With your consent, we store data on your device to analyze your usage behavior and seamlessly integrate external content for you. You can revoke your consent or make specific settings at any time.
The dieS Summer School 2026 will take place this year from 18 to 20 June at the University of Education in Karlsruhe.
It is dedicated to the theme of ‘Networked Writing Pedagogy’ and thus focuses on the diverse intersections between writing research, educational science, psychology and digital technologies. The focus is on interdisciplinary perspectives on writing processes, as well as the methodological challenges arising from the increasing complexity of data, AI-supported writing environments and collaborative practices.
The summer school offers a varied programme comprising lectures, methodology workshops, discussion forums and advisory colloquia, in which quantitative and digital approaches to writing research are presented and examined. Topics such as AI-assisted writing processes, multimodal storytelling and collaborative writing in subject-specific teaching provide up-to-date insights into the breadth and dynamics of empirical writing didactics.
The dieS Summer School is aimed at all interested researchers, but particularly at PhD students and postdocs who wish to further develop their projects, gain methodological confidence and discover opportunities for interdisciplinary collaboration.
We look forward to welcoming you here at the Karlsruhe University of Education (PHKA) and working together on a writing pedagogy that brings together different perspectives, opens up research and embraces methodological diversity. You can register until 15 May.
The summer school combines various formats that enable a wide range of participation in the current discourse on writing pedagogy at all stages of academic development: lectures and discussions, workshops in which empirical data is analysed, advisory colloquia for PhD projects, Open Science sessions and a supporting programme offering plenty of opportunities for individual discussions.
We ask that you attend all sessions!
Programme for the dieS Summer School 2026
Research into writing instruction interventions faces the challenge of generating findings that are both as causally interpretable as possible and ‘practice-oriented’. Whilst the former is often associated with laboratory experiments that are detached from real-life contexts, ‘practice-oriented’ approaches are regarded by many as being more evaluative in nature.
The lecture begins by introducing the distinction between internal, external, construct and statistical validity, using examples from the field of writing pedagogy. It will then outline which characteristics of study designs (randomisation, matching, random sampling, etc.) contribute to enhancing these forms of validity, and which typical threats to them (test practice, nested data, maturation processes, etc.) are present in everyday research practice.
In the second part, the lecture discusses traditional and more recent study designs that make it possible to simultaneously enhance internal and ecological validity in the didactics of writing. It will also examine the extent to which such ambitious projects can benefit from inter- and transdisciplinary cooperation.
Research into writing instruction interventions faces the challenge of generating findings that are both as causally interpretable as possible and ‘practice-oriented’. Whilst the former is often associated with laboratory experiments that are detached from real-life contexts, ‘practice-oriented’ approaches are regarded by many as being more evaluative in nature.
The lecture begins by introducing the distinction between internal, external, construct and statistical validity, using examples from the field of writing pedagogy. It will then outline which characteristics of study designs (randomisation, matching, random sampling, etc.) contribute to enhancing these forms of validity, and which typical threats to them (test practice, nested data, maturation processes, etc.) are present in everyday research practice.
In the second part, the lecture discusses traditional and more recent study designs that make it possible to simultaneously enhance internal and ecological validity in the didactics of writing. It will also examine the extent to which such ambitious projects can benefit from inter- and transdisciplinary cooperation.
This presentation addresses the methodological requirements and challenges of didactic-empirical writing research in interdisciplinary research designs, using the example of a cluster-randomised intervention study on multimodal storytelling. The starting point is the question of how complex, digitally supported writing arrangements can be empirically investigated in such a way that both a close focus on the process and methodological control are ensured.
An intervention study is presented on the effectiveness of digital storytelling (DST) on the narrative text production skills of pupils in the orientation stage. The study examined a writing arrangement for revising narrative texts, in which the writing task remained constant across all groups, whilst only the revision phase was varied. Three conditions were compared: (1) a DST group using multimodal storytelling, (2) an intervention group revising their work based on an analysis of a model text, and (3) a control group without any specific support measures.
This design illustrates key methodological tensions: the operationalisation of text quality in multimodal contexts, the combination of holistic (NAEP) and analytical rating procedures (RANT), the handling of pre-post dependencies, and the consideration of nested data structures in class-level randomisation. The effects of the writing arrangements on the text quality of the final drafts were modelled using linear regression models, controlling for baseline performance; cluster-robust standard errors were used to adjust for the class structure. The presentation uses the study as a case study to shed light on analytical decisions, issues of quality and the limitations of quantitative effectiveness research in complex, digitally driven writing settings. In the outlook, measurement validity for multimodal products and the comparability of support formats are discussed in particular.
This presentation addresses the methodological requirements and challenges of didactic-empirical writing research in interdisciplinary research designs, using the example of a cluster-randomised intervention study on multimodal storytelling. The starting point is the question of how complex, digitally supported writing arrangements can be empirically investigated in such a way that both a close focus on the process and methodological control are ensured.
An intervention study is presented on the effectiveness of digital storytelling (DST) on the narrative text production skills of pupils in the orientation stage. The study examined a writing arrangement for revising narrative texts, in which the writing task remained constant across all groups, whilst only the revision phase was varied. Three conditions were compared: (1) a DST group using multimodal storytelling, (2) an intervention group revising their work based on an analysis of a model text, and (3) a control group without any specific support measures.
This design illustrates key methodological tensions: the operationalisation of text quality in multimodal contexts, the combination of holistic (NAEP) and analytical rating procedures (RANT), the handling of pre-post dependencies, and the consideration of nested data structures in class-level randomisation. The effects of the writing arrangements on the text quality of the final drafts were modelled using linear regression models, controlling for baseline performance; cluster-robust standard errors were used to adjust for the class structure. The presentation uses the study as a case study to shed light on analytical decisions, issues of quality and the limitations of quantitative effectiveness research in complex, digitally driven writing settings. In the outlook, measurement validity for multimodal products and the comparability of support formats are discussed in particular.
In lessons, pupils often write in order to record content discussed orally. This allows them to revisit the material later, for example when preparing for exams (= retention writing). However, writing can also support learning. It involves various cognitive processes that improve the storage of information in long-term memory (Silva and Limongi 2019, Graham and Hebert 2011): one must select the most important points, structure and elaborate thoughts, express connections in language, and so on. (= learning through writing). As plausible as this sounds from a cognitive theory perspective, empirical studies do not paint such a clear-cut picture (Bangert-Drowns et al. 2004; Graham et al. 2020). It can therefore be concluded that writing cannot be viewed merely as a cognitive phenomenon; rather, greater focus should be placed on its social component: what do writers do when writing is used as a learning tool in a learning context? Under what conditions can this be successful?
This talk provides insights into current research findings on learning through writing and the role played by the type of writing arrangement. Using specific examples from science lessons, it demonstrates that writing does not automatically promote learning. For the cognitive activities mentioned, which support learning, to be actually elicited through writing, various prerequisites must be met. The presentation examines these on the basis of research findings and examples from an ongoing intervention study (KoSNaWi).
In lessons, pupils often write in order to record content discussed orally. This allows them to revisit the material later, for example when preparing for exams (= retention writing). However, writing can also support learning. It involves various cognitive processes that improve the storage of information in long-term memory (Silva and Limongi 2019, Graham and Hebert 2011): one must select the most important points, structure and elaborate thoughts, express connections in language, and so on. (= learning through writing). As plausible as this sounds from a cognitive theory perspective, empirical studies do not paint such a clear-cut picture (Bangert-Drowns et al. 2004; Graham et al. 2020). It can therefore be concluded that writing cannot be viewed merely as a cognitive phenomenon; rather, greater focus should be placed on its social component: what do writers do when writing is used as a learning tool in a learning context? Under what conditions can this be successful?
This talk provides insights into current research findings on learning through writing and the role played by the type of writing arrangement. Using specific examples from science lessons, it demonstrates that writing does not automatically promote learning. For the cognitive activities mentioned, which support learning, to be actually elicited through writing, various prerequisites must be met. The presentation examines these on the basis of research findings and examples from an ongoing intervention study (KoSNaWi).
Not least because writing, as a subject of research, straddles various disciplines – including linguistics, educational science and psychology – its empirical study is complex. Challenges become particularly apparent when high standards regarding psychometric and (inference) statistical quality criteria clash with limited feasibility in the context of (higher) education. The same applies, however, when theoretical models lose their relevance and interpretability due to insufficient integration into subject-specific didactics. Accordingly, these challenges can be systematised using a four-quadrant matrix that combines two axes: (1) methodological rigour versus feasibility, and (2) theoretical core versus subject-specific didactic integration.
In the methodological rigour × theoretical core quadrant, the focus is on developing conceptually precise and psychometrically robust operationalisations of central writing constructs (Steinhoff, Grabowski, & Becker-Mrotzek, 2017). The guiding question here is which aspects of writing remain difficult to assess whilst maintaining high standards of quality, and where quantitative and qualitative methods reach their limits. By way of example, a distinction is made between ‘hard’ factors (e.g. writing fluency, lexical or syntactic features, text structure) and ‘soft’ factors (e.g. motivation, self-efficacy, creativity), which can often only be measured indirectly and consequently raise issues of validity and comparability.
In the methodological rigour × subject-specific didactic embedding quadrant, the focus is on the application of sophisticated measurement and evaluation frameworks to didactically meaningful task formats and classroom arrangements (Jost, 2017). The key question is how core empirical quality criteria can be ensured without under-specifying theoretical constructs or compromising didactic fit (see, for example, Wild, 2023).
The ‘feasibility × theoretical core’ quadrant addresses the consequences of institutional and organisational constraints within the higher education sector for theory-based research. The central question is which research designs are viable under real-world conditions and how theoretical precision can be ensured despite pragmatic limitations (Grabowski, 2017; Rhoads, 2016). Typical challenges include, for example, a limited number of measurement time points, restricted rater resources or a short intervention period.
In the feasibility × subject-specific didactic embedding quadrant, the focus is on the development and evaluation of practice-oriented interventions under everyday conditions, whilst simultaneously specifying the underlying mechanisms of action. The key question is how measures that are compatible with subject-specific didactics can be implemented and evaluated without the assumptions regarding their effects remaining vague (cf. Jeuk, Schilcher, & Stöger et al., 2017). Relevant challenges arise, amongst other things, from adherence to the intervention protocol, treatment fidelity and the context-dependence of effects.
The aim is to develop a jointly supported framework that articulates the key challenges in writing research in line with the four-field matrix and derives from this criteria for research designs that are evidence-based, suitable for practical application and compatible with subject-specific didactics.
Grabowski, J. (2017). Requirements for research designs. In M. Becker-Mrotzek, J. Grabowski & T. Steinhoff (Eds.), Research Handbook of Empirical Writing Pedagogy (pp. 315–334). Waxmann.
Jeuk, S., Schilcher, A., Stöger, H. et al. (2017): Challenges and opportunities in collaboration between research and educational institutions. On the difficulty of collecting data in schools. In Mercator Institute for Language Promotion and German (eds.), << LOOKING BACK TO MOVE FORWARD >>. Perspectives on language education in teacher training and research (pp. 21–24). Mercator Institute.
Jost, J. (2017). Principles and methods of writing environments that promote learning. In M. Becker-Mrotzek, J. Grabowski, & T. Steinhoff (Eds.), Research Handbook on Empirical Writing Didactics (pp. 173–186). Waxmann.
Rhoads, C. (2016). The Implications of Contamination for Educational Experiments With Two Levels of Nesting. Journal of Research on Educational Effectiveness, 9(4), 531–555. doi.org/10.1080/19345747.2015.1086912
Steinhoff, T., Grabowski, J., & Becker-Mrotzek, M. (2017). Challenges in empirical writing didactics. In M. Becker-Mrotzek, J. Grabowski & T. Steinhoff (Eds.), Research Handbook on Empirical Writing Pedagogy (pp. 9–24). Waxmann.
Wild, J. (2023). Assessing written narrative. Results of the empirical validation of an instrument for assessing the quality of narrative written texts within the framework of the RESTLESS project. In N. Kruse, V. Lemke, T. Steinhoff, & A. Sturm (Eds.), Writing Instruction: Studies and Discourses on Writing and Text Production (pp. 271–289). Waxmann.
Not least because writing, as a subject of research, straddles various disciplines – including linguistics, educational science and psychology – its empirical study is complex. Challenges become particularly apparent when high standards regarding psychometric and (inference) statistical quality criteria clash with limited feasibility in the context of (higher) education. The same applies, however, when theoretical models lose their relevance and interpretability due to insufficient integration into subject-specific didactics. Accordingly, these challenges can be systematised using a four-quadrant matrix that combines two axes: (1) methodological rigour versus feasibility, and (2) theoretical core versus subject-specific didactic integration.
In the methodological rigour × theoretical core quadrant, the focus is on developing conceptually precise and psychometrically robust operationalisations of central writing constructs (Steinhoff, Grabowski, & Becker-Mrotzek, 2017). The guiding question here is which aspects of writing remain difficult to assess whilst maintaining high standards of quality, and where quantitative and qualitative methods reach their limits. By way of example, a distinction is made between ‘hard’ factors (e.g. writing fluency, lexical or syntactic features, text structure) and ‘soft’ factors (e.g. motivation, self-efficacy, creativity), which can often only be measured indirectly and consequently raise issues of validity and comparability.
In the methodological rigour × subject-specific didactic embedding quadrant, the focus is on the application of sophisticated measurement and evaluation frameworks to didactically meaningful task formats and classroom arrangements (Jost, 2017). The key question is how core empirical quality criteria can be ensured without under-specifying theoretical constructs or compromising didactic fit (see, for example, Wild, 2023).
The ‘feasibility × theoretical core’ quadrant addresses the consequences of institutional and organisational constraints within the higher education sector for theory-based research. The central question is which research designs are viable under real-world conditions and how theoretical precision can be ensured despite pragmatic limitations (Grabowski, 2017; Rhoads, 2016). Typical challenges include, for example, a limited number of measurement time points, restricted rater resources or a short intervention period.
In the feasibility × subject-specific didactic embedding quadrant, the focus is on the development and evaluation of practice-oriented interventions under everyday conditions, whilst simultaneously specifying the underlying mechanisms of action. The key question is how measures that are compatible with subject-specific didactics can be implemented and evaluated without the assumptions regarding their effects remaining vague (cf. Jeuk, Schilcher, & Stöger et al., 2017). Relevant challenges arise, amongst other things, from adherence to the intervention protocol, treatment fidelity and the context-dependence of effects.
The aim is to develop a jointly supported framework that articulates the key challenges in writing research in line with the four-field matrix and derives from this criteria for research designs that are evidence-based, suitable for practical application and compatible with subject-specific didactics.
Grabowski, J. (2017). Requirements for research designs. In M. Becker-Mrotzek, J. Grabowski & T. Steinhoff (Eds.), Research Handbook of Empirical Writing Pedagogy (pp. 315–334). Waxmann.
Jeuk, S., Schilcher, A., Stöger, H. et al. (2017): Challenges and opportunities in collaboration between research and educational institutions. On the difficulty of collecting data in schools. In Mercator Institute for Language Promotion and German (eds.), << LOOKING BACK TO MOVE FORWARD >>. Perspectives on language education in teacher training and research (pp. 21–24). Mercator Institute.
Jost, J. (2017). Principles and methods of writing environments that promote learning. In M. Becker-Mrotzek, J. Grabowski, & T. Steinhoff (Eds.), Research Handbook on Empirical Writing Didactics (pp. 173–186). Waxmann.
Rhoads, C. (2016). The Implications of Contamination for Educational Experiments With Two Levels of Nesting. Journal of Research on Educational Effectiveness, 9(4), 531–555. doi.org/10.1080/19345747.2015.1086912
Steinhoff, T., Grabowski, J., & Becker-Mrotzek, M. (2017). Challenges in empirical writing didactics. In M. Becker-Mrotzek, J. Grabowski & T. Steinhoff (Eds.), Research Handbook on Empirical Writing Pedagogy (pp. 9–24). Waxmann.
Wild, J. (2023). Assessing written narrative. Results of the empirical validation of an instrument for assessing the quality of narrative written texts within the framework of the RESTLESS project. In N. Kruse, V. Lemke, T. Steinhoff, & A. Sturm (Eds.), Writing Instruction: Studies and Discourses on Writing and Text Production (pp. 271–289). Waxmann.
The development and widespread adoption of generative AI marks a historic turning point in science communication, as it has brought an end to the human monopoly on the production of scientific texts. Phenomena such as the Deep Research features of ChatGPT and Gemini illustrate that generative AI can rapidly produce texts that fulfil the essential characteristics of typical scientific publications, including complex ones. What does this mean for the future of scientific writing? We would like to discuss three possible scenarios, along with their implications for academic theses: Should human scientific writing be handed over to generative AI (‘delegation’)? Should it be preserved in distinction from it (‘conservation’)? Or should it be reshaped in response to it (‘transformation’)?
Bibliography
Lehnen, K. (2025): Media Constellations in Flux: University Writing Practices and AI. To be published in: Ballestracci, Sabrina/Foschi, Marina/Lefèvre, Michel/Steinhoff, Torsten (eds.): Digital Academic Communication in Germany, France and Italy: The Case of German Linguistics. Florence: Lingue e Letterature d’Oriente e d’Occidente. www.doi.org/10.13140/RG.2.2.15573.49129
Steinhoff, Torsten (2026): On the Future of Writing in Schools and Universities: Delegation, Preservation or Transformation? To be published in: Ballestracci, Sabrina/Foschi, Marina/Lefèvre, Michel/Steinhoff, Torsten (eds.): Digital Academic Communication in Germany, France and Italy: The Case of German Linguistics. Florence: Lingue e Letterature d’Oriente e d’Occidente. Preprint. http://www.doi.org/10.13140/RG.2.2.14937.17761/1
The development and widespread adoption of generative AI marks a historic turning point in science communication, as it has brought an end to the human monopoly on the production of scientific texts. Phenomena such as the Deep Research features of ChatGPT and Gemini illustrate that generative AI can rapidly produce texts that fulfil the essential characteristics of typical scientific publications, including complex ones. What does this mean for the future of scientific writing? We would like to discuss three possible scenarios, along with their implications for academic theses: Should human scientific writing be handed over to generative AI (‘delegation’)? Should it be preserved in distinction from it (‘conservation’)? Or should it be reshaped in response to it (‘transformation’)?
Bibliography
Lehnen, K. (2025): Media Constellations in Flux: University Writing Practices and AI. To be published in: Ballestracci, Sabrina/Foschi, Marina/Lefèvre, Michel/Steinhoff, Torsten (eds.): Digital Academic Communication in Germany, France and Italy: The Case of German Linguistics. Florence: Lingue e Letterature d’Oriente e d’Occidente. www.doi.org/10.13140/RG.2.2.15573.49129
Steinhoff, Torsten (2026): On the Future of Writing in Schools and Universities: Delegation, Preservation or Transformation? To be published in: Ballestracci, Sabrina/Foschi, Marina/Lefèvre, Michel/Steinhoff, Torsten (eds.): Digital Academic Communication in Germany, France and Italy: The Case of German Linguistics. Florence: Lingue e Letterature d’Oriente e d’Occidente. Preprint. http://www.doi.org/10.13140/RG.2.2.14937.17761/1
The workshop is aimed at participants who are in the planning phase of an intervention study and offers a concise introduction to key issues relating to study design and data collection.
Together, we will explore how research questions can be translated into suitable designs, the role played by comparison and control groups, and how to schedule measurement timepoints effectively. Using examples, we will discuss typical challenges encountered when planning data collection and demonstrate the implications these decisions have for subsequent analysis.
The aim of the workshop is to develop a sound methodological foundation for planning one’s own intervention studies and to prepare participants for further quantitative analyses.
The workshop is aimed at participants who are in the planning phase of an intervention study and offers a concise introduction to key issues relating to study design and data collection.
Together, we will explore how research questions can be translated into suitable designs, the role played by comparison and control groups, and how to schedule measurement timepoints effectively. Using examples, we will discuss typical challenges encountered when planning data collection and demonstrate the implications these decisions have for subsequent analysis.
The aim of the workshop is to develop a sound methodological foundation for planning one’s own intervention studies and to prepare participants for further quantitative analyses.
Quantitative analyses are also an important tool in language teaching for investigating the effects of interventions. This workshop offers an introduction to fundamental approaches to quantitative data analysis, with a particular focus on regression methods in intervention studies.
Using sample data, the workshop will provide an insight into regression analyses that can be used to investigate the effects of interventions whilst taking covariates into account. It will address the assumptions underlying these analyses and which conclusions can – and cannot – be drawn from the results. The analyses and presentations will be carried out using the statistical software JASP as an example; participants are asked to install this in advance.
The aim of the workshop is to develop a basic understanding of how quantitative analyses of intervention studies work, how to interpret the results of regression analyses, and how to realistically assess their significance. The focus is not on participants carrying out complex analyses independently.
Quantitative analyses are also an important tool in language teaching for investigating the effects of interventions. This workshop offers an introduction to fundamental approaches to quantitative data analysis, with a particular focus on regression methods in intervention studies.
Using sample data, the workshop will provide an insight into regression analyses that can be used to investigate the effects of interventions whilst taking covariates into account. It will address the assumptions underlying these analyses and which conclusions can – and cannot – be drawn from the results. The analyses and presentations will be carried out using the statistical software JASP as an example; participants are asked to install this in advance.
The aim of the workshop is to develop a basic understanding of how quantitative analyses of intervention studies work, how to interpret the results of regression analyses, and how to realistically assess their significance. The focus is not on participants carrying out complex analyses independently.
The Open Science Session is designed as an open forum for research projects. It is a new format introduced at this year’s summer school and will take place on Saturday morning from 9 am to 12 noon. It offers researchers at all stages of their careers – from initial project ideas to well-advanced PhD theses or postdoctoral projects – the opportunity to present their research, discuss open questions and receive methodological feedback from the community.
Participants can register for one of three presentation formats:
The Open Science Session is a new format that we have deliberately designed to be accessible and open to collaboration: whether it’s an initial idea, a chapter in the making, a methodological challenge or a project nearing completion – you are all invited to take part in the discussion and raise the profile of your research.
Please send the fully completed form to dies-tagung@ph-karlsruhe.de by 15 May 2026.
Once you have registered and submitted an abstract, a programme for the entire time slot will be drawn up and published here.
Individual support for dissertations
As part of the summer school, you will have the opportunity to further develop your thesis in the field of didactic-empirical writing research through a dedicated advisory colloquium. The seminars take place in small groups of up to three postgraduate students and are facilitated by two members of the dieS group, who will work with you to examine your research, your research questions and possible next steps.
This format provides space for:
To enable us to put together a bespoke advisory package for you, we require some details about your research project. Please also indicate which dieS members you would like to receive advice from. We will do our best to accommodate your preferences.
Please send the fully completed form by 15 May 2026 to: dies-tagung@ph-karlsruhe.de.
Registration for the conference is via LimeSurvey. The registration deadline is 15 May 2026.
The conference fees cover all meals and refreshments throughout the summer school: coffee breaks, drinks, refreshments at the social evening on Thursday, and a snack for the journey home. Only dinner on Friday is on a self-pay basis, so that all participants can choose flexibly.
Participation fees are divided into two categories:
Rooms are available at various centrally located hotels. Full details of each hotel are listed here.
| Hotel | Price per night | Allocation available | Distance to the University of Education |
| Hotel Ambassador | €93.00 including breakfast and city tax | until 7 May 26 | approx. 10 mins’ walk |
| Berliner Hof | €86.00 including breakfast | until 15 April 2026 | approx. 5 mins’ walk |
| Hotel am Markt | €95.00 including breakfast | until 15 April 26 | approx. 12 mins’ walk |
| Motel One | €88.00 excluding breakfast (€17.90) | until 19 May 2026 ( → please use the enquiry form!) | approx. 15 mins’ walk |
| Hotelwelt Kübler | €68.00 excluding breakfast (€15.00) | until 7 May 2026 | approx. 2 mins’ walk |
The information in the programme should be understood as follows (by way of example):
1.013 (Building 1, Floor 0, Room 13)
3,101 (Building 3, Floor 1, Room 1)
Organisation
Karlsruhe University of Education
Building
3
Room
209
Phone
Availability
By arrangement
Organisation
Karlsruhe University of Education
Building
3
Room
102
Phone
Availability
See Stud.IP
Organisation
Karlsruhe University of Education
Building
3
Room
105
Phone
Availability
Mondays to Wednesdays (and by appointment)
We look forward to seeing you!
Updated on 1 October 2026 by Nadine Anskeit
