count and the overall impression (B001) (p < .000 **). This indicates that letters with more substantial content tend to be longer and are evaluated more positively. The students’ self-assessments, on the other hand, show only weak and statistically non-significant correlations with both the overall impression (B001) and the content assessment (C001). In addition to the analyses presented above, the students’ feedback discussions on their texts were examined. These feedback discussions, which focused on the students’ initial drafts (pre-texts), were evaluated using the same content checklist that guided the feedback sessions. For the present analysis, each piece of content feedback was dichotomously coded by the raters as either “fully achieved” (1) or “not achieved” (0). Audio recordings were available for feedback sessions involving 43 students for the application letters and 35 students for the unsolicited offers. However, some recordings were excluded from analysis due to technical issues such as missing or incomplete recordings or poor audio quality. Although all inter‐ vention students participated in the feedback interviews, only some of these were analyzed in detail. For this purpose, a systematic random sample of ten students (n = 10) from the intervention group, five ninth-grade students and five vocational trainees, was selected. These students took part in the feedback interviews for both the application and the offer. They received content-focused feedback, which they could then use to revise and improve their pre-texts. A total of 64 pieces of content-related feedback were identified across the two feedback sessions for these ten students. The dichotomous content-analytical rating (0 = not achieved; 1 = achieved) revealed that 36 out of 64 pieces of content feedback are effective (improvement of the text), nine pieces of feedback are neutral (no change to the text), and 19 are not effective (deterioration of the text). The most frequent improvements in the application letters were observed in the following aspects: starting date (BZ1) - 4 times, how the company was found (B23_2) - 3 times, reason for application (B25) - 4 times, and concluding sentence (B28) - 3 times. In the offers, the most frequent improvements included: P.S. (AZ1) - 7 times, investment/ attachment (AZ3) - 7 times, discount (AZ2) 196 4 Results <?page no="197"?> - 6 times, customer’s attention/ interest (A23_2) - 6 times, and invitation to purchase (A28) - 5 times. All of these listed content aspects showed imeasurable improvement in the students’ texts from pre-test to post-test based on the content checklist. Fifth conclusion: Both vocational school students and students from gen‐ eral education schools benefit from learning about the content criteria for writing application letters and unsolicited offers. The evaluation of the peer group discussions clearly shows that students who actively participate in the feedback discussion are able to use the positive output of these discussions for their texts. The sample indicates that feedback discussions have a positive influence on text quality, provided that students learn how to engage with the feedback and use it to enhance their own writing. This demonstrates that student feedback was, in most cases, effectively incorporated into the revision of the exercise texts and the post-texts. The third area of investigation, Content Learning, was guided by the following research question: • Q 3 : What do students know about the content of application letters and un‐ solicited offers at the beginning and end of the writing project (intervention)? Furthermore, the findings for this area allow the following conclusion: when pupils are asked about their level of knowledge of the respective text types, they rate themselves as having a good understanding on average and this self-assessment does not change significantly in the post-test, even though their knowledge of application letters is higher than that of unsolicited offers. The regression analysis shows no significant correlations, only slight positive coefficients. Pupils with a low understanding of text types tend to write shorter and lower-quality application letters and unsolicited offers, both in terms of content and overall impression. Furthermore, there is a clear relationship between the number of content aspects and the length of the text. A low number of content elements is associated with shorter texts, which in turn corresponds to lower ratings in the overall impression. The evaluation of the sample of feedback interviews shows that these discussions significantly contributed to improving the text content. The results of the Content Learning study area support the following hypothesis: 4.3 Key findings from the content learning study area 197 <?page no="198"?> ✓ AH 3 : Compared to the control group, students in the writing project demonstrate greater knowledge of text types “cover letters” and “unsolicited offers” at the end of the writing project than at the beginning. 4.4 KEY FINDINGS FROM THE METACOGNITION RESEARCH AREA The results on planning are presented first (MIN 1, MAX 4). Table 8 below shows that the intervention group rated the planning for the quality of the letters at a medium level before the intervention (M = 2.9/ SD = 0.89). After the intervention, it remains at this average level (M = 2.7/ SD = 0.95), which does not differ significantly from the other subgroups. The control group also rated planning at a medium level (M = 2.8/ SD = 1.00), which is also maintained after the brief intervention (M = 2.5/ SD = 1.04). In the intervention group, the proportion of those who consider planning important for the quality of the letter increases from 22 students (32 %) to 31 students (46 %). However, after the intervention, the intervention group showed a slight decrease from 67 % to 54 % when asked whether planning was important for the quality of the letter. The attitude towards revising did not differ before the intervention. In both the intervention group (M = 3.0/ SD = 0.84) and the control group (M = 3.0/ SD = 0.71), the students’ attitude towards revising was above average. After the intervention, a slight decline in this assessment can be observed in the control group (M = 2.5/ SD = 1.04). The intervention group tended to maintain a positive attitude towards revising (M = 2.9/ SD = 0.94). However, there is no significant difference in the mean values between the students in the intervention and control groups. Table 8 shows further results of the survey on the sub-areas of planning and revision before and after the intervention: - Pretest Posttest - M (SD) M (SD) - IG CG IG CG - (n = 68) (n = 35) (n = 68) (n = 35) Planning for quality 2.91 (0.89) 2.77 (1.00) 2.68 (0.95) 2.54 (1.04) Revision for quality 3.04 (0.84) 3.03 (0,71) 2.91 (0.94) 2.51 (1.04) Handwritten Planning 2.12 (0.99) 2.11 (1,21) 1.90 (1.02) 1.97 (1.04) 198 4 Results <?page no="199"?> PC Planning 1.99 (1.04) 2.03 (1,18) 1.90 (1.04) 2.03 (1.12) Clear Sequence 2.84 (0.77) 2.77 (0.73) 2.78 (0.88) 2.37 (1.03) Think beforehand 2.97 (1.02) 2.86 (0.81) 2.68 (0.94) 2.54 (0.98) Start immediately without planning 2.28 (1.03) 2.63 (1.03) 2.49 (0.99) 2.69 (1.13) Write in one go, then revise 2.74 (1.10) 2.60 (1.00) 2.78 (1.03) 2.69 (1.08) Revise at the end 1.72 (0.93) 1.89 (0.99) 2.18 (1.05) 2.26 (1.01) Revise during writing 2.90 (0.93) 3.00 (0.91) 2.75 (0.99) 2.43 (1.07) Revise alone 2.87 (0.93) 2.91 (0.82) 2.96 (0.87) 2.80 (1.02) Revise with family 2.00 (0.89) 2.06 (0.94) 2.29 (1.05) 2.14 (1.14) Revise with classmates 2.04 (0.92) 2.14 (0.94) 2.13 (1.00) 1.89 (0.90) Revise at work 1.87 (1.04) 2.03 (0.92) 2.12 (1.13) 2.00 (1.09) Revision by teacher 1.96 (0,92) 2.26 (0.98) 2.18 (0.99) 1.80 (0.96) Tab. 8: Mean values and standard deviations of the items on writing strategies (meta‐ cognition) Note: MIN 1/ MAX 4 (student survey), IG = intervention group, CG = control group Before the intervention, the intervention students planned less by hand (M = 2.17/ SD = 0.99) or on the PC (M = 2.0/ SD = 1.04), but they tended to think ahead (M = 3.0/ SD = 1.02) and knew the order in which to write (M = 2.8/ SD = 0.77). Two-thirds of the intervention group considered planning important for text quality (M = 2.9/ SD = 0.89). However, the majority would still start writing immediately without planning (M = 2.3/ SD = 1.03). Before the intervention, writing in one go followed by revision was the dominant approach (M = 2.7/ SD = 1.10), while most revision occurred during the writing process (M = 2.9/ SD = 0.93). When they did revise, they did so alone (M = 2.9/ SD = 0.93) and less often with others (M = 1.9/ SD = 1.04), including classmates (M = 2.0/ SD = 0.92), parents (M = 2.0/ SD = 0.89), or teachers (M = 2.0/ SD = 0.92). Pupils were aware that revising helps improve their writing (M = 3.0/ SD = 0.84). After the intervention, the intervention group indicated, with relatively stable values, that they predominantly considered revising as important for text quality (M = 2.9/ SD = 0.94). From the students’ perspective, teachers (M = 2.2/ SD = 0.99), family (M = 2.3/ SD = 1.05), classmates (M = 2.1/ SD = 1.00), or the company (M = 2.1/ SD = 1.13) are less likely to help with revision. As before the intervention, students still primarily revised on their own (M = 3.0/ SD = 0.87), during the 4.4 Key Findings from the metacognition research area 199 <?page no="200"?> writing process (M = 2.8/ SD = 0.99), and less frequently at the end (M = 2.2/ SD = 1.05), although interestingly many reported writing and revising in one go (M = 2.8/ SD = 1.03). Both the control group and the intervention group showed relatively high levels of agreement regarding the importance of planning, although revising was rated slightly more relevant. The intervention group demonstrated consis‐ tent results for the item Pre-planning helps text quality, both before (M = 2.9/ SD = 0.89) and after the intervention (M = 2.7/ SD = 0.95). In comparison, the control group also showed similar average results (M = 2.8/ SD = 1.00; M = 2.5/ SD =1.04). However, a change was observed for the item Revising helps: while the intervention group maintained stable mean values between pretest (M = 3.0/ SD = 0.84) and posttest (M = 2.9/ SD = 0.94), the control group showed a clear decrease from the pretest (M = 3.0/ SD = 0.71) to the posttest (M = 2.5/ SD = 1.04). Nevertheless, there were no significant differences in the mean values of the intervention and control groups. First conclusion: From the students’ perspective, planning and revising were rated above average. There was no significant difference between the intervention and control groups over the course of the project. Calculations using Kendall Tau and Spearman Rho in the pretest application showed that the students do not perceive themselves as planning and revising their texts. In some cases, negative correlations were found between the number of words and the writing strategies such as revising (r = -0.06; p-2-sided = .650 ). However, there were significant correlations between the variables revising and planning (r = 0.36; p-2-sided = .003), which remained consistent after the intervention and also in the pretest offer. A chi-square test and the associated cross-tabulations showed no correlation between the strategies planning and revising as well as low level overall impression. The regression analyses carried out for planning and revising were also unable to predict text quality. Second conclusion: In the sample drawn, no correlation was found between planning and revising strategies, word count and overall impression. This means that although students continued to report planning and revising their texts, without a significant increase in these self-reports, they did, in fact, revise their pretexts and produced longer texts as a result of the intervention. 200 4 Results <?page no="201"?> This led to improved text quality. They also rated planning above average and revising even slightly higher as a relevant writing strategy. Their use of planning and revising strategies did not decrease after the intervention. The relevant items for this remain at a medium to high level. Although the number of words and text quality increased, no significant correlation was found between overall impression and the strategies of planning or revising and the number of words. This is followed by a more detailed analysis of the performance of planning and revising with the results of the keylog data in the intervention group (n = 40; see Table 9): - Pretest Posttest - M (SD) M (SD) Word production Application 170.26 (51.19) 197.02 (69.92) Word production Offer 161.87 (66.30) 226.43 (62.05)** Active writing time Application (in min.) 22.55 (6.28) 21.95 (8.15) Active writing time Offer (in min.) 19.49 (7.92) 25.38 (10.19)** Total writing time Application (in min.) 38.38 (9.31) 39.73 (13.73) Total writing time Offer (in min.) 41.20 (15.05) 46.90 (15.27)** Pause time in the Application process (in min.) 15.28 (6.73) 18.34 (8.88) Pause time in the Offer process (in min.) 38.93 (118.19) 20.98 (9.48)** Number burts Application 54.10 (18.36) 48.23 (15.78) Number of burts Offer 52.98 (18.63) 56.88 (21.26) Number of pauses Application 53.31 (18.44) 47.44 (16.32) Number of pauses Offer 52.15 (19.17) 56.24 (21.41) Tab. 9: Mean values and standard deviations of the keylogging results Note: ** (p < .000) A comparison of the words produced in the pretest (M = 170/ SD = 51) and posttest (M = 197/ SD = 69) shows that students wrote more words when writing the applications. This is also confirmed for the offers; in the pre-test (M = 161/ SD = 66) and post-test (M = 226/ SD = 62), significantly more words were produced, with a higher word count in the offers than in the applications. The comparison of the pretests shows that students wrote almost as much in the unfamiliar text type “offer” as in the familiar text type “application”. A significant increase in the 4.4 Key Findings from the metacognition research area 201 <?page no="202"?> mean values for the offers was determined using a T-test for a random sample (p-2-sided < .000**). The word proportion for all four writing tasks ranges from M = 0.81 (SD = 0.42) to M = 0.84 (SD = 0.38), meaning that just under 80 % of all written words are used for the final product, indicating that revision occurred. The active writing time remains almost the same between the pretest (M = 23 min./ SD = 6.28) and posttest application (M = 22 min./ SD = 8.15). The total writing time for applications remained stable at 38 minutes (SD = 9.31) in the pretest and 39 minutes (SD = 13.73) in the posttest. In contrast, a significant increase in mean values was observed for the offers (p < .000**): while students spent an average of M = 41 minutes (SD = 15.05) in the pretest offer, this increased by six minutes to M = 47 minutes (SD = 15.78) in the posttest offer. The average active writing time for the offers also increased significantly from M = 19 (SD = 7.92) to M = 25 minutes (SD = 10.19; p-2-sided < .000**). Third conclusion: The writing project demonstrates positive effects. Over the course of the project, the intervention students wrote significantly more words in the application and offer posttests. They revised their texts, discarding nearly 20 % of the original content. The writing time increased significantly only for the offer. The analysis of the number of pauses and the pause time as variables shows planning phases through the increase in pauses at the beginning of the writing process (the first two intervals). Revision phases are perceived by an increase in the number of words written, the active writing time, and the proportion of words. In addition, pauses associated with revision tend to occur towards the end of the writing process (the last two intervals). Each writing process is thus divided into five equal writing intervals. This procedure is carried out automatically by the Inputlog software. This setting allows for a detailed examination of pause and writing behavior across all writing intervals. The explanations of the results (see Table 8) and the corresponding figures (38 and 39) follow. The results show that the number of pauses across all writing tasks averages eleven per writing interval over the course of the writing project. In the applications, the number of pauses between the pre-test (M = 43) and the post-test (M = 33) decreased, but the average pause duration increased slightly from 18 to 22 seconds (see Tab. 8). 202 4 Results <?page no="203"?> Fig. 38: Number of breaks for offers & applications (pre/ post test, n = 40) Fig. 39: Break time for offers & applications (pre/ post test, n = 40) 4.4 Key Findings from the metacognition research area 203 <?page no="204"?> The number of pauses at the beginning of the writing process for the applications is just under eleven, with an average pause time of 16 seconds (see Fig. 38 and 39). In the posttest, the number of pauses decreases by almost two, but the pause time in the first interval doubles to 35 seconds. Consequently, the pause time per pause doubles in the first interval from the pre-test and post-test. The second interval is similar to the pre-test, with eleven pauses and 18 seconds per pause. At the end of the writing process, the number of pauses ranges from just under ten in interval 4 to nine pauses in interval 5 during the pre-test. The length of the pauses increases from 22 seconds in interval 4 to 31 seconds in interval 5. In the posttest, the number of pauses decreases again in contrast to the beginning intervals, ranging from just under ten in interval 4 to seven in interval 5. However, the pause time increases significantly in interval 5 compared to interval 4. As a result, there are fewer breaks at the end, but they tend to be longer than in interval 4. The number of pauses between the pretest and posttest increased slightly for the offers. Compared to the posttest application, there are just under ten more pauses over the entire writing process (see Table 8). The number of pauses at the beginning of the writing process is just under eleven across the pretest offer in the first two intervals, with an average pause time of 27 to 29 seconds (see Fig. 38 and 39). The number of pauses at the end of the writing process in the fourth interval is just under eleven with an average pause length of 23 seconds. In the fifth and final interval, however, the number of pauses decreases to seven, while the average pause length increases to just under 38 seconds. In the posttest offer, the number of pauses remains pretty much the same, but the pause duration per pause doubles, from 23 to 49 seconds in the fourth interval, and from 37 to over 52 seconds in the fifth interval. Fourth conclusion: The results indicate that the writing process becomes less linear from the first pre-test to the last post-test. Notably, towards the end of the writing process, there is a clear increase in the length of the pauses, accompanied by a decrease in the number of pauses. The following four figures summarize the relationship between the number of breaks and the length of breaks for each writing task. The first figure depicts the writing process for the pretest application (see Fig. 40). It shows that the task is not planned for long during the first intervals. In contrast to the end of the 204 4 Results <?page no="205"?> writing process, the number and duration of pauses are short at the beginning of the writing process, while longer pauses are observed towards the end. Fig. 40: Ratio of number of pauses to length of cover letter (pretest, n = 40) Figure 41 graphically illustrates the writing process of the application post-test. This clearly stands out from the other writing processes. There are very long pauses at the beginning and end. The number of pauses remains fairly linear, although they are most frequent in the middle of the writing process. 4.4 Key Findings from the metacognition research area 205 <?page no="206"?> Fig. 41: Ratio of number of pauses to length of cover letter (post-test, n = 40) Figure 42 shows the writing process of the offer pretest. Compared to the application pretest, there are many long pauses at the beginning and end of the writing process. Fig. 42: Ratio of number of breaks to length of offer (pretest, n = 40) 206 4 Results <?page no="207"?> 22 This point is discussed in section 5.1. Figure 43 presents the writing process of the post-test offer graphically. There are few pauses at the beginning. Long pauses can be observed towards the end of the writing process. Fig. 43: Ratio of number of breaks to length of breaks offered (post-test, n = 40) Fifth conclusion: When examining all four illustrations of the writing processes, a prototypical scissor model can be observed for familiar writing tasks and their associated routine writing actions. This means that the number of pauses is lower at the beginning of the writing phase than at the end of the writing process. Pause times are also shorter at the beginning than at the end, with the shortest pause times and lengths occurring toward the middle of the writing process. This can be seen in the figure Pretest application and Posttest offer. A valley model, on the other hand, appears with less routined tasks, as shown in the pretest offer. Even though the application is familiar, a delay in the writing process was observed in the posttest, which can be attributed to the very long pause time in the first interval. 22 Both at the beginning and at the end of the writing process, long pause times were observed, contrary to the number of pauses. In the middle of the writing process, the values for pause time and number of pauses converge, creating a valley model. 4.4 Key Findings from the metacognition research area 207 <?page no="208"?> In addition to the calculations presented, instructional videos on the modeling phase were also evaluated. The seating arrangement was frontal. The teacher’s desk was centrally located and equipped with a projector, laptop, blackboard, and flipchart displaying the ABC list and checklists. The students sat at small group tables and had their project documents with them. The evaluation of the videos for the ninth graders shows that they needed 17 minutes to plan the first part of the application letter and 25 minutes for the revisions in the following lesson, resulting in a total of 42 minutes of class time was needed for the modeling text. A total of 361 speech acts were transcribed. In addition to the teacher who carried out the modeling, 14 students (four female and ten male) actively participated in the creation of the text, with the teacher contributing the highest proportion of speech acts, 181 in total. Each act was coded and could receive several codes depending on its content. Suggestions (151) predominated in terms of frequency. Questions (101) and answers (99) balanced each other out as complementary speech acts. The majority of the suggestions (151) were accepted (102) and incorporated into the text; only a few rejections (18) were identified. Interruptions were very low (25), especially considering that whole classes attended the modeling phase. During the project, the intervention students worked with three checklists (content, formal, and language), which were also used for planning and revising during modeling. The analysis of the videos shows that all 19 content-related aspects of an application were addressed during modeling. The formal struc‐ ture was considered with a total of 18 aspects relevant to designing a DIN standard-compliant letter. Ten of the 14 linguistic aspects from the last checklist were addressed in the modeling. No explicit reference was made to foreign words, punctuation marks, nominalizations, or rhetorical issues. The reasons for the application, self-introduction, request for an invitation, introductory sentence, structure of the letter, length, font size, greeting line, language style, sentence structure, spelling, and avoidance of repetition were discussed in great detail as evidenced by a high number of speech acts. A total of 27 minutes of teaching time was required for the modeling text offer. In the first part, nine minutes were allocated for the planning phase and 18 minutes for the revision and 237 speech acts were recorded. Notably, the teacher had the most speech acts (98) of all speakers due to her modeling role, with one student being the most involved with 53 speech acts. Ten pupils actively participated in writing the model text: six female and four male. In contrast to the first modeling phase for the application letter, the number of speech acts and the overall duration were reduced, by almost 15 minutes less teaching time. In this phase, the suggestions (78) predominated in this analysis, the majority of which 208 4 Results <?page no="209"?> were accepted (57) and, as above, only a few were rejected (17). 57 questions (item) and 28 answers (response) were recorded, indicating a content-enriched modeling phase dominated by suggestions. However, unlike the first application phase, interruptions were higher (40) than the number of speech acts. The qualitative analysis of the transcript confirms that more than 20 con‐ tent-related aspects were discussed in detail. The aspects such as customer benefits, offer details, P. S., closing sentence with purchase encouragement, pricing, target group orientation, and subject line were especially emphasized. Formally, all 15 aspects were addressed, focusing on letter structure, recipient address, subject line, effective use of emphasis, and the greeting line were explicitly pointed out in detail. In terms of language, the product also paid attention to all 14 aspects listed by students, with particular attention paid to language style, grammar, sentence structure, typing conventions, creative formulation, and rhetorical questions. Sixth conclusion: The evaluation of the modeling phases showed how letters can be written on the PC with attention to content, form and, language. The active participation of the pupils was not disruptive, but somewhat helpful, as more suggestions were accepted than rejected. The number of interruptions was therefore not detrimental to the final result. A closer examination of the resulting text products reveals that almost all aspects of content, form, and language were taken into account. Notably, students were also able to make observations about formality, language, text planning, or text revision. This approach is therefore highly valuable for observing the writing process. The following question was asked for the fourth area of investigation: • Q 4 : What strategy do students use in the planning and revision phase when writing letters? The results lead to the following answer: The intervention group shows in the metacognition test area that the number of words in the application process has increased, though not significantly. The active writing time remains relatively constant at 22 minutes. The students revise their applications, as almost 20 % of the words are deleted. This can be explained by the constant survey results at level 3. However, the pausing behavior changes between the pre-test and post-test. In the application post-test, students tend to pause at the beginning compared to the pre-test. In the offers, the number of words in the process 4.4 Key Findings from the metacognition research area 209 <?page no="210"?> increases significantly. The active writing time increases between the pre-test and post-test. The beginning and end of the writing process are characterized by long pauses in the pretest offer. In contrast, in the posttest, students take more time to revise at the end of the writing process than at the beginning. By answering the fourth question, the following hypothesis applies: ✓ AH 4 : Text quality and length improve and increase through the writing tasks set in the project as students plan and revise their “cover letters” and “unsolicited proposals” using the writing strategies promoted in the eleven-week writing project (90 minutes once a week). 210 4 Results <?page no="211"?> 23 In my opinion, the term ‘motivation' and the attitude to writing associated with it in my work could be divided even more sharply in further studies into the terms of Hayes’ model (1996), which assigns the sub-areas ‘Attitudes & Beliefs’, ‘Cost/ Benefit Estimates’, ‘Goals’ and ‘Predispositions’ under ‘Motivation/ Affect’. In the current writing models by Hayes (2012) and Bachmann & Becker-Mrotzek (2017), only the term ‘motivation’ is used without a precise definition and assignment to other terms (see section 2.10). 5 DISCUSSION The following fifth chapter is divided into five sections. First, the results obtained are discussed (see Section 5.1). This is followed by a detailed discussion of the study, which addresses the quality criteria. The strengths and weaknesses of the study are also presented transparently (see Section 5.2). In the Outlook chapter, further desiderata are presented as suggestions for follow-up projects (see Section 5.3), before an epilogue concludes the fifth chapter (see Section 5.4). 5.1 DISCUSSION OF THE RESULTS The fundamental question of the doctoral project was posed at the beginning of the writing project: How does writing competence develop in students of general and vocational schools with the help of the “Self-Regulated Strategy Development” approach, using the two professional text types ‘cover letter’ and ‘unsolicited proposal’ over the course of a writing project? This question can now be answered with the help of the various results from the four areas of investigation: The results of the Affect section show that students with a positive attitude towards writing business letters tend to produce application letters and offers of a medium to high quality more effectively than students with a negative attitude towards writing. New text types such as the offer therefore have a negative influence on text quality if the attitude towards the task is low. Therefore, the attitude towards writing can have either a motivating or demotivating effect on writing performance (Weinert, 2001: 27). This result is also confirmed Hayes’ models (1996: 14; 2012: 371), which emphasize that motivation influences the writing process 23 influences the writing process (see Section 2.5). In the same way, this is also confirmed by Fix’s (2000: 325) writing didactics cube for writing didactics as an influencing factor on text quality. In this context, the willingness to make an effort to complete the writing task should also be emphasized (Kellogg, 1994: 59 f.). <?page no="212"?> The writing models of Baurmann and Pohl (2011: 97), as well as Baurmann and Becker-Mrotzek (2017) also consider motivation as an important influencing factor for writing competence, an insight that is confirmed by the present study. The correlation between low motivation and weaker writing competence is further supported by other studies (Efing, 2008: 28 f.). However, the intervention did not result in a significant improvement in students’ attitudes. This aligns with findings from other studies, which indicate that attitudes do not necessarily improve immediately following an intervention (Glaser, 2005: 230; Keßler, 2010: 198 ff.; see chapter 2.6), and in some cases, no observable changes occur at all (Hoefele et al., 2015). In the Behavior section of the study, it was found that students participating in the project wrote application letters and unsolicited offers at the beginning of the writing project predominantly at an intermediate level in terms of overall impression, language pragmatics, language system, and word count, a level that they maintained or slightly improved over the course of the writing project. Although the ninth graders had only completed job application training and internships in the school year, the more experienced students showed better writing skills in these text types, likely due to long-term memory and professional experience. The intervention appears to have brought about only slight improvements in the overall impression, primarily through practice and increased writing in the practice test, with more noticeable gains in the practice and post-test offers. The control group of trainees clearly showed that, even without targeted intervention, writing competence for applications and offers improved slightly, likely as a result of their professional environment (for e.g., working in the reception area, reading offers in the catering area). In contrast, the results from the secondary school control group show a decline in both the job application post-test and the job offer post-test when compared to the pre-test. The offer proved to be the more challenging text type for all participants. The offer post-test shows that, despite the intervention, the professionally experienced (trainees) achieved better writing results than the intervention group of ninth graders. There may be various reasons for these results. For example, Beer-Scharwächter (2008: 205), in her intervention study (n = 48) with lower secondary school pupils, observed that text quality may initially decline after the first phase of intervention, only to improve again after the second part of the intervention and the second posttest. This may explain why the text quality of the applications in the intervention group did not improve significantly. From a psychological perspective, this refers to stabilization through repetition (Winkel/ Petermann & Petermann, 2006: 119). The challenge 212 5 Discussion <?page no="213"?> in writing performance is that action-oriented tasks require not only content knowledge but also ability to solve problems, communicate effectively with others and use new media, which have an immense impact on the writing process and, in turn, on the product (Schaper et al., 2000: 132; Hayes, 2012: 372; see chapter 2.2). Mastering such situated writing tasks generally involves complex cognitive processes including transfer, transformation, reflection, and construction, which are some of the most challenging educational objectives for students (Kessler & Ziener, 2004). Writing competence, as a latent construct, does not always appear in writing performance (Neumann, 2014: 58 f.). Pupils with writing experience perform writing tasks more effectively than those without experience. This distinction is based not on age but on writing experience, as those with writing experience are better able to transform, revise, and reflect on their texts (Kessler & Ziener, 2004; Neumann, 2007: 194). This observation is also confirmed by the knowledge-tell‐ ing and knowledge-transforming strategies described by Bereiter & Scardamalia (1987). The present study still coincides with the age of the pupils, but writing development is not confined to the age of majority (Bereiter, 1980; Hayes, 2012: 372). The intervention intended to improve the unfamiliar text type offer, which is consistent with the results of other writing intervention studies (Kellogg, 1994; Glaser, 2005). Improvements were observed in the intervention groups in the linguistic-pragmatic and linguistic-systematic areas. These positive outcomes were brought about by the structure of the intervention, particularly, the use of prototypical patterns and checklists highlighting key text features, an approach that has already proven effective in other SRSD studies (see chapter 2.6). For the area of Content Learning, it can be stated that although the number of content-related text elements or aspects determines how closely a text type aligns with a standard, there is no prescribed number that needs to be achieved. Furthermore, the number reflects the degree of prototypicality and thus the approximation to a text type (Adamzik, 2008: 173). Adamzik (2008: 167) also discusses the fact that while certain linguistic characteristics are necessary for determining a text type, not all of them can or need to be considered, depending on their relevance. Teaching content aspects of text types such as application letter and offer letter proves useful and effective as an intervention in the ninth grade and vocational school contexts, as declarative knowledge supports performance (Kellogg, 1994: 59 ff.). This also explains the need for theoretical and practical final examinations. Secured knowledge has a positive influence on performance (Winkel/ Petermann & Petermann, 2006: 229; Thomas, 2007: 94; Sieland & Rahm, 2007: 199). This also applies to writing skills and the associated declarative knowledge of textual features (Bauermann & Pohl, 2011: 5.1 Discussion of the Results 213 <?page no="214"?> 97). Professional experience with application letters and job offers can positively influence writing performance as it offer prior knowledge for the pretest (see trainees compared to secondary school students in the pretest). However, the intervention shows a stronger influence than age and professional experience alone (see secondary school students in the control and intervention groups). Although it could be assumed that all participants had already mastered certain aspects before the intervention, as they had already written numerous applications in German for internships or apprenticeships, the intervention enabled the students to apply these aspects more consistently. By the end of the writing project, pupils in the intervention group demonstrated greater knowledge of text types, particularly, application letters and unsolicited offers, than the control group, compared to the beginning of the project. In the post-test, the ninth graders in the intervention addressed more content-related aspects than the control group in the second year of training, despite being, on average, four years younger, have no professional experience, and have only completed one internship. All intervention groups benefited from the intervention and showed improvement in the post-test compared to the pre-test. Several content-related aspects were successfully learned through the in‐ tervention including repetition of the subject in the introduction, request for an invitation to an interview, concluding sentence encouraging the reader, the inclusion ofdate of entry, relevant attachments such as CV and copies of certificates, concrete descriptions of products or services, and specification of time limit on offers. However, despite intervention,some aspects were still not implemented successfully. These included description of current activities, justification for why the job position is of interest to applicants, consideration of the target group (such as gender and age), and in some cases, inconsistent implementation of repetition of the subject in the introduction. It is likely that students and career starters strugggle to describe the job as they often lack experience or adequate information. Formulating a reason for applying for the position remains one of the most challenging tasks, even for professionals as it requires a high degree of target audience orientation, a realistic and positive self-assessment, and argumentative language. Similarly, considering the target group when drafting an unsolicited offer also demands a high degree of audience awareness, a skill nthat eeds to be acquired, particularly in marketing contexts, and can be enriched through personal vacation experiences or professional exposure to the hotel industry. As already explained, written language skills were fostered through the use of three checklists during peer feedback interviews. These checklists addressed content, formal, and linguistic text features, and their effectiveness is proven 214 5 Discussion <?page no="215"?> by other studies (Troia et al., 1999; Fix, 2000; Adamzik, 2004; Graham, Harris & Zito, 2005). The feedback discussions were valueable for the students and led to revisions of their initial drafts. Numerous other studies emphasize the effectiveness of discussing one’s writing, as text features can be transferred into revisions (de Beaugrande & Dressler, 1981; Kellogg, 1994, 2008; Fix, 2000; Augst et al, 2007; Nitz, 2010; Hayes, 2012; Braaksma et al, 2012). However, there were still some students who received extensive factual feedback, but implemented it minimally, the reasons for which may be manifold. One important factor is how individual students address feedback. Those with prior positive feedback experience are more likely to use it to their advantage. In contrast, a lack of experience or a negative attitude, may lead students to interpret feedback negatively, on a relational level (My classmates don’t like me) or on a motivational level (I still have so much to change),rather than as constructive input on the factual level, as intended through the use of the three checklists. These discussions occured in the context of “social interaction” (Heinemann, 2008: 132) and were, to a large extent, used effectively. This indicated that positive emotions enhanced the feedback process. Other studies also confirmed this observation: “Collaboration with other students in the writing conference is perceived as emotionally positive” (Fix, 2000: 121) and leads to improved writing through revision (Steendam et al., 2012: 131). It is also important to consider that students in German lessons are part of a class and thus engage in the language of instruction. This language is influenced by both the school context and students’ own linguistic backgrounds. Therefore, in the analysis of the recorded feedback interviews, communication in written language may be desirable, but breaches of standard language norms are natural in the school context (Feilke, 2012a.). Teachers should not perceive such deviations from linguistic norms merely as disruptions, since the content of the conversations, despite inconsistencies, proved to be productive, as this study demonstrates. Collaboration, as Hayes observes, positively influences writing process (2012: 371). In conclusion, the Metacognition area of investigation shows that the quality and length of the texts produced by the writing tasks improved when the students planned and revised their application letters and unsolicited proposals using the writing strategies promoted in the eleven-week writing project (90 minutes once a week). While factors such as breaks, writing duration, and text length alone cannot serve as indicators of metacognitive activity, the strategic use of planning and revision was the central component of the intervention. The students also confirmed in the accompanying survey that they use these strategies more frequently than average. 5.1 Discussion of the Results 215 <?page no="216"?> The writing performance, as reflected in the writing phase mapping, also shows distinct patterns of development - the scissor model for relatively automated writing actions and the valley model for writing actions and tasks that are not yet automated, as was particularly evident in the pretest offer. The intervention groups initially appeared confused by the new task and the rather unfamiliar text type, “offer”. This explains why the writing phase illustration of the offer pretest, with its frequent pauses throughout, appears so contrary to the other three writing phase illustrations. In their study, Rijlaarsdam and van den Bergh (2006: 41 ff.) note that longer pause times occur when ideas have to be generated and time has to be allocated for repeated rereading. This suggests that writing strategies such as planning, writing, and revising must not only be practiced inidvidually, but also repeatedly accordance with a structured procedure (Feilke, 2014: 15; Feilke, 2017: 158). However, this approach achieved better results after a longer intervention period, when comparing the pre-test with the post-test offer. One reason for this is that the offer text type was rather unfamiliar to the students, and no prototypical text samples were available. Consequently, the learning outcome for this text type was particularly evident, compared to job applications. The current representation model according to Bachmann and Becker-Mrot‐ zek (2017: 46) was thus empirically tested. As Bachmann and Becker-Mrotzek (2017) argue, it was confirmed in this study that text production is not only a problem-solving activity, but also an “adaptive resource-guided process” in which text patterns are used. The new writing task, the offer, entered the working memory as an adapted text pattern. By the end of the writing project, the students, particularly those in the intervention groups, were able to produce offers of higher text quality than their applications. In the present study, the eleven-week project resulted in a degree of autom‐ atization in long-term memory. The observations are empirically supported by the higher word count, improved use of text elements, longer writing time, higher content evaluations, and distinct writing phase mappings. Additionally, these findings are consistent with previous studies (Hayes & Flower, 1986; Rijlaarsdam & van den Bergh, 1996: 207 ff.). The teacher’s modeling added value by showing how a writing process can be carried out with all its sub-activities. The use of strategies also supported working memory resources. The teacher’s use of “think-aloud” technique revealed an authentic writing process, marked by mistakes, pauses, idea generation, revisions, repeated rereading, and reflection on content, form, and language. It became clear during the writing process that text production is not linear; revisions occur at various points in the text and often involve deletions. Such 216 5 Discussion <?page no="217"?> revisions are not easily possible on paper. Compared to writing with pencil and paper, the planning phase tends to be less pronounced when writing on a computer. Revisions are easier to carry out on a PC (Kellogg, 1994: 33), even though the intervention students did not fully realize this as a result of the intervention. The significant advantage lies in the opportunity for revision during the process. If new ideas or thoughts are generated while writing, the writer can immediately incorporate them and place them where they should ideally be embedded in the text (Becker-Mrotzek & Böttcher, 2012: 41). This indicates that revisions are constantly made during the process, but particularly towards the end of the writing process (Rijlaarsdam & van den Bergh, 1996: 207 ff.). This form of motor relief can only occur if “[…] motor familiarity with the writing tool is present” (Schneider & Anskeit, 2017: 284), which young people bring with them due to the influence of new media in their living environment and which tends to encourage revisions as a suitable tool (Hayes, 2012: 371; Berge et al., 2016: 180). It is important to provide students with guidance for revising, especially in the form of text samples or text style conventions, as business letters are subject to specific DIN standards that cannot be ignored: In summary, it can be stated that the specification of text type norms slowed down the general motivation to write, but led to more revisions in the area of functional appropriateness (especially for students with weak writing skills), while the freedom from specifications increased the general motivation, but resulted in more surface-related revisions (Fix, 2000: 308, translated by WKG). This can be supported by the schema theory from learning theory, because patterns help to reduce the complexity of reality and provide orientation within it (Kellogg, 1994: 10). It is debatable whether writing strategies should only be used when students are faced with a rather unfamiliar writing task. Inexperienced writers are more likely to benefit from this approach when solving writing tasks. In my opinion, writing strategies become automated when the writing task has been practiced from the writer’s perspective and can be solved without long pauses at the beginning. Automation blurs the individual work steps, which are otherwise perceived more distinctly on a mental level. This is comparable to swimming. Example: Experienced swimmers will no longer think about their swimming movement, the movement is automated. On the other hand, learning to swim and the associated first movements are much slower. Especially at the 5.1 Discussion of the Results 217 <?page no="218"?> beginning, mentally conscious strategies help to internalize the swimming technique and learn the swimming movement. Nevertheless, conclusions about metacognition should be drawn with caution and will provide more insights in the coming years of research. Overall, it is clear that the writing process models have become increasingly complex over the years and that the circle of knowledge has not yet been closed. The comparison of the writer with a “switchboard operator” (Flower & Hayes, 1980: 30) confirms the diverse operations involved in writing. The models, therefore, provide a recurring basis for discussion and research and show how many variables influence writing competence. SRSD, as a remedial approach, involving the teaching of text elements, writing strategies, modeling of the writing process, and collaboration through feedback conversations, is, as demonstrated in this paper, a suitable method for improving text elements, text quality and length, and writing attitudes, and promoting these through planning and revision strategies (Troia et al., 1999; Saddler et al., 2004; Graham, Harris & Zito, 2005). 5.2 CRITICAL REFLECTION ON THE STUDY In empirical educational research, studies are approved only if they are expected to produce a positive effect for pupils, for reasons of ethics and morality. However, unlike scientific research conducted in laboratory settings, field research is particularly prone to methodological errors because of the large number of variables that cannot be fully controlled (Grabowski, 2017: 321). Quasi-experimental research forms the middle ground, with its strengths and weaknesses, some of which are illustrated in the present study: 1. It is critical to note that the sample was not randomized as in a true experiment (Rost, 2007: 125; Graham & Harris, 2017: 14; Grabowski, 2017: 322). Randomization requires the allocation of learning groups at random. Since the group formation is not influenced from the outside, a high external validity can be achieved (Marx & Steinhoff, 2017: 260). This approach was not used for the study because it is “[…] didactically and ethically questionable if some of the learners receive maximum support while others receive only minimal support” (Marx & Steinhoff, 2017: 259, translated by WKG). The division of dependent groups such as a class could only succeed through two different interventions conducted in parallel. However, this would require splitting the class, 218 5 Discussion <?page no="219"?> which could itself influence the results. Even working with a smaller learning group can positively affect intervention outcomes. Moreover, an additional teacher would be needed to supervise the second half of the class, requiring both additional staffing and financial resources. Randomization is particularly suitable in studies with large numbers of participants and in laboratory situations, as randomization in a classroom context may lead to unfavourable and unbalanced compositions. For example, one group might randomly include a disproportionate number of boys or students with a migration background. This composition can potentially result in advantages or disadvantages when comparing outcomes between groups. However, not all of the design requirements of an experiment can usually be realized in the case of school and teaching-related research questions, if only because the composition of a school class cannot be manipulated for research purposes (Grabowski, 2017: 323, translated by WKG). However, this would be the only way to ensure internal validity. With a larger number of test subjects, the four-group plan would have been possible, in which the control group also receives alternative training (Grabowski, 2017: 324; Glaser & Meyer, 2017: 374). In the present study, the pretests showed that the group differences were not significant despite the lack of randomization, which makes comparability feasible. A final reason is that interventions with whole classes have external validity and ecological relevance, as they can be replicated and used in the classroom if successful, as Graham and Harris point out: True-and quasi-experiments when replicated provide a more trustworthy approach for identifying effective teaching practices in writing than the experiences and insights of professional writers or even those who teach writing (Graham & Harris, 2017: 14, translated by WKG). 2. Furthermore, the number of test subjects is too small for the statements to be generalized. In this sense, quantitative research results require representativeness. However, as the school system in Germany is federal, each federal state has the full authority over the lower secondary level. Only the vocational school system with its dual vocational training is legally regulated at the federal level. This makes comparisons with other federal states difficult. Ninth graders in Bavaria are taught differently than those in Lower Saxony, even though there are educational standards according to the Conference of Ministers of Education and Cultural Affairs. Another problem is time. For example, when the survey was conducted 5.2 Critical reflection on the study 219 <?page no="220"?> in 2013, there were secondary schools in Lower Saxony that have since been phased out due to the implementation of secondary schools and the expansion of comprehensive schools. For a comparison, the German courses at secondary school level would have to be examined in secondary schools. Replication studies would have to take this into account. It should be emphasized that an adequate sample size along with the use of diverse test instruments should also be supported by several researchers. In this project, the human and financial resources were only available through personal commitment. Text corpora with over 1000 texts are desirable from a researcher’s point of view, but can only be managed with financial backing (Steinhoff, 2017: 358). 3. The test instruments could have been supplemented. In order to strengthen the validity of the test subjects, the use of an intelligence test, spelling test, or a keyboard typing test, among others, would have been useful (Glaser & Meyer, 2017: 373). 4. There was no follow-up test to check the learning gains after the inter‐ vention (Marx & Steinhoff, 2017: 260). Follow-up tests are extensions of a pretest-posttest control group design. The aim here is to check the sustainability of the intervention (Grabowski, 2017: 328). It is a sign of the “transfer effect” (Glaser & Meyer, 2017: 378, translated by WKG) if the requirements in the follow-up test are modified by other writing tasks. 5. Quasi-experimental interventions in the field are often influenced by ex‐ ternal factors, especially over a longer period of time. Teaching disruptions, illnesses, and practice effects are just some of the problems that can be listed here (Marx & Steinhoff, 2017: 257). Although no selection bias or selection effect could be identified, the intervention was integrated into everyday school life. Effects from other subjects could have an influence. However, this can no longer be ascertained, as otherwise the entire school weeks would have to be documented. Practice effects from repeating the intervention twice, albeit with different writing tasks and text types, could also have an effect, as the task structure and approach became more com‐ prehensible for the pupils over the course of the writing project. Contrary to this, fatigue effects should also be listed as a potential impairment of internal validity. However, this could not be determined on the basis of the results, especially with the posttest offer (Grabowski, 2017: 323 f.). In addition to the weaknesses mentioned, strengths of the study can also be emphasized: 220 5 Discussion <?page no="221"?> 1. One strength lies in the wide selection of test instruments (questionnaire, writing tasks, video and tape recordings), which not only focus on the writing product but also examine the writing process when writing on the PC, during peer discussions, and in interaction with the teacher. 2. Although the study was conducted in 2013, it has not lost its topicality. Even if school structures or curricula have changed, writing on a PC will play an increasingly important role at the end of lower secondary level in the future. Researchers who examine the school of the future in 2050 confirm in their interviews that new teaching methods in schools are needed in response to digitalization (Chassa, 2016). Digitalization and the associated provision of tasks, inputs, or learning materials could increase the individuality of pupils due to an adapted learning pace, which still requires further research. Pupils are picked up where they are, such as the various stations in the writing project. Learning in lockstep is avoided, enabling more individualization. The use of computers is already a common occurence in elementary school and is expected to increase (ibid.). The physicist Mitra permanently installed computers for children in the middle of sidewalks in Indian slums. His theory is that when a child observes something new, its curiosity prompts it to learn and explore the use of computers, even without instruction, teachers, or manuals. The results of his study confirm that these Indian children independently acquired basic computer skills themselves in a very short time (ibid.). Teachers should therefore move away from a frontal approach and promote self-organized learning, taking into account the increasing heterogeneity and inclusion in schools (Giera/ Meyer & Süphke, 2012). For many pupils, phases such as modeling, including a frontal phase, in which the teacher sets the tone, add value to their learning process. 3. The study can be replicated, as the intervention was conducted in whole classes, an approach commonly practiced in schools. Furthermore, the test instruments were presented transparently in this study and can also be found in the appendix (see Giera, 2020). The increasingly required criteria of ecological, social, and external validity are therefore met (Glaser & Meyer, 2017: 380). This means that the effects determined can be tested empirically or applied experimentally in the classrooms: “Ultimately, neither empirical research results that are ultimately far removed from the context nor arbitrary research results that can be interpreted are suitable for implementation in practical contexts” (Glaser & Meyer, 2017: 380, translated by WKG). The field of professional writing has not yet been fully researched. 5.2 Critical reflection on the study 221 <?page no="222"?> Overall, writing studies are associated with “[…] high personnel, financial and time expenditure […]” (Marx & Steinhoff, 2017: 258, translated by WKG). The lead time alone, for approval by the school principal, parents, pupils, university, and, above all, the education authority, is time-consuming. While the writing project has certain has weaknesses, it also possesses strengths that make it a worthwhile contribution to further research in this field. This study adds value to empirical writing didactics. Although the research gap outlined at the beginning cannot be fully closed by such a project, it can provide an impetus for further research into the SRSD approach using new media in whole-classe settings, particularly at the interface of lower and upper secondary level through the integration of professionally relevant text types. For school practice, this study presented a feasible and effective teaching concept that takes into account a heterogeneous student body and the effectiveness of writing lessons. This is particularly beneficial for vocational schools, where German lessons are limited. Therefore, the SRSD approach can be implemented across schools in both lower and upper secondary levels. After only a few lessons, the number of student texts (190) increased significantly. SRSD addresses the writing process in its entirety through a variety of strategies, and weak writers can develop stronger writing attitudes, text type knowledge, writing experience, and communicative skills. Further research should be carried out into writing on the PC as part of German lessons at general education schools in the higher grades of lower secondary education. Increased attention from the business community and chambers of commerce could ensure that such writing projects are carried out at several schools with teachers who are well-trained in the subject. Additional writing projects and research in this area are highly desirable and should not be left behind as a one-man show without financial support, as happened in the present writing project. Coopertion within the research network is essential in order to conduct well-funded field studies with a variety of test instruments in inclusive classroom settings. 5.3 OUTLOOK Our society will continue to face new challenges in the future, which will inevitably also have an impact on educational institutions. The following con‐ siderations are relevant for German didactics and German lessons, particularly within the subfield of writing instruction: In the age of new media, it is the responsibility of researchers in the field of German didactics to address the emerging topic - keyword Industry 4.0. This new era, advancing not only in 222 5 Discussion <?page no="223"?> Germany, but globally, especially in industrialized nations, is characterized by increasing reliance on robotics, bots, and computers to support work across many domains. In Japan, for instance, robots already exist that can understand not only human language but also human emotions. These robots are program‐ med with facial expressions and gestures that are similar to those of humans (WDR, 2015: n.p.). School education and professional life are expected to evolve with the integra‐ tion of technological innovations. In the coming years, the central question may shift from how humans write to how computers generate written texts. In this context, text patterns will become increasingly important for computer science particularly in the field of artificial intelligence and programming. Consumers may be supported in writing various text types, such as cover letters, through commands or keyword-based prompts. Text modules and subsequent control through revision loops would be essential for this. The DIN standard for writing semi-private and business letters is especially well suited to such applications. The text types used in this study, application letters and unsolicited offers, could become even more important in the future for writing on PCs, tablets, or cell phone. Writing text messages is already facilitated by autocomplete functions with an immense vocabulary and pictograms. Why shouldn’t similar support mechanisms extend to more formal writing tasks? Further considerations in this area still need to be developed and remain open research questions: • Which text types used in professional writing remain relevant in the everyday working life of commercial professions in the age of Industry 4.0? • In what ways will computers take over the writing of business correspond‐ ence or private and semi-private letters in the future? • What new fields of research can German didactics investigate in the context of Industry 4.0? • How will the teaching of writing at general and vocational schools in Germany, the EU, and globally evolve in the age of Industry 4.0? • What challenges and opportunities will Industry 4.0 present for students in the area of writing in the future? Writing on a PC, in particular, benefits many types of writers. In German schools, class tests and examinations still largely require writing by hand. However, this does not suit writers who create non-linear texts like a puzzle. From my point of view, computers can significantly support standardized writing. Users could benefit from templates, text modules, and macros, especially in the area of business correspondence. A number of texts such as offers or complaints can 5.3 Outlook 223 <?page no="224"?> be composed quickly for customers. In large companies, internal writing styles aligned with corporate identity and procedures based on corporate behavior are often already standardized, leaving little room for individual employees to formulate their own sentences. Business correspondence is subject to economic pressure in terms of time and the rapid onboarding of employees. However, professional writing also helps to mentally sort and develop oneself to gain new insights, solely through the process of writing. It should remain in human hands to use, adapt, or even break away from standardized patterns, to be creative or subversive, if both the sender (writer) and recipient (e.g., consumer, business partner, future employer) of a message (letter) accept this. Even today, application letters are not only sent by post, but also submitted via e-mail or uploaded to company intranet portals. Application documents can, for example, be uploaded to the employment agency and made visible to potential employers on the agency’s platform. Nevertheless, the cover letter must still be submitted during the application process and remains with the company to which one is applying, and is archived. On the contrary, the unsolicited offer follow a slightly different logic. Some companies still format these on A4 paper and send them to potential clients. Banks, car dealerships, hotels, and travel agencies often rely on the core structure of an unsolicited offer. These documents may be visually enhanced with images and resemble a flyer when opened, but many of the core textual elements of the traditional offer remain intact. The studies conducted so far on professional writing before and during the training period still show a need for further research, although (job-oriented) writing skills are one of the eight key competences identified by the European Commission and part of lifelong learning for EU citizens (Settelmeyer & Widera, 2018; OECD, 2013: 143; European Union, 2006). Skilled workers with strong writing skills are essential for the labor market (Klein & Schöpper-Grabe, 2018; Philipp, 2018; Jakobs, 2005). By the end of lower secondary school, young people should be able to compose texts in a way that is appropriate to the target audience, comprehensible, coherent, and well-structured (Bundesagentur für Arbeit, 2006). Moreover, writing skills in vocational education and training should be increasingly promoted as a communicative action and cross-cutting educational goal (Nill, 2018; Efing, 2018; Hoefele & Konstantinidou, 2018; Hoefele et al., 2015; Baumann & Siemon, 2013; Kimmelmann, 2013; Efing, 2013; Giera, 2011). “Writing plays a key role in academic success and in an individual’s professional and social development” (Fidalgo et al., 2017: 3). German didactics must take a closer look at this academic field. 224 5 Discussion <?page no="225"?> Strong research designs for interventions with pre-, post-, and follow-up tests, including randomizations with a larger number of test subjects will be necessary for future studies. Particularly in this context, profiled writing taskssituated within realistic, action-oriented contexts are essential. In some federal states, such as Bavaria or Lower Saxony, students may leave school at the end of ninth grade. A status quo survey, e.g., in the VERA tests, should therefore be carried out in all ninth grades and extended into upper secondary schools throughout Germany. This would help discover discrepancies and strengths in students’ vocational writing abilities and could be followed by targeted interventions at general education schools, vocational schools, and upper secondary schools. High school students should also be familiar with writing professional texts, as they are they are involved in tasks like making declarations of intent, signing contracts, and dealing with official procedures. Knowledge of the associated text types such as the offer or complaint is essential. Every pupil should be prepared for writing job applications. Although this is currently prescribed in the curriculum at general education schools in Germany, emphasis should be placed on the actual writing process, incorporated not only into German lessons, but also in business lessons, career-oriented programs. It is not enough for extracurricular experts to give a lecture on job applications or for the teacher to spend only one or two hours on the application letter. Supporting writing in and outside the classroom by setting up study groups or writing studios in well-equipped computer rooms with color printers, still lacking in many schools, is relevant because writing applications requires time that is not always available during regular class hours. It also requires individual time with each student, which is not feasible in a full classroom setting. Furthermore, writing applications is also particularly relevant at upper secondary level. Although it is not always specified in the curriculum, it is obvious that trainees must prepare job applications within the training period for positions after completing their vocational training. It is also obvious that not every secondary school graduates pursue higher education, but many aspire to do apprenticeships or enter workforce, and must also apply for a job in writing. Therefore, upper secondary schools should also prepare students adequately for this and support them with writing studios. Furthermore, this requires coordinated efforts of all teachers at schools. Writing can play a role not only in German, but is expected in almost all subjects. Nevertheless, the task of teaching writing skills tends to be limited to the subject of German. However, this should change and be anchored across the curriculum. Norway serves as a compelling example, having implemented the Wheel of Writing incorporating writing instruction across subjects (Berge et al., 2016). An increase in statistically 5.3 Outlook 225 <?page no="226"?> meaningful studies is necessary in this field. Due to the heterogeneity of the student body at general education schools and at upper secondary level, it makes sense to create writing profiles for various cohorts. At the same time, the need for precise diagnostic tools that can also be used by normal teachers is becoming more urgent due to the fact that cultural developments (migration, parallel societies, both parents working, precarious employment) and, more recently, the concept of inclusion of pupils with disabilities in ‘normal’ class groups is increasing the ‘spread’ of the learning group in terms of their level of development and performance. The result must be a clear internal differentiation and individualization of lessons (Eichler, 2017: 146, translated by WKG). Furthermore, competence structure models for writing are not only important in German, but in all subjects (Neumann & Giera, 2018, translated by WKG). The ongoing process of inclusion at all lower and upper secondary schools in particular will make the student body increasingly heterogeneous in the coming years. The focus will be shift toward the individual learner rather than the class. Competence profiles of individual pupils will be required. It will therefore hardly be possible to continue with predominantly traditional frontal teaching. In terms of language didactics, Knopp and Becker-Mrotzek (2018: 95) call for further research into the areas of “theoretical concepts” (ibid.), “didactic-methodological concepts” (ibid.), and “empirical studies” (ibid.) in inclusive German lessons. This project requires more research with interdisciplinary teams from the fields of German didactics, text linguistics, business didactics, or psychology with a focus on writing, who also contribute their internationally networked contacts and knowledge. This endeavour also requires funding and support from the federal government in order to promote writing skills as a basic and participatory skill for the integration of all German citizens into our democratic society and the economic market. 226 5 Discussion <?page no="227"?> 5.4 EPILOGUE “Excellent research, including and especially good application-oriented research, cannot be bought at auction from cheap Jakobs, is not available from the research aldi, and does not come for free” (Rost, 2007: 242, translated by WKG). Since 2011, I have been teaching all types of texts with SRSD as a support approach as a German teacher at lower secondary level. There are more and more inclusion classes at secondary schools. It therefore helps to structure lessons according to the SRSD approach in order to be able to respond to all pupils. When there are 20 or more pupils in a PC room, this methodical framework ensures that all pupils know the lesson steps and don’t have to wait for each other. I occasionally carry out the modeling phase for all students. The writing lessons are perceived as very positive by the pupils, as they are allowed to work on the PC and have a freer hand in deciding which step they carry out, when, and with whom. I can therefore only recommend the writing project from my point of view and advise integrating other types of text using motivating writing tasks based on this principle. For me, it is a matter close to my heart to show pupils as a writing role model that writing is fun - even at lower secondary level. On the one hand, this study pushed me to my limits, but on the other, it was an instructive time for me overall. It was particularly important to me to organize my own doctoral project and to incorporate ideas that had been buzzing around in my head for a long time. The organization and planning were in my hands, from the initial discussions with the head teachers to the individual lessons. It certainly wasn’t always easy and I had to learn a lot in a short space of time, such as how to write an application to the education authority. I really enjoyed the lessons in the very different classes, although I initially assumed that I would not be accepted as a teacher straight away, but this turned out to be a fallacy. Another challenge was evaluating the data. At this point, it was no longer possible for me to work alone. I was able to have a team of students process some of the data. Their critical questions in particular gave my work new impetus. Revising my thoughts and my work was omnipresent. In addition to my actual job as a full-time teacher, organizing the writing phases was another hurdle. For months, I got up at 4: 44 in the morning (at least the number motivated me) to write before starting work and waking up my family. 5.4 Epilogue 227 <?page no="228"?> It was hard, but led to a writing routine that I can recommend to anyone doing a PhD with a child. The most important thing, however, was the exchange with my closest advisor - my counselor. She turned my thoughts upside down, put them right again and always gave me a positive boost of energy after the discussions. The search for regular appointments, especially when you’re not doing a doctorate directly at the university, was my motivation to work regularly at home and to finish the doctorate with all its ups and downs after five years. Although this thesis was initially created by me alone as a vision during my studies, as an idea during the school internships and ultimately as a goal following my Master’s degree, and although it was exhausting at many times due to the fact that I was mainly planning, implementing, and evaluating it alone, and sometimes also very lonely at my desk, I had motivating and strength-building supporters whom I would like to thank very much. Thanks to them, I was able to take the long road to my doctorate. First of all, I would like to thank the head teachers, teachers, pupils, and parents who immediately put their trust in me for the project. Thanks also to my students who supported me in guessing the texts and the student interviews. I would like to thank my research colloquium and my colleagues at the university and school who were able to provide valuable advice. Finally, my biggest thanks go to my thesis supervisors, because without critical reflection and advice I would have overlooked some points. 228 5 Discussion <?page no="229"?> REFERENCES Abraham, U./ Frederking, V. (2017). Deutsch und Deutschdidaktik. In: Bayrhuber, H./ Abraham, U./ Frederking, V./ Jank, W./ Rothgangel, M./ Vollmer, H. J.: Auf dem Weg zu einer Allgemeinen Fachdidaktik. Münster: Waxmann, 53-73. Adamzik, K. (2004). Textlinguistik. Eine einführende Darstellung. Tübingen: Niemeyer. Adamzik, K. (2008). Textsorten und ihre Beschreibung. In: Janich, Nina (Hrsg.): Textlinguistik. 15 Einführungen. Tübingen: Narr, 145-175. Akoun, A./ Pailleau, I. (2014). Besser Lernen mit Positiver Pädagogik. Der Ratgeber für Lehrer, Eltern und Schüler. München: mvg. Antos, G. (2008). Schriftliche Textproduktion: Formulieren als Problemlösung. In: Janich, Nina (Hrsg.): Textlinguistik. 15 Einführungen. Tübingen: Narr, 237-254. Arbeitsgemeinschaft für Kinder- und Jugendhilfe (AGJ) (2017). Deutscher Kinder- und Jungend(-hilfe) Monitor 2017. Abrufbar unter: https: / / www.agj.de/ fileadmin/ files/ arb eitsfelder/ Monitor_Dt_Kinder-jugendhilfe_FINAL.pdf (Zugriff: 08.06.2018). Arnold, R./ Lipsmaier, A./ Münk, D. (1995). Berufspädagogische Kategorien didaktischen Handelns. In: Arnold, R./ Lipsmaier, A. (Hrsg.): Handbuch der Berufsbildung. Opladen: Leske + Budrich, 13-28. Atwell, N. (1987). In the middle. Reading, writing, and learning from adolescents. Portsmouth, NH. Heinemann. Aufgabenstellung für kaufmännische Abschluss- und Zwischenprüfungen (AkA) (2011). Wir über uns. Abrufbar unter: http: / / www.ihk-aka.de/ wir-ueber-uns (Zugriff: 27.05.2011). Augst, G./ Disselhoff, K./ Henrich, A./ Pohl, T./ Völzing, P.-L. (2007). Text-Sorten-Kompe‐ tenz. Eine echte Longitudinalstudie zur Entwicklung der Textkompetenz im Grund‐ schulalter. Frankfurt a. M.: Peter Lang. Austin, J. L. (1979). Zur Theorie der Sprechakte [How to do things with words]. Deutsche Bearbeitung von Eike von Savigny. 2. Auflage. Stuttgart: Reclam. Bachmann, T. (2002). Kohäsion und Kohärenz: Indikatoren für Schreibentwicklung. Zum Aufbau kohärenzstiftender Strukturen in instruktiven Texten von Kindern und Jugendlichen. Innsbruck: StudienVerlag. Bachmann, T. (2014). Texte produzieren: Schreiben als soziale Praxis. In: Bachmann, T./ Feilke, H. (Hrsg.): Werkzeuge des Schreibens. Beiträge zu einer Didaktik der Textprozeduren. Stuttgart: Fillibach bei Klett, 35-61. Bachmann, T./ Becker-Mrotzek, M. (2010). Schreibaufgaben situieren und profilieren. In: Pohl, T./ Steinhoff, T. (Hrsg.): Textformen als Lernformen. Kölner Beiträge zur Sprachdidaktik (7). Duisburg: Gilles & Francke, 191-210. <?page no="230"?> Bachmann, T./ Becker-Mrotzek, M. (2017). Schreibkompetenz und Textproduktion mod‐ ellieren. In: Becker-Mrotzek, M./ Grabowski, J./ Steinhoff, T. (Hrsg.): Forschungshand‐ buch empirische Schreibdidaktik. Münster, New York: Waxmann, 24-53. Baddely, A. D. (1986). Working Memory. Oxford: Clarendon. Baethge, M (2008). Das berufliche Bildungswesen in Deutschland am Beginn des 21. Jahr‐ hunderts. In: Cortina, K. S. / Baumert, J. / Leschinsky, A. / Mayer, K. U. / Trommer, L. (Hrsg.): Das Bildungswesen in der Bundesrepublik Deutschland. Reinbek: Rowohlt. Bandura, A. (1977). Social learning theory. Englewood Cliffs, New Jersy: Prentice Hall. Basler, O./ Schulz, H. (1981). Deutsches Fremdwörterbuch. Berlin: de Gryter. Baumann, K./ Siemon, J. (2013). Wie viel schriftsprachliche Fähigkeit ist für eine er‐ folgreiche Berufsausbildung erforderlich? . In: Die berufsbildende Schule (BbSch), 65.-Jahrgang, Heft 10, 285-288. Baumann, K. (2014). „Man muss schon ein bisschen mit dem Schreiben zurechtkommen! “. Eine Studie zu den Schreibfähigkeiten von Auszubildenden im unteren beruflichen Ausbildungssegment im Kontext von Ausbildungsreife. Paderborn: Eusl. Baumert, A. (2008). Professionell texten. Grundlagen, Tipps und Techniken. 2. Auflage. München: dtv. Baurmann, J./ Pohl, T. (2011). Schreiben - Text verfassen. In: Bremerich-vos, A./ Granzer, D./ Behrens, U./ Köller, O. (Hrsg.): Bildungsstandards für die Grundschule. Deutsch konkret. 3. Auflage. Berlin: Cornelsen, 75-103. Becker-Mrotzek, M./ Böttcher, I. (2006). Schreibkompetenz entwickeln und beurteilen. Praxishandbuch für die Sekundarstufe I und II. Berlin: Cornelsen-Scriptor. Becker-Mrotzek, M./ Böttcher, I. (2012). Schreibkompetenz entwickeln und beurteilen. 4., und überarbeitete Neuauflage. Berlin: Cornelsen. Beer-Scharwächter, A. (2008). Weil Deutschland alle Schüler braucht! Ein Forschung‐ sprojekt zwischen Pädagogischer Psychologie und Deutschdidaktik nach PISA. Dis‐ sertation. Baltmannsweiler: Schneider Verlag Hohengehren. Behrens, U. (2014). Zuhörfähigkeit testen. In: Neumann, A./ Mahler, I. (Hrsg.): Empirische Methoden der Deutschdidaktik: audio- und videografierende Unterrichtsforschung. Baltmannsweiler: Schneider Verlag Hohengehren, 38-52. Biedebach, W. (2006): Der Modellversuch „Vocational Literacy (VOLI) - Methodische und sprachliche Kompetenzen in der beruflichen Bildung“. Konzeption - Erfahrungen - bisherige Ergebnisse. In: Efing, C./ Janich, N. (Hrsg.). Förderung der berufsbezogenen Sprachkompetenz. Befunde und Perspektiven. Paderborn: Eusl, 15-31. Berge, K. L./ Evensen, L. S./ Thygesen, R. (2016). The Wheel of Writing: a model of the writing domain fort he teaching and assessing of writing as a key cometency, The Curriculum Journal, 27: 2, 172-189.Abrufbar unter: https: / / www.tandfonline.com/ doi / abs/ 10.1080/ 09585176.2015.1129980 (Zugriff: 13.05.2018). 230 References <?page no="231"?> Bereiter, C. (1980). Development in writing. In: Gregg, L./ Steinberg, E. R. (Hrsg.): Cognitive Processes in Writing. Hillsdale. New Jersey: Erlbaum, 73-93. Bereiter, C./ Scardamalia, M. (1987). The Psychology of Written Composition. Hillsdale, New Jersey: Erlbaum. Berninger, V./ Cartwright, A./ Yates, C./ Swanson, H. L./ Abbott, R. (1994). Developmental skills related to writing and reading acquisition in the intermediate grades: Shared an unique variance. Reading and Writing: An Interdisciplinary Journal, 6, 161-196. Berufsbildungsgesetz (BBiG) (2005). Berufsbildungsgesetz (BBiG). (https: / / www.gesetze -im-internet.de/ bbig_2005/ BBiG.pdf, Zugriff: 06.08.2018). Beaugrande, R.-A. de/ Dressler, W. U. (1981): Einführung in die Textlinguistik. Tübingen: De Gryter. Birkenbihl, V. F. (2003). Stichwort Schule: Trotz Schule lernen! . 16. Auflage. München: mvg. Birkenbihl, V. F. (2010). Trotzdem Lehren. 4. Aufl. München: mvg. Böhme, K./ Schipolowski, S./ Canz, T./ Krelle, M./ Bremerich-Vos, A. (2017). Kompetenzstu‐ fenmodelle im Bereich Schreiben. In: Becker-Mrotzek, M./ Grabowski, J./ Steinhoff, T. (Hrsg.): Forschungshandbuch empirische Schreibdidaktik. Münster, New York: Waxmann, 55-74. Böttcher, I./ Wagner, M. (1992). Kreative Texte bearbeiten. In: Praxis Deutsch, H. 119, 24-26. Braaksma, M./ Rijlaarsdam, G./ Bergh, H. van den. Oberservational Learning in Argu‐ mentative Writing. In: Torrance, M./ Alamargot, D./ Castelló, M./ Ganier, F./ Kruse, O.Tolchinksy, L./ Van Waes, L. (Hrsg.) (2012): Learning to Write Effectively: Current Trends in European Research. Amsterdam: Emerald, 123-126. Brinker, K. (2001). Linguistische Textanalyse. Eine Einführung in die Grundbegriffe und Methoden. 5. Auflage. Berlin: Erich Schmidt. Brinker, K. (2010). Linguistische Textanalyse. Eine Einführung in die Grundbegriffe und Methoden. 7., durchgesehene Auflage. Berlin: Erich Schmidt. Brommer, S. (2018). Textsortenspezifische sprachliche Variation ermitteln. Muster und Musterhaftigkeit aus korpuslinguistischer, texlinguistischer sowie stilistischer Per‐ spektive. In: Adamzik, K./ Mateusz, M. (Hrsg.): VARIATIONslinguistik trifft TEXTlin‐ guistik. Tübingen. Narr, 61-80. Bühler, K. (1934, 1999). Sprachtheorie. Die Darstellungsfunktion der Sprache. 3. Auflage. Jena: Lucius & Lucius. Bühner, M. (2011). Einführung in die Test- und Fragebogenkonstruktion. 3., aktualisierte und erweiterte Auflage. München: Pearson Studium. Bundesagentur für Arbeit (Hrsg.) (2006). Nationaler Pakt für Ausbdilung und Fachk‐ räftenachwuchs. Kriterienkatalog zur Ausbildungsreife.Abrufbar unter: http: / / www. arbeitsagentur.de/ zentraler-Content/ Veroeffentlichungen/ Ausbildung/ Kriterienkata‐ log-zur-Ausbildungsreife.pdf (Zugriff: 19.06.2012). References 231 <?page no="232"?> Bundesagentur für Arbeit (Hrsg.) (2017). Ausbildungsstellenmarkt - Die aktuellen Entwiclungen im Berichtsjahr 2016/ 2017 in Kürze. Abrufbar unter: https: / / sta‐ tistik. arbeitsagentur. de/ Navigation/ Statistik/ Statistik-nach-Themen/ Ausbildungs‐ stellenmarkt/ Ausbildungsstellenmarkt-Nav.html (Stand: 08.06.2018). Bundesagentur für Arbeit (Hrsg.) (2018). Ausbildungsstellenmarkt - Die aktuellen Entwiclungen im Berichtsjahr 2017/ 2018 in Kürze. Abrufbar unter: https: / / sta‐ tistik. arbeitsagentur. de/ Navigation/ Statistik/ Statistik-nach-Themen/ Ausbildungs‐ stellenmarkt/ Ausbildungsstellenmarkt-Nav.html (Stand: 08.06.2018). Bundesinstitut für Berufsbildung (BiBB) (Hrsg.) (2014). Top 10 der dualen Ausbildungs‐ berufe nach Neuabschlüssen zum 30. September 2014. Abrufbar unter: https: / / www. bibb.de/ de/ 25322.php (Zugriff: 08.06.2018). Bundesinstitut für Berufsbildung (BiBB) (Hrsg.) (2016). Datenreport zum Berufsbildungs‐ bericht 2016. Informationen und Analysen zur Entwicklung der beruflichen Bildung. Bielefeld: Bertelsmann. Bundesinstitut für Berufsbildung (BiBB) (Hrsg.) (2017). Datenreport zum Berufsbildungs‐ bericht 2017. Informationen und Analysen zur Entwicklung der beruflichen Bildung. Bielefeld: Bertelsmann. Bundesinstitut für Berufsbildung (BiBB) (Hrsg.) (2018). Datenreport zum Berufsbildungs‐ bericht 2018. Informationen und Analysen zur Entwicklung der beruflichen Bildung. Bonn: Budrich. Bundesministerium der Justiz (BMJ) (1998). Anlage (zu § 9) Ausbildungsrahmenplan für die Berufsausbildung im Gastgewerbe. Bundesministerium für Bildung und Forschung (BMBF) (Hrsg.) (2010a). Ausbildung & Beruf. Rechte und Pflichten während der Berufsausbildung. Abrufbar unter: http: / / w ww.bmbf.de/ de/ 550.php (Zugriff: 23.05.2011). Bundesministerium für Bildung und Forschung (BMBF) (2010b). Aus- und Fortbildung‐ sordnungen. Abrufbar unter: http: / / www.bmbf.de/ de/ 550.php (Zugriff: 23.05.2011). Calkins, L. (1986). The art of teaching writing. Portsmouth, NH: Heinemann. Calmbach, M./ Thomas, P./ Borchard, I./ Flaig, B. (2017). Wie ticken Jugendliche: Lebens‐ welten von Jugendlichen im Alter von 14 bis 17 Jahren. Düsseldorf: Haus Altenberg. Castells, N./ Neumann, A./ Solé, I. (2014). Advantages and difficulties of conducting thinking-aloud protocols in the school setting. In: Neumann, A./ Mahler, I. (Hrsg.) (2014): Empirische Methoden der Deutschdidaktik: audio- und videografierende Unterrichtsforschung. Baltmannsweiler: Schneider Verlag Hohengehren, 204-221. Chassa, P. (2016). Schule von morgen. Dream the future. Staffel 1, Folge 5. Abrufbar unter: https: / / www.Chassa.de/ Transport-von-morgen/ dp/ B01N5K3YDL/ ref=sr_1_1? s=insta nt-video&ie=UTF8&qid=1490159278&sr=1-1&keywords=dream+the+future (Zugriff: 21.03.2017). 232 References <?page no="233"?> Chenoweth, N. A./ Hayes, J. R. (2001). Fluency in writing: Generating text in L1 and L2. Written Communication, 18 (1), 80-98. Cleveland, K. P. (2013). Jungen unterrichten. Der 6-Punkte-Plan gegen Schulversagen. Weinheim: Beltz. Couzijn, M./ Rijlaarsdam, G. (1996). Learning to write by reader observation and written feedback. In: Rijlaarsdam, G./ Van den Bergh, H./ Couzijn, M. (Hrsg.): Effective Teach‐ ing and Learning of Writing. Current Trends in Research. Amsterdam: University Press, 224-252. Couzijn, M./ Rijlaarsdam, G. (2004). Learning to Read and Write Argumentative Text by Observation of Peer Learners. Abrufbar unter: https: / / www.researchgate.net/ pu blication/ 225236690_Learning_to_Read_and_Write_Argumentative_Text_by_Obser‐ vation_of_Peer_Learners (Zugriff: 06.08.2018). In: Rijlaarsdam, G./ Bergh, H. van den/ Couzijn, M. (Hrsg.): Effective Learning and Teaching of Writing. A Handbook of Writing in Education, 241-258. Czycholl, R./ Ebner, H. G. (1995). Handlungsorientierung in der Berufsbildung. In: Arnold, R./ Lipsmeier, A. (Hrsg.): Handbuch der Berufsbildung. Opladen: Leske + Budrich, 39-49. Dehn, M./ Merklinger, D./ Schüler, L. (2011). Texte und Kontexte. Schreiben als kulturelle Tätigkeit in der Grundschule. Seelze: Kallmeyer. Deutscher Industrie- und Handelskammertag (Hrsg.) (2012). Ausbildung 2012. Ergeb‐ nisse einer DIHK-Online-Unternehmensbefragung. In: www.dihk.de/ ressourcen/ …/ d ihk-ausbildungsumfrage-2012.pdf, Zugriff 04.03.2013. Berlin, Brüssel. Dijk, T. A. van (2008). Discourse and Context. A sociocognitive approach. Cambridge: University Press. Dürscheid, C. (2007). Texte aus kommunikativ-pragmatischer Sicht. In: Zeitschrift für Angewandte Linguistik (46), 3-18. Dweck, C. S. (1986). Motivational processes affecting learning. In: American Psychologist, 41(10), 1040-1048. Abrufbar unter: http: / / dx.doi.org/ 10.1037/ 0003-066X.41.10.1040 (Stand: 14.08.2018). Efing, C. (2006). „Viele sind nicht in der Lage, diese schwarzen Symbole da lebendig zu machen.“ - Befunde empirischer Erhebungen zur Sprachkompetenz hessischer Berufsschüler. In: Efing, C./ Janich, N. (Hrsg.). Förderung der berufsbezogenen Sprach‐ kompetenz. Befunde und Perspektiven. Paderborn: Eusl, 33-62. Efing, C./ Janich, N. (2006). Einleitung der Herausgeber. In: Efing, C./ Janich, N. (Hg). Förderung der berufsbezogenen Sprachkompetenz. Befunde und Perspektiven. Pader‐ born: Eusl, 7-11. Efing, C. (2008). „Aber was halt schon schwer war, war, wo wir selber schreiben sollten.“ Defizite und Förderbedarf in der Schreibkompetenz hessischer Berufsschüler. References 233 <?page no="234"?> In: Jakobs, E.-M./ Lehnen, K. (Hrsg.): Berufliches Schreiben. Ausbildung, Training, Coaching. Frankfurt a. M.: Peter Lang, 17-34. Efing, C. (2010). Kommunikative Anforderungen an Auszubildende in der Industrie. In: Fachsprache, 4 -19. Efing, C./ Häußler, M. (2011). Was soll der Deutschunterricht an haupt- und Realschu‐ len vermitteln? - Empirisch basierte Vorschläge für eine Ausbildungsvorbereitung zwischen zweckfreier und zweckgerichteter Bildung. In: bwp@Spezial 5 - HT 2011. Abrufbar unter: https: / / www.bwpat.de/ ht2011/ ft18/ efing_haeussler_ft18-ht2011.pdf (Zugriff: 25.08.2018). Efing, C. (2012). Sprachliche oder kommunikative Fähigkeiten - was ist der Unterschied und was wird in der Ausbildung verlangt. In: Berufsbildung in Wissenschaft und Praxis. Sprache und Beruf. H. 20155, 6-9. Efing, C. (2013a). Ausbildungsvorbereitender Deutschunterricht an allgemeinbildenden Schulen? -Legitimation und Definition. In: Efing, C. (Hrsg.): Ausbildungsvorbereitung im Deutschunterricht der Sekundarstufe I. Sprachlich-kommunikative Facetten von “Ausbildungsfähigkeit”. Frankfurt/ Main: Peter Lang, S.-11-38. Efing, C. (2013b). Sprachlich-kommunikative Anforderungen in der betrieblichen Aus‐ bildung. In: Efing, C. (Hrsg.): Ausbildungsvorbereitung im Deutschunterricht der Se‐ kundarstufe I. Die sprachlich-kommunikativen Facetten von „Ausbildungsfähigkeit“. Frankfurt am Main: Peter Lang, 123-145. Efing, C. (2013c). (Wie) Bereitet der bisherige Deutschunterricht auf die Ausbildung vor? Eine Schulbuchanalyse. In: Efing, C. (Hrsg.): Ausbildungsvorbereitung im Deut‐ schunterricht der Sekundarstufe I. Die sprachlich-kommunikativen Facetten von „Ausbildungsfähigkeit“. Frankfurt am Main: Peter Lang, 239-256. Efing, C. (2015). Berufsweltbezogene kommunikative Kompetenz in Erst- und Fremdspr‐ ache - Vorschlag einer Modellierung Efing, C. (Hrsg.): Sprache und Kommunikation in der beruflichen Bildung. Modellierung - Anforderungen - Förderung. Frankfurt am Main: Peter Lang, 17-46. Efing, C. (2018). Formulare als textuelle Herausforderung in der beruflichen Ausbildung. In: Efing, C./ Kiefer, K.-H. (Hrsg.): Sprache und Kommunikation in der beruflichen Aus- und Weiterbildung. Ein interdisziplinäres Handbuch. Tübingen: Narr, 373-382. Ehlich, K. (1994). Funktion und Struktur schriftlicher Kommunikation, In: Günther/ Ludwig (Hrsg.): 1. Halbbd./ HSK 10.1, 18-41. Eichler, W./ Henze, W. (1998). Sprachwissenschaft und Sprachdidaktik. In: Lange, Günter/ Neumann, Karl/ Ziesenis, Werner (Hrsg.): Taschenbuch des Deutschunterrichts. Grundfragen und Praxis der Sprach- und Literaturdidaktik. Band-1: Grundlagen -Sprachdidaktik - Mediendidaktik. 8. Aufl. Baltmannsweiler: Schneider,101-123. Eichler, W. (2017). Zur sprachkognitiven Diagnostik anhand von Leistungen im Text-Ver‐ fassen in der Grundschule. In: Siekmann, K./ Toro, I. C. del/ Hoffmann-Erz, R. (Hrsg.): 234 References <?page no="235"?> Schriftsprachliche Kompetenzen in Theorie und Praxis. Festschrift für Günther Thomé. Tübingen: Stauffenberg, 145-164. Electric Paper Evaluationssysteme GmbH (2018). EvaSys. Abrufbar unter: https: / / www. evasys.de/ evasys.html (Zugriff: 28.07.2018). Europäische Union (2018): Allgemeine und berufliche Bildung, Jugend. In: Abrufbar un‐ ter: https: / / europa.eu/ european-union/ topics/ education-training-youth_de (Zugriff: 27.07.2018). European Union (2006). Lebensbegleitendes Lernen - Schlüsselkompetenzen. Abrufbar unter: https: / / eur-lex.europa.eu/ legal-content/ DE/ TXT/ ? uri=LEGISSUM%3Ac11090 (Zugriff: 31.07.2018). European Research Network on Learning to Write Effectively (ERN-LWE) (2011). Results of 2nd Workshop 2011. Potsdam Triangel. Potsdam: [unveröffentlicht]. Fachstelle Berufsbezogenes Deutsch (o. J.). Fachdiskussion. Abrufbar unter: http: / / www. deutsch-am-arbeitsplatz.de/ fachdiskussion/ b/ berufsbezogenes-deutsch.html (Zugriff: 19.04.2017). Feilke, H. (2010). „Aller guten Dinge sind drei“ - Überlegungen zu Textroutinen & literalen Prozeduren. In: Fritz, G. (Hrsg.): Fest-Platte für Gerd Fritz. Abrufbar unter: ht tp: / / www.festschrift-gerd-fritz.de/ files/ feilke_2010_literale-prozeduren-und-textrou‐ tinen.pdf (Zugriff: 06.05.2013). Feilke, H. (2011). „Bitte im ganzen Satz! “. Überlegungen zur Schulsprache. Vortrag die Sommerschule Oldenburg. Abrufbar unter: https: / / www.uni-giessen.de/ fbz/ fb0 5/ dies/ veranstaltungen/ tagungen/ oldenburg2011/ abstracts/ praesi-feilke.pdf (Zugriff: 01.05.2018). Feilke, H. (2012a). Bildungssprachliche Kompetenzen - fördern und enwickeln. In: Praxis Deutsch (233), 4-13. Feilke, H. (2012b). Schulsprache - Wie Schule Sprache macht. In: Günther, S./ Imo, W./ Meer, D./ Schneider, J. G. (Hrsg.): Kommunikation und Öffentlichkeit. Sprachwis‐ senschaftliche Potenziale zwischen Empirie und Norm. Berlin: de Gryter, 149-175. Feilke, H. (2012c). Was sind Textroutinen? - Zur Theorie und Methodik des Forschungs‐ feldes. In: feolke, H./ Lehnen, K. (Hrsg.): Schreib- und Textroutinen. Frankfurt am Main: Peter lang, S.-1-31. Feilke, H. (2014). Argumente für eine Didaktik der Textprozeduren. In: Bachmann, T./ Feilke, H. (Hrsg.): Werkzeuge des Schreibens. Beiträge zu einer Didaktik der Textprozeduren. Stuttgart: Fillibach bei Klett, 11-34. Feilke, H. (2017). Schreibdidaktische Konzepte. In: Becker-Mrotzek, M./ Grabowski, J./ Steinhoff, T. (Hrsg.): Forschungshandbuch empirische Schreibdidaktik. Münster, New York: Waxmann, 153-171. Fidalgo, R./ Torrance, M./ Robledo, P./ García, J.-N. (2012). Evaluating Cognitive Self-Reg‐ ulation Instruction for Developing Students’ Writing Competence. In: Torrance, M./ References 235 <?page no="236"?> Alamargot, D./ Castelló, M./ Ganier, F./ Kruse, O./ Tolchinksy, L./ Van Waes, L. (Hrsg.): Learning to Write Effectively: Current Trends in European Research. Amsterdam: Emerald, 75-77. Fidalgo, R,/ Harris, K. R./ Braaksma, M. (2017). Design Principles for Teaching Effective Writing: An Introduction. In: Fidalgo, R,/ Harris, K. R./ Braaksma, M. (Hrsg.): Studies in Writing Series: Vol. 34. Design Principles for Teaching Effective Writing. Leiden: Brill, 3-12. Fix, U. (2008). Text und Textlinguistik. In: Janich, N. (Hrsg.): Textlinguistik. 15 Einfüh‐ rungen. Tübingen: Narr, 15-34. Fix, M. (2000). Textrevisionen in der Schule. Prozessorientierte Schreibdidaktik zwischen Instruktion und Selbststeuerung. Empirische Untersuchungen in achten Klassen. Baltmannsweiler: Schneider Verlag Hohengehren. Fix, M. (2008). Texte schreiben. Schreibprozesse im Deutschunterricht. 2. Aufl. Paderborn: Schöningh. Fleuchaus, I. (2004). Kommunikative Kompetenzen von Auszubildenden in der berufli‐ chen Ausbildung. Ausprägungen, Förderung und Relevanz im Urteil von Ausbildern, Lehrern und Auszubildenden. Dissertation. Hamburg: Dr. Kovač. Flower, L. S./ Hayes, J. R. (1980). The Dynamics of Composing: Making Plans and Juggling Constraints. In: Gregg, L. W./ Steinberg, E. R. (Hrsg.): Cognitive Process in Writing. Hillsdale, New Jersey: Erlbaum, 31-50. Foxworth, L./ Mason, L. H. (2017). Writing-to-Learn Instruction that Works. In: Fidalgo, R,/ Harris, K. R./ Braaksma, M. (Hrsg.): Studies in Writing Series: Vol. 34. Design Principles for Teaching Effective Writing. Leiden: Brill, 66-86. Fritzsche, J. (2003). Schriftlicher Sprachgebrauch. In: Lange, Günter/ Neumann, Karl/ Ziesenis, Werner (Hrsg.): Taschenbuch des Deutschunterrichts. Grundfragen und Praxis der Sprach- und Literaturdidaktik. Band-1: Grundlagen - Sprachdidaktik - Mediendidaktik. 8. Auflage. Baltmannsweiler: Schneider, 201-225. Fürstenau, S. (2009). Lernen und Lehren in heterogenen Gruppen. In: Fürstenau, S./ Gomalla, M. (Hrsg.): Migration und schulischer Wandel: Unterricht. Wiesbaden: VS: 61-84. Gansel, C./ Jürgens, F. (2008). Textgrammatische Ansätze. In: Janich, N. (Hrsg.): Textlin‐ guistik. 15 Einführungen. Tübingen: Narr, 55-83. Gewerkschaft Erziehung und Wissenschaft im Deutschen Gewerkschaftsbund Erziehung und Wissenschaft (2017). GEW 03/ 2017. Frankfurt am Main: Stamm Verlag. Gibbons, P. (2009). English Learners, Academic Literacy, and Thinking. Portsmouth, NH: Heinemann. Giera, W.-K. (2010). Berufliche Textsorten. Schreibanforderungen in der beruflichen Bildung. Bachelorarbeit. Lüneburg: Leuphana Universität [unveröffentlicht]. 236 References <?page no="237"?> Giera, W.-K./ Meyer, P.-L./ Süphke, J. (2012). Der Index für Inklusion und seine Derivate. In: Saldern, M. von (Hrsg.): Inklusion. Deutschland zwischen Gewohnheit und Men‐ schenrecht. Norderstedt: Books on Demand,145-164. Giera, W.-K. (2013). Schülerinnen Schüler mit und ohne Migrationshintergrund fördern: Ein außerschulisches Schreibprojekt nach dem SRSD-Modell von Steve Graham und Karen Harris in der 7. Klasse einer Realschule. Masterarbeit. Lüneburg: Leuphana Universität [unveröffentlicht]. Giera, W.-K./ Mahler, I./ Neumann, A. (2013). Wissenschaftliches Schreiben und Forschen fördern. Ein Schreibtutorium für Studienanfänger_innen des Faches Deutsch. In: Tschirpke, S./ Peters, N./ Liebetanz, F./ Kreitz, D. (Hrsg.): JoSch - Journal der Schreib‐ beratung 01/ 2013. Ausgabe 6, 38-47. Giera, W.-K. (2014). Evaluierung zweier kompetenzorientierter Beobachtungs- und Reflexionsaufgaben für einen Besuch der KZ-Gedenkstätte Neuengamme in einer neunten Realschulklasse im Rahmen der Politischen Bildung. 2. Staatsexamensarbeit. Verden: Studienseminar Verden. Giera, W.-K. (2020). Berufsorientierte Schreibkompetenz mithilfe von SRSD fördern: Evaluation eines schulischen Schreibprojekts (Doctoral dissertation). Universität Lüneburg. Narr Francke Attempto. https: / / d-nb.info/ 1203961464 Glaser, C. (2005). Förderung der Schreibkompetenz von Grundschülern. Effekte einer integrierten Vermittlung kognitiver Schreibstrategien und selbstregulatorischer Fer‐ tigkeiten. Dissertation. Abrufbar unter: http: / / opus.kobv.de/ ubp/ volltexte/ 2005/ 217/ p df/ GLASER.PDF (Zugriff: 19.11.2012). Potsdam: Universität Potsdam. Glaser, C./ Meyer, D. (2017). Evidenz-basierte Trainingsforschung. In: Becker-Mrotzek, M./ Grabowski, J./ Steinhoff, T. (Hrsg.): Forschungshandbuch empirische Schreibdidak‐ tik. Münster, New York: Waxmann, 369-383. Goldin, I. M./ Ashley, K. D. (2012). Eliciting formatice assessment in peer review. In: Journal of Writing Research 4(2). Abrufbar unter: http: / / www.jowr.org/ abstracts/ vol 4_2/ Goldin_Ashley_2012_4_2_abstract.html (Zugriff: 31.10.2017), 203-237. Goldin, I. M./ Ashley, K. D./ Schunn, C. D. (2012). Redesigning Educational Peer Review Interactionas Using Computer Tools. An Introduction. In: Journal of Writing Research 4(2). Abrufbar unter: http: / / www.jowr.org/ abstracts/ vol4_2/ Goldin_et_al_2012_4_2_ abstract.html (Zugriff: 31.10.2017), 111-119. Grabowski, J. (2011). Grundlagen empirischen Forschens: eine Auffrischung. DieS Som‐ merschule 2011 Präsentation. Oldenburg: unveröffentlicht. Grabowski, J. (2017). Anforderungen an Untersuchungsdesigns. In: Becker-Mrotzek, M./ Grabowski, J./ Steinhoff, T. (Hrsg.): Forschungshandbuch empirische Schreibdidaktik. Münster, New York: Waxmann, 315-334. References 237 <?page no="238"?> Graham, S./ Harris, K. R. (1987). Improving composition skills of inefficient learners with self-instructional strategy training. In: Aspen Publishers (Hrsg.): Topics in Language Disorders. H. 7, 66-77. Graham, S./ Harris, K. R. (1996). Self-regulation and strategy instruction for children who find writing and learning challenging. In: Levi, M./ Ransdell: (Hrsg.): The science of writing: Theoriey, methods, individual differences, and applications, 347-360. Graham, S./ Harris, K. R./ Troia, G. A. (1998). Writing and Self-Regulation: Cases from the Self-Regulated Strategy Development Model. In: Schunk, D. H./ Zimmerman, B. J. (Hrsg.): Self-regulated learning: From teaching to self-reflective practices, 20-41. Graham, S./ Harris, K. H./ Zito, J. (2005). Promoting Internal and External Validity: A Synergism of Laboratory-Like Experiments and Classroom-Based Self-Regulated Strategy Development Research. In: Empirical Methods for Evaluating Educational Interventions, Academic Press, 235-265. Graham, S./ Harris, K. R./ Mason, L. (2005). Improving the writing performance, knowl‐ edge, and self efficacy of struggeling young writers: The effects of self-regulated strategy development. In: Contemporary Educational Psychology (30), 207-241. Graham, S./ Perin, D. (2007). Writing next: Effective strategies to improve writing of adolescents in middle and high schools - A report to Carnegie Corporation of New York.Washington, DC: Alliance for Excellent Education. Graham, S./ Harris, K. R. (2017). Evidence-Based Writing Practices: A Meta-Analysis of Existing Meta-Analyses. In: Fidalgo, R./ Harris, K. R./ Braaksma, M. (Hrsg.): Studies in Writing Series: Vol. 34. Design Principles for Teaching Effective Writing. Leiden: Brill, 13-37. Graves, D. (1983). Writing. Teachers and children at work. Excter, NH: Heinemann. Grice, H. P. (1975). Logic and conservation. In: Cole, P./ Morgan, J. L. (Hrsg.): Syntax and Semantics. Bd.-3: Speech Acts. New York: Academic Press, 41-58. Große, E. U. (1976). Text und Kommunikation. Eine linguistische Einführung in die Funktionen der Texte. Stuttgart: Kohlhammer. Grün, K. (2013). Der Geschäftsbrief. Gestaltung von Schriftstücken nach DIN 5008, DIN 5009. 5., überarbeitete Aufl. Berlin: Beuth. Haarmann, H. (2007). Geschichte der Schrift. 3. Aufl. München: Beck. Hammill, D./ Larsen, S. (1996). Test of written language-3. Austin, TX: Pro-Ed. Handwerksordnung (HwO) (1998). Gesetz zur Ordnung des Handwerks (Handwerksord‐ nung). Abrufbar unter: https: / / www.gesetze-im-internet.de/ hwo/ HwO.pdf (Zugriff: 06.08.2018). Harris, K. R./ Santangelo, T./ Graham, S. (2010). Metacognition and Strategies Instruction in Writing. In: Waters, H. S./ Schneider, W. (Hrsg.): Metacognition, Strategy use & Instruction. New York: The Guilford Press, 226-256. 238 References <?page no="239"?> Harris, K. R./ Graham, S. (2017). Self-Regulated Strategy Development: Theoretical Bases, Critical Instructional Elements, and Future Research. In: Fidalgo, R,/ Harris, K. R./ Braaksma, M. (Hrsg.): Studies in Writing Series: Vol. 34. Design Principles for Teaching Effective Writing. Leiden: Brill, 119-151. Hayes, J. R./ Flower, L. (1980). Identifying the Organization of Writing Processes. In: Gregg, L. W./ Steinberg, E. R. (Hrsg.): Cognitive Processes in Writing. Hillsdale: Lawrence Erlbaum, 3-30. Hayes, J. R./ Flower, L./ Schriver, K. A./ Stratman, J. F./ Carey, L. (1987). Cognitive processes in revision. In: Rosenberg, S. (Hrsg.): Advances in applied psycho-linguistics. Cam‐ bridge: Cambridge University Press, 176-240. Hayes, J. R. (1996). A New Framework for Understanding Cognition and Affect in Writing. In: Levy, C. Michael / Ransdell, Sarah E. (Hrsg.): The Science of Writing: Theories, Methods, Individual Differences, and Applications. New York, London: Erlbaum, 1-27. Hayes, J. R. (2006). New Directions in Writing Theory. In: MacArthur, Charles A./ Graham, Steve/ Fitzgerald, Jill (Hrsg.): Handbook of Writing Research. New York, London: Guilford Press, 28-40. Hayes, J. R. (2012). Modelling and Remodelling Writing. In: Sage (Hrsg.): Written Communication. Abrufbar unter: http: / / wcx.sagepub.com/ content/ 29/ 3/ 369 (Zugriff: 28.08.2012), 29 (3), 369-388. Heinemann, M./ Heinemann, W. (2002). Grundlagen der Textlinguistik. Interaktion - Text - Diskurs. Tübingen. Heinemann, W. (2008). Textpragmatische und kommunikative Ansätze. In: Janich, N. (Hrsg.): Textlinguistik. 15 Einführungen. Tübingen: Narr, 113-143. Hirschi, A. (2011). Die Textsorte Brief für Direct-Marketing und die Anleitung zur Erstellung solcher Briefe. In: Demarmels, S./ Kesselheim, W. (Hrsg.): Textsorten in der Wirtschaft. Zwischen textlinguistischem Wissen und wirtschaftlichem Handeln. Wiesbaden: VS, 49-62. Hoefele, J./ Konstantinidou, L./ Kruse, O./ Dieterich, S. (2015). Förderung der allgemeinen Schreibkompetenz an Berufsschulen. Prozessorientierte Schreibdidaktik zwischen Deutsch als Mutter- (DaM) und Zweitsprache (DaZ). Valorisierungsbericht. Abrufbar unter: http: / / www.sbfi.admin.ch/ berufsbildung/ 01528/ 01536/ 02163/ 02164/ index.html ? lang=de (Zugriff: 17.05.2016). Hoefele, J./ Konstantinidou, L. (2018). Förderung der allgemeinen Schreibkompetenz im Bereich der beruflichen Bildung. In: Efing, C./ Kiefer, K.-H. (Hrsg.): Sprache und Kommunikation in der beruflichen Aus- und Weiterbildung. Ein interdisziplinäres Handbuch. Tübingen: Narr, 339-348. Hurrelmann, B./ Groeben, N. (2006). Textwissenschaftliche Grundlagen. In: Hurrelmann, B./ Groeben, N. (Hrsg.): Empirische Unterrichtsforschung in der Literatur- und Lese‐ didaktik. Weinheim/ München: Juventa, 31-51. References 239 <?page no="240"?> Industrie- und Handelskammer (IHK) Nürnberg für Mittelfranken-AkA (2010). Abschlus‐ sprüfung Sommer 2010. Hotelfachmann Hotelfachfrau 6116. Solingen: U-Form-Verlag. Industrie- und Handelskammer zu Leipzig (Hrsg.) (2011). Das deutsche Bildungssystem. Abrufbar unter: http: / / www.leipzig.ihk.de/ desktopdefault.aspx/ tabid-841/ 1073_read -19590/ usetemplate-druckansicht/ (Zugriff: 23.05.2011). Ingenkamp, K./ Lissmann, U. (2008). Lehrbuch der Pädagogischen Diagnostik. 6. Auflage. Weinheim: Beltz. Institut der deutschen Wirtschaft Köln (IWD) (2016). Jeder Dritte wechselt den Job. Abrufbar unter: https: / / www.iwd.de/ artikel/ jeder-dritte-wechselt-den-job-281979/ (Zugriff: 16.08.2018). Institut für Qualitätsentwicklung im Bildungswesen (IQB) (2015). IQB-Bildungstrend 2015 in der Sekundarstufe I. Abrufbar unter: https: / / www.iqb.hu-berlin.de/ bt/ BT2015 (Zugriff: 14.08.2018). Isenberg, H. (1974). Texttheorie und Gegenstand der Grammatik. Berlin. Isenberg, H. (1976). Einige Grundbegriffe für eine linguistische Texttheorie. In: Daneš/ Viehweger (Hrsg.): Ebenen der Textstruktur. Bd. I, Berlin: Akademie der Wissenschaf‐ ten,119-146. Jäger, A. (2007). Erfolgreich schreiben im Beruf. Darmstadt: Wissenschaftliche Buchge‐ sellschaft. Jakobs, E.-M. (2008). Berufliches Schreiben: Ausbildung, Training, Coaching. Überblick zum Gegenstand. In: Jakobs, E.-M./ Lehnen, K. (Hrsg.): Berufliches Schreiben. Ausbil‐ dung, Training, Coaching. Frankfurt a. M.: Peter Lang, 1-14. Janich, N. (2012). Fachsprache, Fachidentität und Verständigungskompetenz - zu einem spannungsreichen Verhältnis. In: Berufsbildung in Wissenschaft und Praxis. Sprache und Beruf. H. 20155, 10-13. Janssen, D./ Van Waes, L/ Van den Bergh, H. (1996). Effects of thinking aloud on writing processes. Mahwah, NJ: Lawrence Erlbaum. Jones, D./ Christensen, C. (1999). The relationship between automaticity in handwriting and students’ ability to generate written text. In: Journal of Educational Psychology, 91, 44-49 Jost, J. (2017). Prinzipien und Methoden lernförderlicher Schreibumgebungen. In: Becker-Mrotzek, M./ Grabowski, J./ Steinhoff, T. (Hrsg.): Forschungshandbuch empiri‐ sche Schreibdidaktik. Münster, New York: Waxmann, 173-186. Kaufer, D. S./ Hayes, J. R./ Flower, L. (1986): Composing written sentences. Research in the Teaching of English, 20, 2, 121-140. Kellogg, R. T. (1994). The Psychology of Writing. New York: Oxford University Press. Kellogg, R. T. (1996). A model of working memory in writing. In: Levy, C. M./ Ransdell, S. (Hrsg.): The science of writing: Theories, methods, individual differences, and applications. Mahwah, NJ: Erlbaum. 240 References <?page no="241"?> Kellogg, R. T. (2008). Training writing skills. A cognitive developmental perspective. In: Journal of Writing Research 1(1). Abrufbar unter: http: / / www.jowr.org/ articles/ vol1_ 1/ JoWR_2008_vol1_nr1_Kellogg.pdf (Zugriff: 31.10.2017), 1-26. Kessler, M./ Ziener, G. (2004). Woran kann man kompetenzorientiertes Unterrichten erkennen. (http: / / www.ptz-stuttgart.de/ uploads/ media/ RS-kompet-or_Unterrichten_ 02.pdf, Zugriff 09.03.2017). Keßler, C. (2006). Textrevision und Selbstregulation beim Verfassen von Aufsätzen. Eine Trainingsstudie bei Realschülern der 6. Klasse. Unveröffentlichte Diplomarbeit. Gießen: Justus-Liebig-Universität. Keßler, C. (2010). Förderung von Revisionsfertigkeiten bei Sechstklässlern. Prüfung der Effektivität und Wirkmechanismen eines selbstregulatorischen Schreibtrainings anhand leistungsbezogener, strategienaher und subjektiver Maße der Schreibkompe‐ tenz. Dissertation. Abrufbar unter: http: / / geb.uni-giessen.de/ geb/ volltexte/ 2010/ 7793/ index.html (Zugriff: 18.11.2012). Gießen: Justus-Liebig-Universität. Kimmelmann, N. (2013). Sprachlich-kommunikative Herausforderungen in niedrig quali‐ fizierten gewerblichen Berufen - Herausforderungen für die berufsbildenden Schulen. In: Die berufsbildende Schule (BbSch) 65.-Jahrgang. Heft 10, 293-297. Klein, H. E./ Schöpper-Grabe, S. (2018). Die Perspektive der Wirtschaft. In: Efing, C./ Kiefer, K.-H. (Hrsg.): Sprache und Kommunikation in der beruflichen Aus- und Weiterbildung. Ein interdisziplinäres Handbuch. Tübingen: Narr, 95-104. Knopp, M./ Becker-Mrotzek, M. (2018). Theoretische und empirische Perspektiven auf Inklusion. Ein Systematisierungsversuch aus Sicht der Deutschdidaktik. In: Pieper, I./ Pohl, T./ Wieser, D./ Kern, F. (Hrsg.): Didaktik Deutsch (44). Baltmannsweiler: Schneider Verlag Hohengehren, 84-100. Koch, H. H./ Pielow, W. (1984). Schreiben und Alltagskultur. Voraussetzungen und Haltungen des Schreibens in Schule, Hochschule und in außerschulischen Bereichen. Baltmannsweiler: Pädagogischer Verlag. Koch, P./ Oesterreicher, W. (1985). Sprache der Nähe-Sprache der Distanz. Mündlichkeit und Schriftlichkeit im Spannungsfeld von Sprachtheorie und Sprachgeschichte. In: Romanisches Jahrbuch, Band-36,15-43. Koch, P./ Oesterreicher, W. (2008). Mündlichkeit und Schriftlichkeit von Texten. In: Janich, N. (Hrsg.): Textlinguistik. 15 Einführungen. Tübingen: Narr, 199-215. Konrad, K. (2008). Erfolgreich Selbstgesteuert Lernen. Theoretische Grundlagen, For‐ schungsergebnisse, Impulse für die Praxis. Bad Heilbrunn: Klinkhardt. Konstantinidou, L./ Hoefele, J./ Kruse, O. (2016). Assessing writing in vocational education and training schools: Results from an intervention study. In: Göpferich, S./ Neumann, I. (Hrsg.): Developing and Assessing Academic and Professional Writing Skills. Frankfurt am Main: Peter Lang, 73-101. References 241 <?page no="242"?> Krelle, M. (2014). Videographie von Deutschunterricht - Methodische Anmerkun‐ gen und Hinweise. In: Neumann, A./ Mahler, I. (Hrsg.): Empirische Methoden der Deutschdidaktik: audio- und videografierende Unterrichtsforschung. Baltmanns‐ weiler: Schneider Verlag Hohengehren, 78-97. Langer, E. J. (1989). Mindfulness. Reading. Massachusetts: Addison-Wesley Publishing. Lehmann, R. H. (1992). The achieved quality of essay ratings. In: Purves, A. C. (Hrsg.): The IEA study of achievement in written composition II. Education and performance in fourteen countries. Oxford: Pergamon, 239-274). Lehmann, R. H./ Hunger, S./ Ivanov, S./ Gänsfuß, R. (2004). LAU 11. Aspekte der Lernaus‐ gangslage und der Lernentwicklung Klassenstufe 11. Ergebnisse einer längsschnittli‐ chen Untersuchung in Hamburg. Abrufbar unter: http: / / bildungsserver.hamburg.de/ lau/ (Zugriff: 17.05.2016). Lehmann, R. H./ Ivanov, S./ Hunger, S./ Gänsfuß, R. (2005). ULME I. Untersuchung der Leistungen, Motivation und Einstellungen zu Beginn der beruflichen Ausbildung. [gaebler.info/ hamburg/ ulme-1.pdf], letzter Zugriff 17.05.2016. Lehnen, K./ Schindler, K. (2010). Repertoires erweitern. Für andere Domänen trainieren. In: Perrin, D./ Böttcher, I./ Kruse, O./ Wrobel, A. (Hrsg.): Schreiben. Von intuitiven zu professionellen Schreibstrategien, Wiesbaden: Springer, 151-167. Lehnen, K. (2017). Kooperatives Schreiben. In: Becker-Mrotzek, M./ Grabowski, J./ Steinh‐ off, T. (Hrsg.): Forschungshandbuch empirische Schreibdidaktik. Münster, New York: Waxmann, 299-314. Leijten, M./ Van Waes, L. (2013). Keystroke Logging in Writing Research: Using Inputlog to Analyze and Visualize Writing Processes. In: Written Communication, 30(3) (http: / / www.inputlog.net/ download_software.html, Zugriff : 06.08.2018), 358-392. Leischner, D. (1993). Bildungssysteme in Europa. Bonn: Europa Union. Lewis, E. St. E. (1903). Catch Line and Argument. In: The Book-Keeper, Vol. 15, S.-124. Linnemann, M. (2014).Kognitive Prozesse der Adressatenantizipation beim Schreiben. Dissertationsschrift, [http: / / kups.ub.uni-koeln.de/ 5859], letzter Zugriff 01.05.2018, Disseration.Köln: Universität Köln. Linnemann, M. (2017). Erfassung von Schreibprozessen: Methoden, Techniken, Tools. In: Becker-Mrotzek, M./ Grabowski, J./ Steinhoff, T. (Hrsg.): Forschungshandbuch empirische Schreibdidaktik. Münster, New York: Waxmann, 335-352. Linke, A./ Nussbaumer, M./ Portmann, P. R. (2004): Studienbuch Linguistik.Tübingen: Narr. MacArthur, C. A. (2017). Thoughts on What Makes Strategy Instruction Work and How It Can Be Enhanced and Extended. In: Fidalgo, R,/ Harris, K. R./ Braaksma, M. (Hrsg.): Studies in Writing Series: Vol. 34. Design Principles for Teaching Effective Writing. Leiden: Brill, 235-252. 242 References <?page no="243"?> Marquardt, D. (2009). Brief. In: Lange, Günter/ Marquardt, Doris/ Petzoldt, Leander/ Zie‐ senis, Werner: Textarten - didaktisch. Eine Hilfe für den Literaturunterricht. 5. unveränderte Aufl., Baltmannsweiler: Schneider Verl. Hohengehren, 8-12. Marx, N./ Steinhoff, T. (2017). Unterrichtsbezogene Interventionen. In: Becker-Mrotzek, M./ Grabowski, J./ Steinhoff, T. (Hrsg.): Forschungshandbuch empirische Schreibdidak‐ tik. Münster, New York: Waxmann, 253-266. Mayring, P. (2010). Qualitative Inhaltsanalyse. Grundlagen und Techniken. 11. Aufl. Weinhein: Beltz. Meichenbaum, D. (1977). Cognitive Behavior Modification. An Integrative Approach. New York: Plenum Press. Mirzaii, M./ Aliabadi, R. B. (2013).Direct and indirect written corrective feedback in the context of genre-based instruction on job application letter writing. In: Journal of Writing Research, 5(2) Abrufbar unter: http: / / www.jowr.org/ abstracts/ vol5_2/ Mirzai i_Aliabadi_2013_5_2_abstract.html (Zugriff: 06.08.2018), 191-213. Neumann, A. (2006). Schreibkompetenz Hamburger Schülerinnen und Schüler in der 11. Klasse. Ergebnisse aus LAU 11 und ULME I. Didaktik Deutsch (21), 20-43. Neumann, A. (2007): Briefe schreiben in Klasse 9 und 11. Beurteilungskriterien, Messun‐ gen, Textstrukturen und Schülerleitungen. Münster: Waxmann. Neumann, A. (2010). Subgruppenanalysen der Schreibfähigkeiten anhand der Ergebnisse der DESI-Studie. In: Neumann, A./ Domenech, M. (Hrsg.): Paradoxien des Schreibens in der Bildungssprache Deutsch: Befunde zu Schreibsozialisation, Schreibmotivation und Schreibfähigkeit bei Schülerinnen und Schülern mit nichtdeutscher Mutterspr‐ ache und zum Schreibunterricht im mehrsprachigen Kontext. Hamburg: Kovač, 9-37. Neumann, A./ Matthiesen, F. (2011). Indikatorenmodell des schulischen Schreibens (IM‐ OSS). Testdokumentation 2009, 2010. Lüneburg: Leuphana Universität. Neumann, A. (2014). Gemeinsamkeiten und Unterschiede bei der Kodierung schriftlicher und mündlicher Texte im schulischen Praxisfeld. In: Neumann, A./ Mahler, I. (Hrsg.): Empirische Methoden der Deutschdidaktik: audio- und videografierende Unterrichts‐ forschung. Baltmannsweiler: Schneider Verlag Hohengehren, 53-77. Neumann, A. (2017). Zugänge zur Bestimmung von Textqualität. In: Becker-Mrotzek, M./ Grabowski, J./ Steinhoff, T. (Hrsg.): Forschungshandbuch empirische Schreibdidaktik. Münster, New York: Waxmann, 203-219. Neumann, A./ Giera, W.-K. (2018). Diagnose von Schreibkompetenzen in der beruflichen Bildung. In: Efing, C./ Kiefer, K.-H. (Hrsg.): Sprache und Kommunikation in der beruflichen Aus- und Weiterbildung. Ein interdisziplinäres Handbuch. Tübingen: Narr, 333-340. Neumayer, G./ Rudolph, U. (2008). Geschäftskorrespondenz von A-Z. Kreativ und profes‐ sionell Briefe, Faxe und E-Mails schreiben. 2., vollständig aktualisierte Aufl. Hannover: Humboldt. References 243 <?page no="244"?> Niedersächsisches Kultusministerium (Hrsg.) (2006a). Kerncurriculum für die Haupt‐ schule Schuljahrgänge 5-10. Deutsch. Hannover: Unidruck. Niedersächsisches Kultusministerium (Hrsg.) (2006b). Kerncurriculum für die Realschule Schuljahrgänge 5-10. Deutsch. Hannover: Unidruck. Niedersächsisches Kultusministerium (Hrsg.) (2006c). Kerncurriculum für die Integrierte Gesamtschule Schuljahrgänge 5-10. Deutsch. Hannover: Unidruck. Niedersächsisches Kultusministerium (Hrsg.) (2006d). Kerncurriculum für die Integrierte Gesamtschule Schuljahrgänge 5-10. Deutsch. Hannover: Unidruck. Niedersächsisches Kultusministerium (Hrsg.) (2006e). Kerncurriculum für das Gymna‐ sium Schuljahrgänge 5-10. Deutsch. Hannover: Unidruck. Nill, U. (2018). Die Perspektive der Lehrwerkspraxis. In: Efing, C./ Kiefer, K.-H. (Hrsg.): Sprache und Kommunikation in der beruflichen Aus- und Weiterbildung. Ein inter‐ disziplinäres Handbuch. Tübingen: Narr, 115-124. Nitz, L. (2010). Texte überarbeiten - Das Problem der Kohärenz in Schreibkonferenzen. Eine textlinguistische Untersuchung in der Grundschule. Berlin: LIT. Nystrand, (2006). The Social and Historical Context for Writing Research, In: MacArthur, C./ Graham, S.; Fitzgerald, J. (Hrsg.): Handbook of Writing Research. New York: Guilford Press, 11-27. OECD (2013). OECD Skills Outlook 2013. First Results from the Survey of Adult Skills. Abrufbar unter: https: / / doi.org/ 10.1787/ 9789264204256-en (Zugriff: 15.05.2014). Ortner, H. (2013). Schreiben und Wissen. Einfälle fördern und Aufmerksamkeit staffeln. In: Perrin, D./ Böttcher, I./ Kruse, O./ Wrobel, A. (Hrsg.): Schreiben. Von intuitiven zu professionellen Schreibstrategien. 2., überarbeitete Aufl. Wiesbaden: Westdeutscher Verlag, 63-81. Ossner, J. (2006). Sprachdidaktik Deutsch. Eine Einführung. Weinheim, Basel: Schöningh. Parjares, F./ Valiante, G. (2006). Self-Efficacy Beliefs and Motivation in Writing Devel‐ opment. In: MacArthur, C./ Graham, S./ Fitzgerald, J. (Hrsg.): Handbook of Writing Research. London, New York: Guilford Press: 158-170. Parsons, T. (1951). The social system. New York, London: Free Press. Pätzold, G. (1995). Vermittlung von Fachkompetenz in der Berufsbildung. In: Arnold, R./ Lipsmaier, A. (Hrsg.): Handbuch der Berufsbildung. Opladen: Leske + Budrich, 157-169. Philipp, M. (2012). Motiviert lesen und schreiben. Dimension, Bedeutung, Förderung. Seelze: Klett/ Kallmayer. Philipp, M. (2015a). Because Writing Matters! (Berufliches) Schreiben und seine effektive Förderung. In: Efing, C. (Hrsg.): Sprache und Kommunikation in der beruflichen Bildung. Modellierung - Anforderungen - Förderungen. Frankfurt am Main: Peter Lang, 151-168. 244 References <?page no="245"?> Philipp, M. (2017). Wirksame Schreibförderung - metaanalytische Befunde. In: Becker-Mrotzek, M./ Grabowski, J./ Steinhoff, T. (Hrsg.): Forschungshandbuch empiri‐ sche Schreibdidaktik. Münster, New York: Waxmann,187-202. Philipp, M. (2018). Ausbildungsvorbereitende Schreibförderung in der Sekundarstufe I. In: Efing, C./ Kiefer, K.-H. (Hrsg.): Sprache und Kommunikation in der beruflichen Aus- und Weiterbildung. Ein interdisziplinäres Handbuch. Tübingen: Narr, 349-359. Piaget, J. (1947). La représentation du monde chez l’enfant. Presses Universitaires de France. Piehl, E./ Sellin, B. (1995). Berufliche Aus- und Weiterbildung in Europa. In: Arnold, R./ Lipsmaier, A. (Hrsg.): Handbuch der Berufsbildung. Opladen: Leske + Budrich, 441-454. Pieper, I. (2018). Zumutungen erkennen, Verhältnismäßigkeiten im Blick behalten: Für mehr Balance in einer feldnahen Deutschdidaktik. In: Pieper, I./ Pohl, T./ Wieser, D./ Kern, F. (Hrsg.): Didaktik Deutsch (44). Baltmannsweiler: Schneider Verlag Hohen‐ gehren, 4-9. Pohl, T. / Steinhoff, T. (2010). Textformen als Lernformen. In: Pohl, T./ Steinhoff, T. (Hrsg.): Textformen als Lernformen. Kölner Beiträge zur Sprachdidaktik (7). Duisburg: Gilles & Francke: 5-27. Quasthoff, U./ Heller, V. (2014). Mündlichkeit und Schriftlichkeit aus sprachwissen‐ schaftlicher und sprachdidaktischer Sicht: Grundlegende Ein-/ Ansichten und meth‐ odische Anregungen. In: Neumann, A./ Mahler, I. (Hrsg.): Empirische Methoden der Deutschdidaktik: audio- und videografierende Unterrichtsforschung. Baltmanns‐ weiler: Schneider Verlag Hohengehren, 6-37. Reetz, L./ Seyd, W. (1995). Curriculare Strukturen beruflicher Bildung. In: Arnold, R./ Lipsmaier, A. (Hrsg.): Handbuch der Berufsbildung. Opladen: Leske + Budrich, 203- 219. Rehbein, J. (1983). Zur pragmatischen Rolle des ‘Stils’. In: Sandig, B. (Hrsg.). Stilistik. Band I Probleme der Stilistik. Hildesheim: Olms, 21-48. Rijlaarsdam, G./ van den Bergh, H. (1996). The Dynamics Of Composing. In: Levy, C. M./ Ransdell, S. E. (Hrsg.): The Science of Writing: Theories, Methods, Individual Differences, and Applications. New York, London: Erlbaum, 207-232. Rijlaarsdam, G./ van den Bergh, H. (2006). Writing Process Theory. A Functional Dynamic Approach. In: MacArthur, C. A./ Graham, S./ Fitzgerald, J. (Hrsg.): Handbook of Writing Research. New York, London: Guilford Press, 41-64. Robledo-Ramón, P./ García, J. N. (2017). Description and Analysis of Strategy-Focused Instructional Models for Writing. In: Fidalgo, R,/ Harris, K. R./ Braaksma, M. (Hrsg.): Studies in Writing Series: Vol. 34. Design Principles for Teaching Effective Writing. Leiden: Brill, 38-65. Rolf, E. (1993). Die Funktionen der Gebrauchstextsorten. Berlin, New York: de Gryter. References 245 <?page no="246"?> Rosebrock, C. (2007). Anforderungen von Sach- und Informationstexten, Anforderungen literarischer Texte. In: Bertschie-Kaufmann, A. (Hrsg.): Lesekompetenz - Leseleistung - Leseförderung. Grundlagen, Modelle und Materialien. Seelze: Friedrich, 50-65. Rost, D. H. (2007). Interpretation und Bewertung pädagogisch-psychologischer Studien. Eine Einführung. 2. Aufl. Weinheim, Basel: Beltz. Saddler, B./ Moran, S./ Graham, S./ Harris, K. R. (2004). Preventing Writing Difficulties: The Effects of Planning Strategy Instruction on the Writing Performance of Struggling Writers. In: Exceptionality, 2 (1), Lawrence Erlbaum Associates, Inc., 3-17. Saldern, M. von (1999). Schulleistung in Diskussion. Baltmannsweiler: Schneider Hohen‐ gehren. Saldern, M. von (2011). Schulleistung 2.0. Von der Note zum Kompetenzraster. Norder‐ stedt: Books on Demand. Sandig, B. (2006). Textstilistik des Deutschen. 2. Auflage. Berlin, New York: Walter de Gryter. Saughnessy, M. (1977). Erros and expectations. London: Oxford University Press. Sawyer, R. J./ Graham, S./ Harris, K. R. (1992). Direct Teaching, Strategy Instruction, and Strategy Instruction With Explicit Self-Regulation. Effects on the Composition Skills and Self-Efficacy of Students With Learning Disabilities. In: Journal of Educational Psychology, Vol. 84, No. 3, 340-352. Schaper, R.-H./ Schreiber, R./ Seyd, W. (2000). Der Berufsausbilder. Lehr- und Arbeitsbuch zu den >>Sieben Handlungsfeldern<< Die berufs- und arbeitspädagogischen Qualifi‐ kationen des Ausbilders nach der neuen AEVO. sechste, vollständig neu gefasste Auf. Hamburg: Feldhaus. Schindler, K. (2005). Studierende schreiben beruflich. Beobachtungen einer empirischen Studie. In: Jakobs, E.-M./ Lehnen, K./ Schindler, K. (Hrsg.): Schreiben am Arbeitsplatz. Schreiben-Medien-Beruf. Wiesbaden: VS, 217-232. Schindler, K. (2013). Schreiben im Beruf In: Efing, C. (Hrsg.): Ausbildungsvorbereitung im Deutschunterricht der Sekundarstufe I. Die sprachlich-kommunikativen Facetten von „Ausbildungsfähigkeit“. Frankfurt am Main: Peter Lang, 173-190. Schindler, K. (2017). Studium und Beruf. In: Becker-Mrotzek, M./ Grabowski, J./ Steinhoff, T. (Hrsg.): Forschungshandbuch empirische Schreibdidaktik. Münster, New York: Waxmann, 109-123. Schmickler, B. (2016). Ausbildungsmarkt in der Krise. Wer will schon Bäcker oder Fleischer werden? Tagesschau. Abrufbar unter: www.tagesschau.de/ wirtschaft/ ausbi ldungsjahr-101html (Zugriff: 20.03.2017). Schmitt, M./ Knopp, M. (2017). Prädiktoren der Schreibkompetenz. In: Becker-Mrotzek, M./ Grabowski, J./ Steinhoff, T. (Hrsg.): Forschungshandbuch empirische Schreibdidak‐ tik. Münster, New York: Waxmann, 240-252. 246 References <?page no="247"?> Schneider, H./ Anskeit, N. (2017). Einsatz digitaler Schreibwerkzeuge. In: Becker-Mrotzek, M./ Grabowski, J./ Steinhoff, T. (Hrsg.): Forschungshandbuch empirische Schreibdidak‐ tik. Münster, New York: Waxmann, 283-298. Schründer-Lenzen, A. (2009). Sprachlich-kulturelle Heterogenität als Unterrichtsbedin‐ gung. In: Fürstenau, S./ Gomolla, M. (Hrsg.): Migration und schulischer Wandel: Unterricht. Wiesbaden: Verlag für Sozialwissenschaften, 121-138. Schwitalla, J. (1997). Gesprochenes Deutsch. Eine Einführung. (Grundlagen der Germa‐ nistik, Band-33), Berlin: Erich Schmidt. Searle, S. R. (1971). Sprechakte. Ein sprachphilosophischer Essay. Übersetzt von R. Wiggershaus. Frankfurt a. M.: Suhrkamp. Settelmeyer, A./ Widera, C. (2018). Die Perspektive der Berufsbildungspolitik. In: Efing, C./ Kiefer, K.-H. (Hrsg.): Sprache und Kommunikation in der beruflichen Aus- und Weiterbildung. Ein interdisziplinäres Handbuch. Tübingen: Narr, 105-114. Sieland, B./ Rahm, T. (2007). Personale Kompetenz entwickeln. In: Fleischer, Tho‐ mas/ Grewe, Norbert/ Jötten, Bernd/ Seifried, Klaus/ Sieland, Bernhard (Hrsg.): Hand‐ buch Schulpsychologie. Psychologie für die Schule. Stuttgart: Kohlhammer: 197-210. Söll, L./ Hausmann, F. J. (1985). Gesprochenes und geschriebenes Französisch. (Grundla‐ gen der Romanistik, Band-6), Berlin: Romanistik Spitta, G. (1992). Schreibkonferenzen in Klasse 3 und 4. Ein Weg vom spontanen Schreiben zum bewußten Verfassen. Frankfurt/ a. M.: Cornelsen Scriptor. Stanat, P./ Rauch, D./ Segeritz, M. (2010). Schülerinnen und Schüler mit Migrationshinter‐ grund. In: Klieme E./ Artelt, C./ Hartig, J./ Jude, N./ Köller, O./ Prenzel, M./ Schneider, W./ Stanat, P. (Hrsg.): PISA 2009. Bilanz nach einem Jahrzehnt. Münster: Waxmann, 200-230. Ständige Konferenz der Kultusminister der Länder in der Bundesrepublik Deutschland (KMK) (Hrsg.) (1996/ 2011). Handreichungen für die Erarbeitung von Rahmenlehrplä‐ nen der Kultusministerkonferenz für den berufsbezogenen Unterricht in der Berufs‐ schule und ihre Abstimmung mit Ausbildungsordnungen des Bundes für anerkannte Ausbildungsberufe (http: / / www.kmk.org/ fileadmin/ Dateien/ veroeffentlichungen_be schluesse/ 2011/ 2011_09_23_GEP-Handreichung.pdf, Zugriff 23.06.2016). Ständige Konferenz der Kultusminister der Länder in der Bundesrepublik Deutschland (KMK) (Hrsg.) (2004). Beschlüsse der Kultusministerkonferenz. Bildungsstandards im Fach Deutsch für den Mittleren Bildungsabschluss Beschluss vom 04.12.2003. Köln: Luchterhand. Ständige Konferenz der Kultusminister der Länder in der Bundesrepublik Deutschland (KMK) (Hrsg.) (2005). Beschlüsse der Kultusministerkonferenz. Bildungsstandards im Fach Deutsch für den Hauptschulabschluss ( Jahrgangsstufe 9) Beschluss vom 15.10.2004. Köln: Luchterhand. References 247 <?page no="248"?> Ständige Konferenz der Kultusminister der Länder in der Bundesrepublik Deutschland (KMK) (2010). Gemeinsame Erklärung der Partner des Ausbildungspaktes und der Bundesagentur für Arbeit. Abrufbar unter: https: / / www.kmk.org/ aktuelles/ artikelan sicht/ gemeinsame-erklaerung-der-partner-des-ausbildungspaktes-und-der-bundesa‐ gentur-fuer-arbeit.html (Zugriff: 31.07.2018). Statistisches Bundesamt (2017a) (Destatis). Zahlen und Fakten. Bildung, Forschung, Kultur. Abrufbar unter: https: / / www.destatis.de/ DE/ ZahlenFakten/ GesellschaftStaat/ BildungForschungKultur/ BildungForschungKultur.html (Zugriff: 08.06.2018). Statistisches Bundesamt (Destatis) (2017b). Bevölkerung und Erwerbstätigkeit. Bevöl‐ kerung mit Migrationshintergrund - Ergebnisse des Mikrozensus 2016. Abrufbar un‐ ter: https: / / www.destatis.de/ DE/ Publikationen/ Thematisch/ Bevoelkerung/ Migration Integration/ Migrationshintergrund2010220167004.pdf ? blob=publicationFile (Zugriff: 28.07.2018). Steendam, E. van/ Rijlaarsdam, G./ Sercu, Lies/ Bergh, H. van den (2012). Effective Instructional Strategies in Collaborative Revision in EFL: Two Empirical Studies. In: Torrance, M./ Alamargot, D./ Castelló, M./ Ganier, F./ Kruse, O./ Tolchinksy, L./ Van Waes, L. (Hrsg.): Learning to Write Effectively: Current Trends in European Research. Amsterdam: Emerald, 131-134. Steinhoff, T./ Grabwoski, J./ Becker-Mrotzek, M. (2017). Herausforderungen der empir‐ ischen Schreibdidakik. In: Becker-Mrotzek, M./ Grabowski, J./ Steinhoff, T. (Hrsg.): Forschungshandbuch empirische Schreibdidaktik. Münster, New York: Waxmann, 9-24. Steinhoff, T. (2017). Untersuchung von Textkorpora. In: Becker-Mrotzek, M./ Grabowski, J./ Steinhoff, T. (Hrsg.): Forschungshandbuch empirische Schreibdidaktik. Münster, New York: Waxmann, 353-368. Stein, N./ Glenn, C. (1979). An analysis of story comprehension in elementary school children. In: Freedle, R. (Hrsg.): Advances in discourse processes, Vol. 2. New directions in discourse processing. Norwood, NJ: Ablex. Steinmüller, U. (2006). Deutsch als Zweitsprache - ein Politikum. In: Ahrenholz, B. (Hrsg.): Kinder mit Migrationshintergrund - Spracherwerb und Fördermöglichkeiten. Freiburg i. Brsg.: Fillibach, 322-331. Sturm, A. (2017). Strategiefokussierte Intervention. In: Becker-Mrotzek, M./ Grabowski, J./ Steinhoff, T. (Hrsg.): Forschungshandbuch empirische Schreibdidaktik. Münster, New York: Waxmann, 267-281. Thomas, L. (2007). Lern- und Leistungsdiagnostik. In: Fleischer, T./ Grewe, N./ Jötten, B./ Seifried, K./ Sieland, B. (Hrsg.): Handbuch Schulpsychologie. Psychologie für die Schule. Stuttgart: Kohlhammer, 82-98. Trappen, S. (2013). Repertoires öffnen. Ein Rhetorik-Modell für Schreibtrainings. In: Perrin, D./ Böttcher, I./ Kruse, O./ Wrobel, A. (Hrsg.): Schreiben. Von intuitiven zu pro‐ 248 References <?page no="249"?> fessionellen Schreibstrategien. 2., überarb. Aufl. Wiesbaden: Westdeutscher Verlag, 170-183. Troia, G. A./ Graham, S./ Harris, K. R. (1999). Teaching Students with Learning Disabilities to Mindfully Plan When Writing. In: Exeptional Children, Vol. 65, No.2: 235-252. Van Waes, L./ Leijten, M./ Weijen, D. van (2009). Keystroke logging in writing research. Observing writing processes with Inputlog. Abrufbar unter: http: / / www.gfl-journal. de/ 2-2009/ vanwaes.pdf (Zugriff: 01.05.2018). Wagner, R. W. (2003). Methoden des Unterrichts in mündlicher Kommunikation. In: Bre‐ del, U./ Günther, H./ Klotz, P./ Ossner, J./ Siebert-Ott, G. (Hrsg.): Didaktik der deutschen Sprache. Ein Handbuch. Paderborn: Schöningh, 747-759. Weigle, S. C. (2002). Assessing writing. Cambridge, UK: Cambridge University Press. Weinert, F. E. (Hrsg.) (2001). Vergleichende Leistungsmessung in Schulen - eine umstrittene Selbstverständlichkeit. In: Weinert, F. E. (Hrsg.): Leistungsmessungen in Schulen. Weinheim: Beltz, 17-31. Weinzierl, C./ Wrobel, A. (2017). Schreibprozesse untersuchen. In: Becker-Mrotzek, M./ Grabowski, J./ Steinhoff, T. (Hrsg.): Forschungshandbuch empirische Schreibdidaktik. Münster, New York: Waxmann, 221-237. Wellenreuther, M. (2009). Forschungsbasierte Schulpädagogik. Anleitungen zur Nutzung empirischer Forschung für die Schulpraxis. Baltmannsweiler: Schneider-Verl. Hohen‐ gehren. Wergen, J./ Wörner, A. (2005). Im Griff Bürokommunikation. Stuttgart: Ernst Klett Sprachen GmbH. Westdeutscher Rundfunk (WDR) (2015). Schichtwechsel, die Roboter übernehmen. Abrufbar unter: http: / / www.arte.tv/ guide/ de/ 055859-000-A/ schichtwechsel-die-robo ter-ubernehmen? country=DE (Zugriff: 07.11.2016). Winkel, S./ Petermann, F./ Petermann, U. (2006). Lernpsychologie. Paderborn: Schöningh. Wisskirchen, G. (2006). Das Allgemeine Gleichbehandlungsgesetz - Auswirkungen auf die Praxis. Frechen: Datakontext. Wolf, M. (2009). Das lesende Gehirn. Wie der Mensch zum Lesen kam-und was es in unseren Köpfen bewirkt. Heidelberg: Spektrum Akademischer Verlag. Wood, D./ Bruner, J./ Ross, G. (1976). The Role of Tutoring in Problem Solving. In: J. Child Psychol. Psychiat., Vol. 17. Pergamon Press, 89-100. Wunderlich, D. (1976). Studien zur Sprechakttheorie. Frankfurt am Main: Suhrkamp. Wygotski, L. S. (1987). Ausgewählte Schriften. Arbeiten zur psychischen Entwicklung der Persönlichkeit, Band-2. Köln: Pahl-Rugenstein. Wyss-Kolb, M. (1995). Was und wie Lehrlinge schreiben. Eine Analyse von Schreib‐ gewohnheiten und von ausgewählten formalen Merkmalen in Aufsätzen. Aarau: Sauerländer. References 249 <?page no="250"?> Xiong, W., Litman D., & Schunn, C. (2012).Natural Language Processing techniques for researching and improving peer feedback. In: Journal of Writing Research, 4(2), Abrufbar unter: http: / / www.jowr.org/ abstracts/ vol4_2/ Xiong_et_al_2012_4_2_abstra ct.html (Zugriff: 06.08.2018,155-176). Zentralstelle für Prüfungsaufgaben der Industrie- und Handelskammern (ZPA-Nord-West) (2011). Wir über uns. Abrufbar unter: http: / / www.ihk-zpa.de/ op encms/ Wir-ueber-uns (Zugriff: 27.05.2011). Ziener, G./ Kessler, M. (2012). Kompetenzorientiert unterrichten - mit Methode. Meth‐ oden entdecken, verändern, erfinden. Seelze: Klett, Kallmeyer. 250 References <?page no="251"?> List of Figures Fig. 1: Writing development model (Bereiter, 1980: 85) . . . . . . . . . 21 Fig. 2: Learning processes (Winkel/ Petermann & Petermann, 2006: 229, translated using AI) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25 Fig. 3: Competence grid according to Kessler & Ziener (2004, own illustration) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 Fig. 4: Overview of candidate predictors (Schmitt & Knopp, 2017: 241, translated using AI) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30 Fig. 5: Wheel of Writing I (Berge et al., 2016: 180) . . . . . . . . . . . . . . 31 Fig. 6: Wheel of Writing II (Berge et al., 2016: 182) . . . . . . . . . . . . . 32 Fig. 7: Criteria catalog for training maturity (Federal Employment Agency, 2006: 22, translated by WKG) . . . . . . . . . . . . . . . . . . 34 Fig. 8: Language fields at school (after Feilke, 2012a: 2011: 6, simplified representation, translated by WKG) . . . . . . . . . . 38 Fig. 9: Cooperation partners for dual training (own illustration) . 44 Fig. 10: Current training situation in Germany (own illustration, translated using AI) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50 Fig. 11: Writing components in the workplace (after Jakobs, 2005: 17, translated using AI) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52 Fig. 12: Phases of writing socialization ( Jakobs, 2008: 4, translated by WKG) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54 Fig. 13: Potsdam Triangle Model (own illustration) . . . . . . . . . . . . . 63 Fig. 14: Proximity-distance continuum (Koch & Oesterreicher, 2008: 201, translated by WKG) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67 Fig. 15: Professional writing occasions (after Jäger, 2007: 82 ff., own illustration) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73 Fig. 16: Building blocks of the mailing concept (after Hirschi, 2011: 52 f., own illustration, translated using AI) . . . . . . . . . . . . . . 75 Fig. 17: Factors for correspondence (after Jakobs, 2008: 2 f., own illustration) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80 Fig. 18: Strategy terms according to Sturm (2017: 267 f., own illustration) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82 Fig. 19: Writing process model (Hayes & Flower 1980: 10) . . . . . . . . 83 Fig. 20: Writing model as an ‘individual-environmental model’ (Hayes, 1996) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88 Fig. 21: Three-memory model (Winkel/ Petermann & Petermann, 2006: 36, translated by WKG) . . . . . . . . . . . . . . . . . . . . . . . . . 89 <?page no="252"?> Fig. 22: Writing didactic cube (Fix, 2000: 324 f., translated using AI) 91 Fig. 23: Writing ability development according to Kellogg (2008: 3, own illustration) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93 Fig. 24: Writing process model (Hayes, 2012: 371) . . . . . . . . . . . . . . . 94 Fig. 25: Basic model of text production (Bachmann & Becker-Mrotzek, 2017: 42, translated by WKG) . . . . . . . . . . 96 Fig. 26: Circle II: Representation of the communicative situation and adaptation of the text pattern (Bachmann & Becker-Mrotzek, 2017: 48, translated using AI) . . . . . . . . . . 97 Fig. 27: SRSD process (after Glaser, 2005; own illustration, translated by WKG) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98 Fig. 28: Theory of SRSD (after Harris & Graham, 2017: 123, own illustration) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101 Fig. 29: SRSD instructions (after Harris & Graham, 2017: 126 f., own illustration) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118 Fig. 30: Measuring writing skills (own illustration) . . . . . . . . . . . . . . 123 Fig. 31: The four perspectives of the study (own illustration) . . . . . 130 Fig. 32: Examination perspectives in the basic model of text production by Bachmann & Becker-Mrotzek (2017: 42, translated by WKG) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134 Fig. 33: Research perspectives in the writing process model by Hayes (2012) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134 Fig. 34: Composition of the sample (own illustration) . . . . . . . . . . . 148 Fig. 35: Chronology & test instruments (own illustration) . . . . . . . . 153 Fig. 36: Methods of writing process research as a function of synchronicity, directness and detail (after Linnemann, 2017: 337, translated by WKG) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 159 Fig. 37: KaWa example (own illustration) . . . . . . . . . . . . . . . . . . . . . . 162 Fig. 38: Number of breaks for offers & applications (pre/ post test, n = 40) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 203 Fig. 39: Break time for offers & applications (pre/ post test, n = 40) 203 Fig. 40: Ratio of number of pauses to length of cover letter (pretest, n = 40) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 205 Fig. 41: Ratio of number of pauses to length of cover letter (post-test, n = 40) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 206 Fig. 42: Ratio of number of breaks to length of offer (pretest, n = 40) 206 Fig. 43: Ratio of number of breaks to length of breaks offered (post-test, n = 40) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 207 252 List of Figures <?page no="253"?> List of Tables Tab. 1: Mean values and standard deviations of the items for measuring writing attitude (Affect) . . . . . . . . . . . . . . . . . . . . . . 180 Tab. 2: Mean values and standard deviations of the items for measuring text quality (Behavior) . . . . . . . . . . . . . . . . . . . . . . . . 182 Tab. 3: Correlations (Kendall’s tau-b) for the application (pretest) . . . 183 Tab. 4: Correlations (Kendall’s tau-b) for the offer (post-test) . . . . . . . 183 Tab. 5: Mean values and standard deviations of the items for measuring text quality (Behavior) . . . . . . . . . . . . . . . . . . . . . . . . 186 Tab. 6: Mean values and standard deviations of the items for measuring content learning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 190 Tab. 7: Correlations (Kendall-Tau-b) with content design, overall impression, word count & student assessment of content (offer posttest) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 195 Tab. 8: Mean values and standard deviations of the items on writing strategies (metacognition) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 198 Tab. 9: Mean values and standard deviations of the keylogging results 201 <?page no="254"?> ISBN 978-3-381-15521-7 This volume presents the results of an intervention study involving 9 th -grade high school students and apprentices in the hospitality industry, which empirically examines writing in a professional context. The study focuses on the question of what writing performance the students demonstrate when drafting job applications and advertising letters during the course of the writing project using “self-regulated strategy development”. To answer this question, both the written products and the writing processes were recorded and evaluated primarily using quantitative methods. The goal of the study is to improve the writing competencies of the participating students based on the selected text types, as measured by content, linguistic, and formal aspects. Giera Promoting career-oriented writing skills through SRSD Winnie-Karen Giera Promoting career-oriented writing skills through SRSD An evaluation of a school writing project