Comment on Zigerell (2015): Using Poisson inverse Gaussian regression on citation data

Abstract

Zigerell’s recent article in Research & Politics argues that Maliniak et al.’s findings that women are cited less than men in international relations are not robust to alternative specifications. Using Poisson inverse Gaussian (PIG) regression, in this comment, I demonstrate that both papers’ findings are sensitive to empirical specifications. In many model specifications, women are cited less than men, but other specifications either show a null effect or that women are actually cited more than men in international relations. This illustrates that substantive model selections matter a great deal for the conclusions we can make from our data.