After completing the coursework requirements for my Ph.D. program earlier this year, I began thinking of ways to organize my reading notes in a way that would allow me to easily recall them them when I begin writing my dissertation and would, more immediately, help me study for my qualitative exams. Being a (militant) PC user, I needed software that would either run on Linux or Windows. Consequently, the popular document management system, DEVONthink Pro was not an option. And while I’ve used Zotero, and (until recently) Mendeley for years, I was really looking for something that worked on a more fine-grained level.
After talking to some of my colleagues in the sociology department about their research methods, I decided to check out a few different qualitative data analysis (QDA) software suites to see if they would be up to the task. I tried a number software packages, including Atlas.ti, NVivo 10, and MAXQDA. On the whole, Nvivo was my favorite of the bunch, but I eventually settled on MAXQDA because it was much cheaper, and for it’s very useful “lexical search” feature (more on this later). In this post I’ll describe how I’ve used MAXQDA to organize my notes and highlights when preparing for qualifying exams.
Coding with MAXQDA
Like most other QDA packages, MAXQDA was designed with social scientific coding in mind, but I’ve been using it primarily as a method to tag and capture passages from books and articles. So far, I’ve been working with a body of literature that I’ve already annotated as PDFs, but it has proved an extremely useful way of organizing and systematizing the work I’ve already done.
After you’ve imported your documents into MAXQDA, the first thing you’ll want to do is set up a coding system. Since I’m not approaching this from a social scientific perspective and I’m not working from a standard codebook or using grounded theory, I set this up with two goals in mind: 1) to capture the main arguments and terms from a text for use in exam preparation and 2) to create a reference document on a variety of topics for future use in writing and teaching. To this end, I created two coding hierarchies: one called “Key” and the other called “Topics.” In Key, I code the the most important passages of a text, including:
- Any available abstract (coded as “Abstract”)
- Its main arguments (coded as “Major Arguments”)
- Its central questions (coded as “Questions”)
- Any theoretical and terminological framework it introduces (coded as “Terms”)
- How it engages with the literature (coded as “Gestures to the Literature”)
- Any particularly elegant encapsulation of historical change (coded as “Potted History”)
As you might be able to tell, things coded in the “Key” hierarchy are primarily useful telling you something about and individual text. The “Topics” hierarchy I use is much more “field” oriented. Although you can add codes to your system at any time, if you already know what the major debates and concepts are in a specific field, it’s a good idea to add these to your coding system early so you don’t have to circle back to them later. For example, right now I’m mainly looking at readings related to the history and sociology of biomedicine, and much of my topic hierarchy reflects this:
Lexical Search & “Auto-Gutting” the Literature
If you spend a little time and convert your articles to word documents (I did this with ABBYY FineReader) you can autocode sentences with MAXQDA’s very cool lexical search feature. (Note: you can use lexical search with PDFs, but the program doesn’t support automatic sentence tagging.)
Lexical search allows you to define a set of search strings and automatically (or selectively) code sentences that match those strings, a feature I’ve primarily used to quickly “gut” articles. I work primarily with three lexical search sets I’ve defined, which automatically code sentences for argumentative language, gestures to the literature (see above), and the introduction of new terms (which is very common in the STS literature). By doing this, I can very quickly get a sense of a piece’s major arguments, its position in the historiography, and the language it uses.
Exporting your Data
MAXQDA allows a number of options for exporting your codes and notes after you’ve finished coding your files. For my purposes, I found the program’s Smart Publisher report generator to be perfectly suited for the last stages of exam preparation. Since I organized my code system around topical hierarchies, the final report is a thematically indexed guide to the literature, with my notes and highlights grouped according the the codes I used. This output option is not only perfect for the last, consolidating stages of exam preparation, but I anticipate it will also be useful in writing the historiographical sections of my dissertation, literature reviews, and maybe even lectures.