Qualitative data. The phrase conjures up stacks of spiral-bound field notebooks, a dented-and-scratched voice recorder, and most of all, perpetual disorganization. While chemists have lab notebooks and accountants have spreadsheets, qualitative researchers are often left to invent a data management system entirely from scratch. Not one to turn down a challenge, in the summer of 2011 I tried find a workable and productive solution to managing a busy year of interview-driven research. Here’s what I did.
Interview data – like any James Cameron movie –is 90% pre- and post-production and only 10% on-site action. I needed a centralized and secure system for managing contact information for potential research participants, logging phone calls and emails, and storing research data such as audio files and consent forms.
There are many fragmented solutions out there that handle one – sometimes two – of the demands. But for the sake of data security, research ethics, and easy-of-use none of those options were acceptable.
Instead, I used Filemaker Pro to build a simple database that met the research demands. Filemaker Pro was a good fit because it has a low learning curve and can handle multiple file formats, including Word documents, PDFs, and audio files. As a result, I could track everything I needed for each research participant on one screen.
If you are not familiar with databases (I certainly wasn’t), a book like Filemaker Pro 11: The Missing Manual (most recent version is here) will help you start from scratch. I began by creating a database, which is like filing cabinet for everything associated with my project. Next, I created tables, which are like hanging file folders: they hold similar types of information in a standard format. In the screenshot above, you’ll see my table is for organizations. I created fields for each kind of information I needed from organizations – name, address, phone number, etc. – and created a layout. Finally, I created a new record for each organization as needed. Here’s what it looked like as I created the database:
This process is not strictly linear: on many occasions I had to add or delete fields, rearrange the layout, and add tables as my needs in the field changed. To be clear: databases are designed to be much more sophisticated than what I’ve described here. However, for a researcher trying to be diligent about managing qualitative data, even this simple system is the most comprehensive solution I’ve found so far.
This approach was incredibly useful before, during, and after the fieldwork.
Communication Log: At any moment I could pull up an organization or individual and see every contact I had made with them. This made follow-up calls more personal and effective.
Security: Since the database exists as a single file, it was easy to put it in a secure disk image file for maximum security (password backed-up offline, of course).
Transparency: It was easy to report to my advisor and research team. I could always demonstrate exactly what I had done, even when my time didn’t yield confirmed interviews.
Research (Meta)Data: By the end of the project, I learned a lot about the labor required to obtain and complete qualitative interviews, such as the average number of calls required to confirm an interview. This has pedagogical value, as well.
Ongoing Usefulness: Once constructed, the database can be re-used for future projects. Additionally, I’ve used what I learned about the research process to make a more realistic estimate of the number of interviews I can complete in the field.
The system wasn’t perfect. Here are some problems I ran into.
Knowledge of Databases: I was the only one on the team who knew how to run the software. I had to invest Summer vacation time to learn how databases work.
Database Corruptibility: Occasionally, the database had to auto-rebuild for an unknown reason. This caused some temporary panic on my part, but I back up religiously and always keep duplicate research data.
Geek Factor: I felt guarded skepticism from colleagues about using a database for qualitative research.
All told, the Filemaker Pro solution worked well and was worth the front-loaded investment of time and money. However, I can’t help but think that there is an overlooked market for reasonably priced, easy-to-use software that meets the demands of qualitative researchers. Surely I’m not the first person to run into this problem – right?
How about you? Do you have a system for creating and managing qualitative data?
Photo “Insect Distribution Maps in Robert E. Silberglied’s Field Notes” by Flickr user Smithsonian Institution / No known copyright restrictions (Part of the very cool Field Book Project–@JBJ)