Measuring Ethnic Voting: Do Proportional Electoral Laws Politicize Ethnicity?


John D. Huber is Professor of Political Science, Columbia University, New York, NY 10027 ( An earlier version of the article was presented at “Democracy and Representation: Conference honoring the contributions of G. Bingham Powell and his 40 years of teaching and research in the Department of Political Science at the University of Rochester” 2010, Rochester, New York. I am grateful for research assistance from Tom Ogorzalek, for research support from the National Science Foundation, and for helpful comments from Kate Baldwin, Johanna Birnir, Daniel Bochsler, Dawn Brancati, Bernard Groffman, Simon Hug, and seminar participants at Columbia, Essex, LSE, University of Geneva, ETH Zurich, and the Hebrew University in Jerusalem. I am also grateful to the Russell Sage Foundation, where I was a visiting research scholar while writing an early draft of this article. All data and replication materials can be found at


I develop four related measures of the “ethnicization” of electoral behavior. Each measure increases as ethnic identity becomes more central to vote choice, but the measures differ along two theoretical dimensions. The first dimension contrasts a group-based perspective (which focuses on cohesion in the voting patterns of group members) with a party-based perspective (which focuses on the composition of groups supporting political parties). The second dimension contrasts a fractionalization perspective (which assumes that more groups or parties cause more problems) with a polarization perspective (which assumes that problems are greatest when there are two equal-sized groups or parties). Using survey data to implement the measures in 43 countries, the article shows that proportional electoral laws are associated with lower levels of ethnicization—the opposite of what is widely assumed. I argue that the lower levels of ethnicization in PR systems should be unsurprising.