For the last several posts, I have been discussing the need for complexity science to truly overcome the qualitative/quantitative divide by doing more work to develop qualitative method. The next question, then, is how?
Of the various options available to complexity scientists, I think the best is case-based method. Actually, the better term is cross-case analysis. Cross-case analysis is an inductive approach to scientific inquiry that begins with a set of cases in order to explore what makes them similar to and yet different from one another. Cross-case analysis is very iterative and data-driven: the researcher develops ideas about the non-obvious patterns of relationship amongst a database by exploring its cases.
Perhaps the most well-known cross-case method is grounded theory, which was developed by Glaser and Strauss in the middle 1960s. While their method is referred to in the popular literature as grounded theory, they actually called it (at least initially) the constant comparative method, which they argued could be used to generate grounded theory. In other words, their famous book title, The Discovery of Grounded Theory was meant to imply that, through the constant comparative method one could generate grounded theory. Instead, the name Grounded Theory stuck.
In the sticking of this name, however, a major feat in the history of social science method was lost. In a paper I published in 2003, my colleagues and I made it clear that Glaser and Strauss never meant their method to be limited to narrative data. The constant comparative method could be equally applied to numerical or narrativel data. Grounded theory was not only a breakthrough in the popularization of cross-case analysis, it was a major breakthrough in the blurring of qualitative and quantitative method.
Here is a blurb from their book:
"Our position in this book is as follows: there is no fundamental clash between the purposes and capacities of qualitative and quantitative methods or data. What clash there is concerns the primacy of emphasis on verification or generation of theory—to which heated discussions on qualitative versus quantitative data have been linked historically. We believe that each form of data is useful for both verification and generation of theory, whatever the primacy of emphasis. Primacy depends only on the circumstances of research, on the interests and training of the researcher, and on the kinds of material he needs for his theory (1967:17–18)."
Grounded theory is not the only cross-case method. Others do exist. The problem, however, is these methods have not made it into the mainstream of sociological or social scientific inquiry.
What is fascinating to me is that, while case-based method remained on the margins of sociological inquiry throughout the 1980s and 1990s, over on the other side of the scientific fence, in the natural and computational sciences, cross-case method was being rediscovered. This time, however, it emerged in the form of distributed artificial intelligence, cluster analysis, data mining, decision-tree analysis, artificial neural networking, the self-organizing map algorithm, machine intelligence, genetic algorithms, fuzzy-set theory, fuzzy-set logic, and the host of robots and algorithms running our washing machines, cars, industrial machinery, traffic lights, the internet and, the soon to come, Web 2.0.
And still sociologists sit idle, believing case-based method is something wishy washy that qualitative type people do. Just like sociologists and many social scientists have sat idle and watched complexity science emerge.
We are out of the loop--big time! Trust me, I am not being dramatic. If you approached the average sociology professor or graduate student and asked them if they could implement any of the above methods I just listed from the natural and computational sciences, and could they do so while integrating these methods with qualitative methods to conduct qualitative, cross-case analysis of large, complex databases, they would probably say no.
Hence the need for David Byrne and Charles Ragin's forthcoming book, The SAGE Handbook of Case-Based Methods. Actually, the sub-title of the book should be qualitative, comparative analysis (QCA), because that is the method they have been advocating for several years.
It is great to see this book published. It is also great that it is a handbook, because that means other scholars are working with these ideas; and the fact that SAGE has published it means that QCA has, in some small way, gained the authority it deserves.
A quick review of the chapters in the book demonstrates the broad utility of cross-case analysis and, more specifically, QCA (click here to see the complete index). There are chapters integrating cluster analysis with case-based method, as well as chapters applying QCA to the analysis of large, complex, digital databases.
The book also goes a long way to integrating cross-case analysis with complexity science. Byrne and Ragin are major social science scholars in complexity science. In my book on Sociology and Complexity Science (SACS), for example, I identify them as two of the leading scholars in SACS--see my map of SACS. For example, Byrne wrote a very important book in 1998 titled, Complexity Theory and the Social Sciences. Ragin's related book is Fuzz-Set Social Science (2000).
For those interested in developing a method for studying complex social systems, Byrne and Ragin's book provides the necessary foundation. In the name of QCA, they bring together the best of qualitative and quantitative method in order to overcome both.