Most Recent Posts

Monday, January 12, 2015

Everything You Think You Know About Comparative Politics (May Be) Wrong


Munck, G. and Snyder, R. (2007). “Debating the Direction of Comparative Politics: An Analysis of Leading Journals,” Comparative Political Studies 40(1): 5-31.

Munck and Snyder premise this article on the assumption that comparative politics, as a subfield, is too broad to be amenable to qualitative analysis. In this, they recapitulate the methodological argument within the subfield at a meta-level, as both Blyth and Mahoney note.  Arguing that most comparative politics choices, whether theoretical, methodological, or substantive, are appropriately framed as “both-and” choices rather than “either-or” choices, they present data from three leading subfield-specific journals and argue that this data disconfirms the conventional wisdom about methodological evolution in comparative politics. Instead, they argue that comparative politics still tends toward qualitative methods, inductive models, and focuses on the state. They conclude by urging a greater turn toward quantitative and formal approaches, to avoid the small-N and data-mining problems, and greater focus on sub-national actors.
This article draws its data from three leading comparative-specific journals, but because it limits its analysis to descriptive statistics, it lacks an identifiable dependent and independent variable and has no causal explanations. Using the descriptive statistics, Munck and Snyder analyze the subfield to determine if the worrisome criticisms of some scholars have merit, and conclude that their five recommendations for increased methodological sophistication are merited, and no other.
Unfortunately for this review, the most salient critique of Munck and Snyder was quickly snapped up by James Mahoney in the journal pages immediately following their article. By cherry-picking their journal choices, they’ve engaged in precisely the sort of data-mining they criticize; when Mahoney considered the most prestigious discipline-wide journals, he found that the turn from qualitative and inductive research was real. This argument – that the subfield’s most prestigious research is not being published in the subfield-specific journals that Munck and Snyder analyze – is tremendous, and scatters much of their argument.  Lake’s suggestion that any conglomeration of people constitutes an institution with behavioral consequences gives this argument additional force: since comparative politics is part of political science, the broader disciplinary expectations shape how comparativists think and train.