by Christina Silver, Director of the CAQDAS Networking Project and Host of #CAQDASchat podcast
I’ve published 12 episodes of my #CAQDASchat podcast since November 2022, chatting with qualitative software developers, researchers, methodologists and teachers. I started the podcast for several reasons, not least to capture the thoughts of experts in the field about the history and development of computer-assisted qualitative analysis and about where it is headed. To that end, I conclude each episode by asking guests these two questions:
- if you had to pinpoint one moment that had changed the course of computer-assisted qualitative data analysis over the past 30 years, what would that be?
- what do you think needs to happen in the field moving forwards to further advance methodology?
Capturing perspectives on the history and future of CAQDAS
I’ve thus captured perspectives about the history and future of the field from a small but diverse group of eminent voices in the QDA space. As I’m changing the format for the podcast in 2025, I thought it would be interesting – and fun – to reflect on these perspectives collectively.
It’s a particularly good moment for this because the impact that Generative-AI developments are having on the practice of QDA are being increasingly discussed in a variety of forums (see for example our new QualAI pages, recent discussions on the Qual-Software Jiscmail List, other posts on this blog, and journal publications), and coincidentally the first #CAQDASchat podcast episode aired around the same time that OpenAI released ChatGPT to the world and for some people, everything began changing.
Reflections on ‘past moments’
This post reflects on the first question: If you had to pinpoint one moment that had changed the course of computer assisted qualitative data analysis in that time, what would that be? I’ll reflect on the second question in a later post
The 13 guests on which these reflections are based are listed in the table below.
What’s the point?
As I’m steeped in the Five-Level QDA method in which knowing your strategies always proceeds jumping into your tactics, I began by thinking about my objectives. Why am I doing this? Beyond the overarching reasons of interest and fun I had two broad objectives:
- to understand how this group of CAQDAS developers, methodologists, teachers, and researchers conceptualise the history of the field, and
- to raise awareness about these podcasts and encourage more researchers to listen to, or watch them[i] because they contain much compelling and significant discussion.
Within these objectives I decided my specific goal was to capture what the first 13 #CAQDASchat podcast guests think are the most significant moments in the past circa 30 years that changed the course of computer assisted qualitative data analysis.
In terms of how Harry Wolcott (1994) conceptualises it, I had a descriptive purpose: to capture “what is going on here?”. Given the intention to encourage others to listen to or watch the podcasts, this feels enough. I’m stopping short of an in-depth analysis (“how things work”) or interpretation (“what does it mean”). If you’re interested I how I went about the descriptive process, scroll down to the end of this post where I share the steps.
I’m lucky to know many eminent people in the CAQDAS-field and to have had the privilege of having so many informal chats with them over the years, at conferences, during trainings and webinars, whilst beta-testing products, and so on. My #CAQDASchat podcast is an opportunity to more formally capture those hugely valuable conversations and share them with the CAQDAS community.
I hope that this post will spark your interest to listen to or watch one or more episodes of the podcast.
Many significant ‘past moments’
The first thing of note is that several podcast guests found it tricky to restrict themselves to discussing only one moment that they felt had changed the course of computer-assisted qualitative analysis, and several talked about two or more. In fact, you may think that asking them to distil 30 years of history down to just one significant moment is somewhat unfair. After all, 30 years is a long time and much has happened. My intention is asking them to do so for the podcasts was to reveal what for them captures the essence of historical salience in the CAQDAS field.
Technical Developments and the QDA process
Almost all the significant moments discussed related to two topics: “Technical developments” and/or “the QDA process”, defined in Table 5 below. The only other comments that did not fall into either of those two topics were a couple of comments coded to “Changed thinking/practice” (defined as “discussion about changes in one’s own thinking or practice with respect to QDA and the use of CAQDAS-packages” see Table 4).
I grouped “Technical Developments” into:
- General Technical Developments (those that were not directly related to the CAQDAS field, but which had an impact on it), and
- Specific CAQDAS capability (technical developments directly relating to the CAQDAS-field and/or CAQDAS features).
Given we were discussing key moments in the CAQDAS-field it is unsurprising that most guests discussed some kind of technical development. More interesting is that almost always these technical developments were discussed in terms of their influence on the process of QDA. It was not just the technical developments themselves that was of note, but their influence on how qualitative researchers go about their craft. I identified “the QDA process” as relating to three aspects:
- Opening up QDA (discussion about the process of QDA becoming more open in terms of accessibility)
- Collaboration (discussion about researchers’ ability to and practice of working in teams to undertake QDA, with or without the use of CAQDAS-packages or other technology)
- Rigorous Process (discussion about the quality of qualitative analysis in relation to the potential contribution of CAQDAS-packages)
Table 2 summarizes the key moments across all podcast guests in five categories: General technical developments, Specific CAQDAS capability, Opening up QDA, Collaboration, and Rigorous process.
The following Tables and Figures provide a variety of visualizations, the relative frequencies of each sub-category, summaries of each guest’s responses topics mentioned. I end by describing the steps of my process and outlined my planned next steps with this project.
Table 2: Summary of findings across podcast guests
| General technical developments Broad developments that have impacted the practice of QDA and/or the features of CAQDAS-packages, which may or may not have been developed for those purposes. 9 of the 13 guests mentioned at least one general technical development in relation to ‘past moments’ and several mentioned more than one (see Table 4). | Four general technical developments were mentioned by one guest each: the internet, the advent of graphical user interfaces, the nature and impact of relational databases and advances in automated speech-to-text capabilities. In addition, three guests discussed the impact of developments in increasingly flexible data formats (e.g. use of rich text format instead of plain text; increasing availability of multimedia forms of data, and cross-platform video technology more generally). Two guests discussed the rise of Generative-AI technologies based on Large Language Models (LLMs) and the impact these technologies are having on the practice of qualitative analysis. |
| Specific CAQDAS capability Specific CAQDAS tools and/or their use within particular CAQDAS-packages. 6 of the 13 guests mentioned at least one specific CAQDAS capability in relation to ‘past moments’ and two of them mentioned more than one (see Table 4). | Four guests discussed the development of the REFI-QDA exchange standard and its impact on the practice of QDA in terms of fostering collaboration. One of these guests also discussed cross-platform and web-based capabilities in the context of CAQDAS packages more generally as well.
The other specific CAQDAS capabilities mentioned in terms of ‘past moments’ were the development of visual mapping tools within CAQDAS packages, the specific capability of independent quotations in a particular program (ATLAS.ti), and the moment when the idea of developing computer software to facilitate qualitative data analysis was first mooted. Each of these were mentioned by one guest each. |
| Opening up QDA The process of QDA becoming more open in terms of accessibility. 12 of the 13 guests mentioned opening up QDA in relation to ‘past moments’ (see Table 4) | The idea of ‘opening up QDA’ was relation to both what was being ‘opened up’ and for whom. Most guests that mentioned ‘opening up QDA’ in relation to ‘past moments’ did so in relation to general technical developments and the impact these had on qualitative researchers. This includes opening up in terms of: – whether and how CAQDAS developers harnessed general technical developments or not, and the influence on the popularity of CAQDAS packages when they did (one guest) – understanding amongst researchers of the potential of CAQDAS capability for making connections and how this was fuelled by similar technical capabilities in other areas of life (one guest) – technical capabilities fostering collaboration and connection (e.g. the internet and cross-platform compatibility) (two guests) – practicalities and productivities (including saving time and being able to work with other researchers) that advances in technology afford researchers (three guests) – the development of AI capabilities and their impact on the possibilities for QDA (two guests) Two guests related the idea of ‘opening up QDA’ to their own thinking about qualitative analysis and the use of CAQDAS packages. These guest recounted stories about their experiences as illustrations of how specific technologies or events changed opened up QDA for them personally. |
| Collaboration Discussion about researchers ability to and practice of working in teams to undertake QDA, with or without the use of CAQDAS-packages or other technology. 5 of the 13 guests mentioned collaboration in relation to ‘past moments’ (see Table 4) | Technical capabilities fostering the ability for researchers to collaborate with one another on qualitative analysis projects (four guests). Three guests related these aspects of collaboration specifically to the development of the REFI-QDA exchange standard (a specific CAQDAS capability) and one to a general technical development (the internet) One guest discussed collaboration within the CAQDAS community, particularly amongst developers at conferences and other events |
| Rigorous process Discussion about the quality of qualitative analysis in relation to the potential contribution of CAQDAS-packages. 2 of the 13 guests mentioned ‘rigorous process’ in relation to ‘past moments’ (see Table 4) | Rigorous process relates to the ability to be transparent about analytic process via the use of CAQDAS packages. This included discussion about documenting process, being able to easily access connections within and between concepts and thereby illustrating how conceptual ideas are derived from qualitative data. One guest discussed rigour in process in general terms relating to the use of CAQDAS, the other in more specific terms relation to particular CAQDAS tools (mapping) |
Figure 1 illustrates the relative frequency of the sub-categories of General Technical Developments (coloured green), Specific CAQDAS capabilities (coloured orange), QDA process (coloured purple), and “Changed thinking/practice” (coloured blue) which a couple of guests related to aspects of Specific CAQDAS capabilities (see also Figure 2.).
Figure 1. Cloud visualization showing the relative frequency of mentions of sub-categories
Table 3. presents my summaries of each guest’s responses to the ‘past moments’ question, Table 3 presents topics mentioned by each guest, and Table 4. lists the codes grouped into their categories, along with working definitions.
Table 3. Overview of Guest’s responses to the
Past Moments question: My summaries and codes (subsequently) applied
| CAQDAS chat guests | My summary of guest’s Past Moments responses | My codes |
| ep001 Nigel Fielding Role: Methodologist Country: UK ep001 Ray Lee Role: Methodologist Country: UK | The ability to transfer analysed data from one CAQDAS package to another, which allows “fruitful collaboration” among researchers with different software preferences. This is viewed as increasingly important because of international collaborations, and the implications of the Covid pandemic. Developing the REFI exchange standard was challenging technically and because of the commercial pressures of competing vendors, but valuable for the research community. It was also something that had “long been dreamed of” by Nigel Fielding and Ray Lee. Ray agrees with Nigel about the importance of REFI-QDA, but highlights the distinction between manual qualitative analysis and the use of computer software in discussing significant past moments, suggesting that the idea manual methods are somehow more noble has fallen by the wayside as the capabilities of software have raised the bar in terms of rigour and illustrating process. In addition is the moment when someone first thought there must be a better way of doing analysis than with piles of paper. | Collaboration REFI-QDA exchange standard Opening up QDA Collaboration REFI-QDA exchange standard Initial Development Rigorous Process |
| ep002-Udo Kuckartz Role: Developer Country: Germany | CAQDAS feature development has been triggered by external technical developments, significant examples being the move to graphical user-interfaces, using rich text format text and handling PDF files – these external changes affected how CAQDAS-packages could be used for broader purposes, their usability and attractiveness to researchers | User Interface Opening Up QDA |
| ep003 Kristi Jackson Role: Methodologist Country: USA | The analytic possibilities provided by relational databases in comparison to two-dimensional spreadsheets – their capacity to cross-reference without duplication, which is related to interconnectivity, and the ubiquitousness of relational databases in everyday life (e.g. cell phones) has opened up the potential for the power in CAQDAS-packages | Relational Database Opening Up QDA |
| ep04 Stu Shulman Role: Developer Country: USA | The internet and its widespread usage, because of the connections between researcher’s and opening up of new datatypes for analysis that it afforded | Internet Collaboration Opening Up QDA |
| ep005 Susanne Friese Role: Developer Country: Germany | Recent advances in automated transcription capabilities – especially improved accuracy – even for different languages and dialects – the time this saves and the lessening of the “entrance barrier” to using computers for analysis | Automated transcription Opening Up QDA |
| ep06 Trena Paulus Role: Methodologist Country: USA | Capabilities that facilitated analytic collaboration, the key developments of which were cross-platform CAQDAS-tools which facilitated simultaneous analysis, and the ability to exchange coded data between CAQDAS-programs via the REFI standard. The ability to work together and across platforms is important because many are not lone researchers. In addition, teachers benefit from these capabilities. These capabilities encouraged people who had previously been reluctant, to be more open to what is possible. | Cross-platform or Web-based Collaboration Opening Up QDA REFI-QDA exchange standard |
| ep007 Jack Bowen Role: Developer Country: UK | The technique of “reinforcement learning” that enables the outputs of AI models to be aligned to the preferences of humans, which underlies the increased performance of GPTs | Developments in Gen-AI Opening Up QDA |
| ep008 Eli Lieber Role: Developer Country: USA | Conference gatherings provide opportunity for developers to discuss developments; prompting different ways of thinking and doing from which a lot was learned. Enables a broader understanding of the landscape, which helps to balance the different roles in terms of methods/scholarship/software development and see how the pieces fit together. He is grateful for the CNP for providing those spaces. | Collaboration REFI-QDA exchange standard Opening Up QDA |
| ep009 Liz-Pope Role: Methodologist Country: USA | Gen-AI conversation is shaking up people’s thinking about computer-assisted QDA, to the degree that thought about it is being ‘forced’. This is expressed in the context of being in the field for “only” 9 years. This is both a good and a bad thing. Change is inevitable and can be good, but the concern is that it will lead to a ‘schism’ in the field, so we need to observe what happens and raise awareness. | Developments in Gen-AI Opening Up QDA |
| ep010 Janet Salmons Role: Methodologist Country: USA | Movement from primarily text for QDA to multimedia and multimodal data forms which requires different types of, more holistic analysis. The work of Gunter Kress who emphasised (ahead of his time) the need to understand and take into account the nature of the kinds of data we now have available: “multimodal literacy”. This is not only about data but also how we communicate, which bring opportunities and challenges | More flexible data types Opening up QDA |
| ep011 Ann Lewins Role: Methodologist Country: UK | The independence of the quotation as an architectural element in ATLAS.ti and the possibility to link them separately from coding opened up the potential of software; that work could be done without coding | Independent Quotation Architecture Opening Up QDA |
| ep012 David Woods Role: Developer Country: USA | When cross-platform video technology worked such that same video formats were compatible on MacOS and Windows, which made video much easier to use for research purposes | More Flexible Data Types Opening Up QDA |
Table 4. Topics mentioned by each guest, grouped by category
Technical developments affect the practice of QDA
The relationships between QDA processes (purple codes), General Technical Developments (green codes), and Specific CAQDAS capabilities (orange codes) are illustrated in Figure 1. The thickness of the connections indicates frequency of co-occurrence in coding.
In viewing this map it’s important to remember that there were 13 guests and they had been asked to discuss one significant past moment. Therefore, the frequency of co-occurrence is low. In addition, Figure 2 shows that for a couple of podcast guests, technical developments led to changes in their personal thinking or practice (blue code).
Figure 2. Visual representation of the co-occurrences between sub-categories across the 12 discussions about ‘past moments’
Given that my podcast guests are CAQDAS developers, methodologists, researchers, and teachers, it’s interesting to consider if their perceptions are related to these roles. Although several guests have more than one of these roles, I categorised each of them as either developer/methodologist/researcher/teacher based on my interpretation of the main emphasis of their contribution to the CAQDAS-field. This is clearly open to debate, and they may not agree with my categorisation! As Figure 3 illustrates, there are no clear relationships between role and perception of ‘past moments’.
Figure 3. Visual representation of the sub-categories mentioned by podcast guests about ‘past moments’
Table 5. Categorised Codes / Definitions and # Coded Segments
| Category / Code | Definition |
| Technology | Category that organises types of technical development in the CAQDAS field mentioned by Podcast Guests in response to the question “if you had to pinpoint one moment in the past 30 years, that had changed the course of computer-assisted qualitative analysis, what would be?” |
| General Tech Developments | Sub-Category that organises broad developments that have impacted the practice of QDA and/or the features of CAQDAS-packages, which may or may not have been developed for those purposes. |
| More flexible data types | Changes in the availability of or flexibility in the functioning of different types of qualitative data. For example video data formats, use of multi-media and online data, etc. |
| Relational databases | The architecture of CAQDAS-packages as databases that contain multiple interconnected points that can be navigated |
| Graphical user interface | Move to use of more graphic user-interfaces within CAQDAS-packages |
| Internet | The rise of the internet |
| Automated Transcription | Developments in the accuracy and speed of automated transcription of qualitative materials |
| Developments in Gen-AI | The rise of and developments within Generative-AI capabilities |
| Specific CAQDAS capability | Sub-Category that organized mentions of specific tools and/or their use within particular CAQDAS-packages |
| Cross-Platform or Web-based | The ability of CAQDAS-packages to operate across operating systems, computer platforms and/or the rise of web-based CAQDAS applications |
| Mapping tools | Visual mapping, modeling or networking tools within CAQDAS-packages |
| REFI-QDA exchange standard | The common exchange format developed by a group of CAQDAS developers to enable analysed data to be moved from one program to another |
| Independent quotation architecture | The ability in some CAQDAS-packages to mark, work with, and analyse segments of qualitative data without the need to code them via a technical tagging process |
| Initial development | Reference to the earliest moments when CAQDAS packages began to be developed |
| Changed thinking/practice | Discussion about changes in one’s own thinking or practice with respect to QDA and the use of CAQDAS-packages |
| The QDA process | Sub-Category that organises aspects of the qualitative data analysis (QDA) process discussed in relation to ‚past moments‘ |
| Collaboration | Discussion about researchers ability to and practice of working in teams to undertake QDA, with or without the use of CAQDAS-packages or other technology |
| Rigorous process | Discussion about the quality of qualitative analysis in relation to the potential contribution of CAQDAS-packages |
| Opening up QDA | Discussion about the process of QDA becoming more open in terms of accessibility |
Table 6. Number of Podcast Guests mentioning technological developments in relation to their impact on the QDA process as significant moments in the course of the CAQDAS field
Here’s an overview of my process
- Checked and corrected transcripts of each of the 12 episodes
- Captured the discussion in response to the question about in each transcript at a code called ‘past moments.’
- Read each ‘past moments’ discussion in turn and summarised the response
- Reviewed my summaries and coded them to initial codes
- Retrieved coded segments together with my summaries and re-coded into more specific sub-codes – or gathered into broader categories – where necessary
- Explored the relationship between coded segments quantitatively (Tables 3 and 5), qualitatively (Tables 2 and 4), and visually (Figures 1, 2 and 3)
- Synthesised these explorations into this blogpost.
Would AI do a better job?
While I was preparing this blogpost it occurred to me that this could be an useful opportunity to do a little experiment: to see whether the use of Generative-AI would add anything or reveal anything I hadn’t picked up on. Watch this space for another post on that….
[i] Note that episodes 1 – 6 are only available to listen. From Episode 7 podcasts are available to listen and watch.
AI use: The podcast recordings were captioned for accessibility using automated transcription tools, and those transcripts used for this project. No AI was used for the analysis or writing of this post.
