LAUSR

Measuring inequalities in political participation across social groups is a challenging task as participation is typically coded in dummy variables. For instance, social scientists record whether their respondents have voted in the previous elections (1) or not (0). In this paper, we identify a list of desirable criteria that an inequality index used for empirical comparative studies should meet. Existing inequality indices fail to satisfy one or more of these criteria. Building on our list, we define a new Gini-type index, the Political Inequality in Participation Index (PIPI), suitable for cross-country comparisons. We show that inequalities measured by the PIPI are correlated to, but are qualitatively different from the best-known measurements. In particular, using data simulation techniques, we demonstrate that this correlation is decreasing in the complexity of societies’ structure. Moreover, by replicating an existing study, we further demonstrate that when working with real-world data, the PIPI provides new empirical results.