Qualitative data is an interesting kind of raw material to work with. By itself it already tells a smaller complete narrative independent of analytical framework, but within a larger analysis it can effectively propel an argument and be enhanced in the company of similar data. Of course, the social scientist’s job is to appropriately and efficiently capture the essence of the data and interpret it for general consumption, and as expected this is no simple task.
The bulk of the problem lies in, unsurprisingly, the nature of qualitative data itself. Like all forms of data, it is inert, passive and neutral. However, qualitative data in particular also has the added difficulty of being extremely ambiguous both in its construction and its manipulation.
For example, let’s say you’re trying to understand the reasoning behind a particular individual’s decision to vote for the green party in… let’s say Lithuania. As is typical with such a study, you begin by looking at quantitative data – stuff like the general spread of voting behaviors in the person’s community, as well as other things like voting trends based on different aspects of demographics (age, race, religion, etc.) You eventually find that the findings you come up with, though informative, lacks depth. So you develop a method to extract qualitative data instead – you choose to conduct an interview with that person.
The two of you meet and talk; in the end, you produce hours and hours of recorded data. Later, you go back to analyze it and eventually you generate a report of your findings (Alternatively, you may choose to codify the data into set variables – making them ‘pseudo-quantitative’; though a different procedure altogether, it’s basically the same task. The researcher is transforming qualitative data into usable findings). To establish a level of reliability, you also have a colleague come in to analyze your recordings and produce an independent report. What you’ll generally find is that though the two reports may overlap in many respects, they will likely also differ in very important and fundamental levels.
This disparity lies in the essential fact that researchers entering the analysis also bring in her or his own sensibilities, philosophies when they take pieces of information and convert them into “facts.” There are many ways to deal with this, of course, and those methods are discussed in this section. Also explored will be various other considerations that should go into the task of collecting, analyzing, and interpreting qualitative data.
Note also that across history social scientists have also divided themselves into different schools of thought on how to handle qualitative data – and data in general – and that aspiring researchers are encouraged to survey the reasoning behind all these competing perspectives.
- Bennett, A., and A. L George. 2001. “Case studies and process tracing in history and political science: similar strokes for different foci.” In Bridges and Boundaries: Historians, Political Scientists, and the Study of International Relations, Cambridge, MA: MIT Press, p. 137–166.
- Lustick, Ian S. 1996. “History, Historiography, and Political Science: Multiple Historical Records and the Problem of Selection Bias.” The American Political Science Review 90(3): 605-618.
- Wolcott, Harry F. 1994. Transforming qualitative data. Beverly Hills, CA: SAGE.