In the latest issue of The Political Methodologist, James S. Krueger and Michael S. Lewis-Beck examine the current standing of the time-honored but oft-dismissed-as-passe ordinary least squares regression model in political science research. Based on their analysis of 1,756 articles in the American Political Science Review, the American Journal of Political Science, and the Journal of Politics, 1990-2005, Krueger and Lewis-Beck report that:
* The OLS regression model accounted for 31% of the statistical methods employed in these articles. (A total of 2,221 methods were counted, because some articles featured more than one statistical method.)
* “Less sophisticated” statistical methods — those that would ordinarily be covered before OLS in a methods course — accounted for 21% of the entries.
* “More sophsiticated” methods, e.g., time series, maximum-likelihood estimation, or latent variable analysis (topics ordinarily covered after the OLS model) accounted for 43% of the entries. (The totals don’t add to 100% because 6% of the methods weren’t identified in the source articles.)
* Just one in six or so of the articles that reported an OLS-based analysis went on to report a “more sophisticated” one as well.
* Over time, even within this relatively narrow time slice, “more sophisticated” statistical methods gained a bit of traction. (This conclusion is based exclusively on articles that reported both an OLS analysis and some other type of statistical analysis. I would have thought that the more telling basis of over-time comparison would be the relative proportions of articles per year that reported just one or the other type of analysis — those that relied on OLS, or on some “less sophisticated” method, or on some “more sophisticated” one.)
The authors’ conclusion? Good news for those whose statistical prowess pretty well ends where “more sophisticated” methods begin.
bq. OLS is not dead. On the contrary, it remains the principal multivariate technique in use by researchers publishing in our best journals. Scholars should not despair that possession of quantitative skills at an OLS level (or less) bars them from publication in these top outlets.
For the full report by Krueger and Lewis-Beck, click here.