Seven Rules for Social Research

This is an unusual and untypical book on social research methods, in which Glenn Firebaugh manages to buck the unwelcome trend within methods texts of creating "How to ..." guides, seasoned liberally with under-theorised outputs from SPSS (Statistical Package for the Social Sciences) tables.

Firebaugh maintains a refreshingly reassuring conversational style with the reader and communicates difficult concepts in a straightforward manner. In doing so, he occasionally veers from the obvious - "A research question in the social sciences is a question about the social world that you try to answer through the analysis of empirical data" - to the relatively inaccessible: "To accommodate additional groups, equation 6.8" (a complex, multiple-model, regression equation) "adds dummy variables (G3, etc.) to capture differences in intercepts, and interaction terms (TxG3, etc.)."

Firebaugh states that the book is aimed at students taking an "upper-level undergraduate methods course", but how many will have covered the mathematics of regression at the first level? Consequently, this is a book to be studied, rather than just read.

The book's title - Seven Rules for Social Research - is also uncommonly to the point; but Firebaugh strikes one as a pretty no-nonsense kind of guy.

Consider these examples, taken from a section in which he is bemoaning what he calls "empirical nihilism", which, he asserts, is "most prominent in schools of thought containing a 'post' prefix: postmodernism, postpositivism, poststructuralism, post-Kuhnian philosophy of science, and so on. In its most extreme version, empirical nihilism in the social sciences denies the possibility of discovering even regularities in human behaviour. That position is obviously silly." (Is it, indeed?) He then adds this advice: "The best response to empirical nihilism is to ignore it and do the research anyway."

My advice is to ignore such unbridled proselytising, and then to engage fully with the text. Its unapologetically and unreflectively positivist take on social research is regrettable; only three pages are devoted to qualitative research, and even then it is presented as the servant to a quantitative overseer - "Qualitative methods are well-suited for providing thick description that can help place quantitative results in proper context."

The strength of Seven Rules for Social Research is its concern with the "logic" of doing good social research; it is not really about "methodology" (thankfully, I suspect).

With this caveat in mind, Firebaugh has produced a lively and insightful contribution. He discusses thought-provoking examples and has created some truly excellent and innovative end-of-chapter exercises, whereby the reader's hand is held while being guided through data sets taken from the author's Princeton website.

He manages to expose the many traps an unsuspecting researcher will encounter when planning and doing research (and not just in the early days of a career).

One of Firebaugh's many "yes, of course, why didn't I say that?" examples is: "Some methodologists and philosophers of social science conceptualise cause as deterministic rather than probabilistic, so a purported cause can be eliminated by a single exception. If so, small samples could be analytically powerful, since each case has the potential to eliminate a different explanation ... small N's can lead to 'big conclusions'."

By the end, his seven rules make a lot of sense:

i) Surprise should be possible

ii) Look for differences that make a difference

iii) Build in reality checks

iv) Replicate

v) Compare like with like

vi) Study change

vii) Let method be the servant.

They are eminently reasonable points and offer sound advice to social scientists. I shall be recommending Seven Rules for Social Research to students, but only after giving them a health warning about the author's mostly undeclared epistemological standpoint.