{"id":60575,"date":"2024-09-20T15:05:03","date_gmt":"2024-09-20T13:05:03","guid":{"rendered":"https:\/\/www.audiotranskription.de\/resources\/hybrid-interpretation-group\/"},"modified":"2026-08-12T11:19:15","modified_gmt":"2026-08-12T09:19:15","slug":"hybrid-interpretation-group","status":"publish","type":"page","link":"https:\/\/www.audiotranskription.de\/en\/resources\/hybrid-interpretation-group\/","title":{"rendered":"Hybrid interpretation"},"content":{"rendered":"

[vc_section][vc_row row_padding=””][vc_column]

\"audiotranskription<\/div>

AI in interpretative qualitative research<\/p>

Hybrid interpretation<\/span><\/h1><\/div><\/div><\/div>[\/vc_column][\/vc_row][\/vc_section][vc_section row_padding=”vpt-7 vpb-7″][vc_row row_padding=”custom” row_padding_bottom=”vpb-3″][vc_column width=”3\/4″ col_sticky=”” offset=”vc_col-sm-offset-2″]

Quick access

<\/div><\/div><\/div>[\/vc_column][\/vc_row][vc_row row_padding=””][vc_column col_sticky=””]
<\/div>

methodological paper<\/span><\/p>

Hybrid Interpretation of Text-Based Data Using Dialogically Integrated LLMs (Kr\u00e4hnke, U., Pehl, T., & Dresing, T. (2025))<\/p>

Jetzt lesen<\/a><\/div><\/div><\/div><\/div>
<\/div>

Practical guidance<\/span><\/p>

Introduction to hybrid interpretation with three LLMs. A practical step-by-step guide<\/p>

Download PDF<\/a><\/div><\/div><\/div><\/div>
<\/div>

Webinars<\/span><\/p>

See hybrid interpretation live or help shape it in a research workshop? Visit our suitable webinars\/courses!<\/p>

Jetzt anmelden<\/a><\/div><\/div><\/div><\/div>
<\/div>

Prompts & example<\/span><\/p>

The prompt templates used and full documentation of a hybrid interpretation of a text segment can be found here:<\/p>

Download DOCX<\/a><\/div><\/div><\/div><\/div>
<\/div>

GDPR-compliant hybrid interpretation<\/span><\/p>

A workshop report on local interpretation with four LLMs on a MacBook Pro with Gemma 3, Qwen 3, Mistral 3.1 and Llama 3.3<\/p>

Jetzt lesen<\/a><\/div><\/div><\/div><\/div>
<\/div>

Serendipity prompting <\/span><\/p>

The methodized curiosity: Using LLMs as a meeutic impulse generator instead of an answer machine in the \"phase 0\" of the analysis.<\/p>

Info-Seite & Prompt<\/a><\/div><\/div><\/div><\/div><\/div><\/div>[\/vc_column][\/vc_row][\/vc_section][vc_section row_padding=”vpt-7 vpb-7″][vc_row row_padding=”custom” row_padding_bottom=”vpb-7″][vc_column width=”2\/3″ col_sticky=”” offset=”vc_col-sm-offset-2″][\/vc_column][\/vc_row][vc_row row_padding=”custom” row_padding_bottom=”vpb-7″][vc_column width=”1\/2″ col_sticky=””][vc_single_image image=”59490″ img_size=”large”][\/vc_column][vc_column width=”1\/2″ col_sticky=”” offset=”vc_col-lg-5 vc_col-sm-offset-1″]

In dialog with 3 AIs
<\/div><\/h2><\/div><\/div>[vc_column_text]Hybrid interpretation is an elaborate method for integrating AI into interpretative qualitative research (Kr\u00e4hnke, <\/a>Dresing, <\/a>Pehl 2025)<\/a>. It enables an insight-evoking dialog between several AIs (Large Language Models \/ LLMs) and the researcher, with the aim of a profound and comprehensible interpretation of the text. This article describes the practical implementation of this procedure, which can be implemented largely free of charge via a browser. <\/strong>[\/vc_column_text][\/vc_column][\/vc_row][vc_row row_padding=””][vc_column width=”2\/3″ col_sticky=”” offset=”vc_col-md-7 vc_col-sm-offset-2″]

It works better than with ChatGPT alone
<\/div><\/h2><\/div><\/div>[vc_column_text]Previously widespread approaches to the use of AI in qualitative social research, such as the use of ChatGPT, often only deliver superficial results (keyword “make a summary”). In addition, strategies of elaborate input to the LLM (prompt engineering) are pursued, which, however, require a high level of development expertise. Nevertheless, automatically generated interpretations by individual prompts are usually not particularly differentiated.<\/span><\/p>\n

With the hybrid interpretation described here, we have found that we can carry out high-quality and differentiated analyses with AI. We ourselves are always surprised and delighted by the depth and sophistication of the interpretation suggestions. This awakens in us the desire to analyze and interpret again and again.<\/p>\n

Here we describe the features of the approach and provide instructions on how anyone interested can try it out for themselves – including using tools that are available free of charge.[\/vc_column_text][\/vc_column][\/vc_row][\/vc_section][vc_section row_padding=”vpt-7 vpb-7″ row_bg_color=”bg–color–bg-1″ row_bg_image_size=”bg–image–size–cover”][vc_row content_placement=”bottom” row_padding=”custom” row_padding_bottom=”vpb-5″][vc_column width=”2\/3″ col_sticky=”” offset=”vc_col-sm-offset-2″]

6 reasons that make this approach so special<\/p>

Hybrid interpretation with dialogic-moderated LLMs
<\/div><\/h2><\/div><\/div>[\/vc_column][\/vc_row][vc_row row_padding=”custom” row_padding_bottom=”vpb-7″][vc_column width=”1\/2″ col_sticky=”” offset=”vc_col-lg-5″]

1. Simulation of authentic interpretation group instead of directional prompting<\/h3><\/div><\/div>[vc_column_text]The approach reduces the need for elaborate prompt design and instead enables a natural discussion language when dealing with AI. This makes the method more accessible and less technically demanding.[\/vc_column_text][\/vc_column][vc_column width=”1\/2″ col_sticky=”” offset=”vc_col-lg-offset-1″][vc_single_image image=”59731″ img_size=”large”][\/vc_column][\/vc_row][vc_row row_padding=”custom” row_padding_bottom=”vpb-7″][vc_column width=”1\/2″ col_sticky=”” offset=”vc_col-lg-5″]

2. Multiple AI models<\/h3><\/div><\/div>[vc_column_text]Three different LLMs (currently ChatGPT, Claude and Gemini) are integrated into the research process and moderated by the researcher. Studies show that the interlinking of different LLMs leads to an improvement in the quality of the output of each individual LLM. This significantly increases the overall quality of the analysis. This variance increases the diversity of perspectives due to the different “bias” of the LLMs involved. The confrontation with different points of view encourages more differentiated answers and mutual reference, which increases the diversity of perspectives. <\/span>[\/vc_column_text][\/vc_column][vc_column width=”1\/2″ col_sticky=”” offset=”vc_col-lg-offset-1″][vc_single_image image=”59709″ img_size=”large”][\/vc_column][\/vc_row][vc_row row_padding=”custom” row_padding_bottom=”vpb-7″][vc_column width=”1\/2″ col_sticky=”” offset=”vc_col-lg-5″]

3. Agency remains with the researcher through active moderation role<\/h3><\/div><\/div>[vc_column_text]The researcher takes on an orchestrating role as a moderator and remains actively involved in the interpretation process. She critically examines the interpretations offered, asks specific questions, gives instructions and reflects on the contributions of the AI models. This active control ensures that the analysis is targeted and in line with the research interest. <\/span>[\/vc_column_text][\/vc_column][vc_column width=”1\/2″ col_sticky=”” offset=”vc_col-lg-offset-1″][\/vc_column][\/vc_row][vc_row row_padding=”custom” row_padding_bottom=”vpb-7″][vc_column width=”1\/2″ col_sticky=”” col_pp_sm=”vc_col-sm-push-6″ col_pp_lg=”vc_col-lg-push-7″ offset=”vc_col-lg-5″]

4. Iterative dialog between LLM and researchers<\/h3><\/div><\/div>[vc_column_text]The analysis takes place in several rounds, with each round imitating the style of a lively group discussion. The AI models are confronted with the answers of the other LLMs and the assessments of the researcher. In the process, the analyses are deepened, argued in a more differentiated way and their interpretations are examined. This iterative approach results in an increasingly refined and multi-layered interpretation of the research material.<\/span>[\/vc_column_text][\/vc_column][vc_column width=”1\/2″ col_sticky=”” col_pp_sm=”vc_col-sm-pull-6″ offset=”vc_col-lg-offset-1″][vc_single_image image=”59713″ img_size=”large”][\/vc_column][\/vc_row][vc_row row_padding=”custom” row_padding_bottom=”vpb-7″][vc_column width=”1\/2″ col_sticky=”” col_pp_sm=”vc_col-sm-push-6″ col_pp_lg=”vc_col-lg-push-7″ offset=”vc_col-lg-5″]

5. Documentation for intersubjective traceability<\/h3><\/div><\/div>[vc_column_text]The complete documentation of the process addresses the often criticized “black box” problem of AI usage. This enables an intersubjective comprehensibility of the interpretation and makes the individual work performance of the researchers in the context of the orchestration of the work process clear. The researcher’s examination of the various interpretative approaches is documented transparently. It can be retraced in terms of plausibility and quality, which increases the validity of the research.<\/span>[\/vc_column_text][\/vc_column][vc_column width=”1\/2″ col_sticky=”” col_pp_sm=”vc_col-sm-pull-6″ offset=”vc_col-lg-offset-1″][vc_single_image image=”59715″ img_size=”large”][\/vc_column][\/vc_row][vc_row row_padding=”custom”][vc_column width=”2\/3″ col_sticky=”” offset=”vc_col-md-7 vc_col-sm-offset-2″]

6. Didactization of qualitative methods through low-threshold practical experience<\/h3><\/div><\/div>[vc_column_text]Hybrid interpretation allows students to gain their first practical experience of qualitative interpretation. The integrated LLMs act as a kind of sparring partner for the interpretation work. <\/span><\/p>\n

Hybrid interpretation helps students to get to know, apply and critically evaluate different perspectives on text interpretation. You do not simply receive a finished result or automatic coding, but rather material for your own examination of the text. <\/span><\/p>\n

By actively engaging with different approaches to interpretation, they are encouraged to develop their own, well-founded positions. This process not only promotes analytical thinking, but also the ability to synthesize and evaluate complex interpretations.<\/span>[\/vc_column_text][vc_row_inner row_padding=”” row_bg_color=”bg–color–bg-1″ row_boxpadding=”vc_boxpadding”][vc_column_inner]

Podcast 'Methods:Suitcase' from 18.02.2025<\/p>

Episode 43: Interpreting qualitative data with generative AI
<\/div><\/h2><\/div><\/div>[vc_raw_html css=””]PGlmcmFtZSBzdHlsZT0iYm9yZGVyLXJhZGl1czoxMnB4IiBzcmM9Imh0dHBzOi8vb3Blbi5zcG90aWZ5LmNvbS9lbWJlZC9lcGlzb2RlLzQyamZidFRSQmZsSEd0SHpIRTY0U3o\/dXRtX3NvdXJjZT1nZW5lcmF0b3IiIHdpZHRoPSIxMDAlIiBoZWlnaHQ9IjE1MiIgZnJhbWVib3JkZXI9IjAiIGFsbG93ZnVsbHNjcmVlbj0iIiBhbGxvdz0iYXV0b3BsYXk7IGNsaXBib2FyZC13cml0ZTsgZW5jcnlwdGVkLW1lZGlhOyBmdWxsc2NyZWVuOyBwaWN0dXJlLWluLXBpY3R1cmUiIGxvYWRpbmc9ImxhenkiPjwvaWZyYW1lPg==[\/vc_raw_html][vc_column_text elem_margin_top=”vmt-1″]In this episode, Thorsten Dresing and Anna-Barbara Heindl discuss how generative AI works in principle and clarify what impact this has on the use scenario of AI in reconstructive research. Thorsten and his colleagues suggest using Large Language Models such as ChatGPT, Claude and Gemini to get inspiration and irritation for different interpretation possibilities in parallel to their own interpretation work – analogous to human interpretation groups.[\/vc_column_text][\/vc_column_inner][\/vc_row_inner][\/vc_column][\/vc_row][\/vc_section][vc_row content_placement=”bottom” row_padding=”custom” row_maxwidth=””][vc_column width=”11\/12″ col_sticky=””][vc_single_image image=”59663″ img_size=”full”][\/vc_column][\/vc_row][vc_section row_padding=”vpt-7 vpb-7″ row_bg_image_size=”bg–image–size–cover”][vc_row content_placement=”bottom” row_padding=”custom” row_padding_bottom=”vpb-7″][vc_column width=”2\/3″ col_sticky=”” offset=”vc_col-sm-offset-2″]

Overview of the process
<\/div><\/h2><\/div><\/div>[vc_column_text]First of all, the user needs (partly) free accounts with all three LLMs and chooses a short one. A simple, open starting question is formulated for the LLMs to initiate the analysis process.<\/strong><\/p>\n

In the first round of analysis, the three LLMs describe the text from different perspectives. The researcher then reflects on the statements and derives the next steps. <\/span><\/p>\n

This is followed by several iterative discussion rounds in which the researcher deepens the interpretations and discusses them with the LLMs. This process is moderated and controlled by interim comments and reflections from the user until the user determines the conclusion of the analysis. <\/span><\/p>\n

The researcher identifies key aspects and develops their own interpretative perspective. Finally, a conclusion is drawn up that summarizes the interpretation developed.<\/span><\/p>\n

To ensure traceability, the entire course of the discussion is documented and commented on in Word or an f4 project. <\/span>[\/vc_column_text][\/vc_column][\/vc_row][vc_row row_padding=”custom” row_padding_bottom=”vpb-5″][vc_column width=”1\/2″ col_sticky=””]

From free of charge to 60\u20ac per month
<\/div><\/h2><\/div><\/div>[vc_column_text]During our test runs, we had by far the best experience with the following three LLMs in terms of the appealing quality of the interpretation suggestions. To use these, each person needs their own account, initially free of charge. You can simply click on the links and register. (as at Dec 2024)<\/span><\/p>\n