Content area
Full Text
Fuzzy-Set Social Science, by Charles C. Ragin. Chicago: University of Chicago Press, 2000. 352 pp. $48.00 cloth. ISBN: 0-- 226-70276-6. $20.00 paper. ISBN: 0-226-- 70277-4.
This is an ambitious book; the author wants to cross the chasm between purely qualitative research and big-N quantitative studies. Ragin claims that this can be accomplished through studies based on an intermediate number of cases (an example might be a study of 20 countries) if fuzzy-set analysis is used. Moreover, he sees this as a way of dealing with what he regards as the present mismatch between theory and research that often occurs under conventional forms of analysis. Appropriately, the early chapters emphasize what fuzzy-set analysis is all about, develop other concepts that he feels a new methodology will need, and criticize conventional forms of analysis. What is there not to like? It would be a shame for us to have different theories about a subject, determined solely on what methods are used. Ultimately, we are all losers if there are different schools of thought about a subject that are based exclusively on what subsets view as the only acceptable sources and ways of gathering evidence. Improvements in any one part of the discipline are ultimately of benefit to all of us.
Although these goals are admirable, I find the product to be rather disappointing. For one, there are virtually no illustrations that actually compare how the new fuzzy-set methods yield something better than conventional procedures. There are plenty of illustrations, but almost all fall short, one way or the other. For example, we are given a critical review of how the conventional procedures deal with a specific research problem, but when Ragin talks about his proposed alternative, he uses a different research problem, so there is no comparison between the product obtained when the two approaches deal with the same type of problem. Or he will have a fine time criticizing conventional procedures (sometimes unjustly; other times with skill and insight), but then these general criticisms are taken to be evidence that his proposed approach is better. Indeed, it is as if this difference is so obvious, that there is no need to compare the pudding that each produces. The implied reasoning is: If I can show that...