This post is also available in: Português (Portuguese (Brazil))
I started doing research when cut and paste literally meant using scissors and glue. Coding (or categorizing) involved transcribing interviews that were recorded with a cassette tape. Then, by hand, we underlined the important themes with colored markers, using different colors for different categories. At a certain point, we cut the highlighted categories, and glued those with similar colors side-by-side into new blank sheets. Then, we compared the pieces to check if our categories made sense and were robust enough.
Yes, I’m old. And luckily, qualitative data analysis counts on many new and handy resources these days. Think of programs like Atlas.ti, NVivo, MAXQDA, and so many others. Using qualitative data analysis software, however, can still seem daunting for many qualitative researchers. It may feel complicated, mechanic, and not adequate to bring up the best out of the in-depth and nuanced material we collect in qualitative interviews. However, going back to the old paper cut and paste or building complex excel tables is not necessarily better or more efficient. On the contrary.
I first used a qualitative data analysis software in 2010, when I started analyzing data for my Ph.D. I used Atlas.ti.
These were my first impressions, as a novice in the tech field:
- More control. The software gives you more control over your data and makes the organization of data much easier than the old paper “cut and paste” style.
- Easier to find and compare data. You have all the data in one single file. The program's searching tools make it easy to find what you need and to make comparisons between categories.
- Interpretation "memory". You can record and trace-back your analytical decisions by logging them into the program (memos) while coding.
- More reliability. This control helps you to be more reliable as a researcher, to learn from your successes and mistakes, and to be clear and transparent on analytical decisions in group work.
- You are still the boss. The interpretative work is done by you: it depends on your brain and interpretation skills; the software facilitates your life but does not interpret things for you.
- It takes time. If it is your first time using a qualitative data analysis software, be aware that it takes time. You will learn while doing it, but it is better to kick off the learning process before you need to do the actual analysis and and are under time pressure to finish the project. It takes time to learn the program's features and to understand how to best benefit from it.
- Too much computer! It is a matter of personal preference, but sometimes I find it useful to work on paper sometimes. You can easily do that by printing lists of categories, quotations, or analytical interpretation memos, and reading or making comparisons outside the computer.
- You may get trapped in the technical bits. If you are a nerd like me, you might risk getting stuck in trying to learn ALL about the program. This can be a life-long task. Have in mind that some program features might not be useful for you, so better not to bother with learning them.
- It is costly. Most researchers rely on the software their institution supports, so they do not need to pay for it. It might happen, however, that you switch institutions and the new one supports a different software. You might have to learn how to handle a new program and might lose access to the data and analysis you had on the previous software. Nowadays, some programs (like NVivo) support transferring databases across different programs; hopefully, this tendency will increase in the future.
All-in-all, considering all the pros and cons, I found the balance to be positive. Up to date, I keep using qualitative data analysis software and experimenting with different programs. To me, the most important is to find a software that works for us as researchers, respecting our style and preferences, as well as the different research projects we have. There is no silver bullet. Some researchers may use the software solely for the process of coding, or as an organizing database for a literature review. Others may use it during the whole research process and to collect all materials. Programs usually support working with a variety of files: text, PDF, image, video, audio, geodata, and social media, which gives you a lot of flexibility. Virtually all software supports group work, which comes in very handy for research projects with partners in different countries.
Whatever you chose, remember that software is a tool developed to help with data analysis. It may cost you more time in the beginning, but in the medium term, it should bring an added value. If that is not the case, it may be time to use a different program or method.
Want more tips? Feel free to comment or contact me. Soon, I'll write a post on how to use Atlas.ti to do discourse analysis.