Abstract
In recent years, many scholars have argued that qualitative and quantitative methods are neither opposites nor enemies; instead, both should be used judiciously. Howard Becker’s new book belongs to this genre, and he claims that the two are not so different. Qualitative studies often use numbers. And quantitative studies are often executed with the meticulous care characteristic of qualitative studies, when it comes to the evidence.
Becker presents his argument in terms of three closely linked concepts: data–evidence–ideas. In brief, you need high-quality data that can support the idea, and this constitutes evidence. It is incumbent on the qualitative or quantitative researcher to ensure that the data are solid. It is, however, easy to make mistakes, and Becker’s book contains many interesting examples of this.
There is also much one can learn from another part of Becker’s book. This is where he discusses the different ways in which data collection can be organized, and what consequences this may have for the evidence. It is always hazardous, for example, to hire others to do the data collection. The reason, consistent with agency theory, is that “hired hands” have other goals than the principal investigator. It is also dangerous to rely on data collected by people whose salary or job performance depends on how many people or incidents they record.
Becker is best known as a qualitative researcher, and it is therefore of special interest to see how he presents this approach. In his view, the sociologist should enter the field without preconceived ideas and specific hypotheses to test. As you make your observations, you often have to change directions, in order to get a good handle on your topic. There will be plenty of surprises, and you need to learn how to zig-zag rather than just march ahead.
This way is contrasted with quantitative sociology, which, according to Becker, lacks the flexibility of the qualitative approach. Before gathering data, you have to decide what to look for and stick to that. A questionnaire, for example, cannot be changed mid-course. This means that unless you know what you are looking for before you start collecting data, you will soon be lost.
Becker’s argument is well presented and convincing. The reader is in for many fine sociological observations, drawn both from Becker’s own research and that of others. The book also contains a number of interesting observations on some key players from the golden days of Chicago Sociology (Herbert Blumer and Everett Hughes).
In two respects, however, my perspective differs a bit from Becker’s. The first has to do with his belief that the distinction between qualitative and quantitative sociology can be partly erased, and perhaps eliminated. Two factors, however, may prevent this. The first involves the coming of Big Data, which Becker does not discuss. There is a good chance that artificial intelligence, data science, etc. will put even more of a premium on quantitative methods.
The second factor is the aggressive attitude some quantitative sociologists display towards qualitative sociology, which in their view stands for an unscientific approach that should be eliminated from sociology. A recent example is the split that has developed in German sociology, whereby a number of quantitative sociologists in 2017 left the national association to found their own.
Now for the second point on which my perspective differs a bit from Becker’s. It has to do with the fact that the battle between qualitative and quantitative sociologists is essentially about methods. Which method should be used—that’s the question. Should it be one that takes the meaning in human behavior into account, or one that is based on statistical analysis? Note, however, that regardless of the answer, both approaches share one thing. This is that neither defines itself in terms of the theory that should be used in a sociological analysis.
Becker does not want to use the term “theory”; he prefers the more folksy “ideas” (p. 4). A good sociologist, he says, enters the field without preconceived ideas, carefully studies the empirical situation, and then backs up the ideas he or she has developed during the research with solid and relevant data (evidence).
One may, however, ask if this is how Becker himself handled the theory part in his many important studies, such as Boys in White (1961), Outsiders (1963), Art Worlds (1982) etc. This is more dubious. Taking Boys in White as an example, Becker says that he (and his co-worker Blanche Geer) did not bring any pre-conceived ideas (read “theories”) to the research. “When Blanche Geer and I began our study of medical students,” he writes, “we had no research hypotheses to test beyond the simple idea that students entered the [medical] school, spent four years, then graduated – and something must have happened to them during that time” (p. 182).
After about a year, however, Becker continues, he had developed a key concept: “student culture.” How did this happen—where did this concept come from; was it from data or theory? Becker’s answer is that for about a year he followed students around, talking to them and observing them, and this resulted in, he says, “what I’d begun to call ‘student culture’, a collection of shared ideas and activities students used to organize their response to the problems medical school created for them” (p. 183; emphasis added). The next sentence reads—and note how Becker explicitly refers to a theory that he already knew: “This [concept of student culture] mirrored a classical understanding of culture I had acquired in graduate school, William Graham Sumner’s description of folkways (1906) as solutions, collectively arrived at, to persistent problems that a group confronted.”
In brief, Becker learned about the concept “culture” from Sumner, gave it a nice creative twist, based on observation, and voila: he had developed the concept student culture. Those familiar with the works of Becker and his mentor Hughes can easily think of other similar examples. From your knowledge of sociological theory, you select a concept which you then change a bit, in order to express your new insight. Here are some other examples of this way of developing theory: “moral entrepreneur” (Becker), “status contradiction” (Hughes), “ethnic division of labor” (Hughes), and “dirty work” (Hughes).
Where does this leave us? My sense is that while ending the battle between quantitative and qualitative sociology is important, more needs to be fixed in contemporary sociology. Making this argument in relation to the book reviewed here, one can put it as follows: good sociology does not only depend on data–evidence–ideas. You also need one more item, and that is the crucial one in any science: theory.