In the prior chapter, I developed a theory about how changes in the heterogeneity of democratic societies should shape party system fragmentation. To empirically assess this theory, measures of the independent variable of social heterogeneity—and hence testable hypotheses—are needed. In other words, which democracies are more socially heterogeneous than others? And what are the ways in which a country's citizenry can change with the passing of time, increasing its social heterogeneity? Answering these questions is my goal in the present chapter.

Yet measuring social heterogeneity is a difficult task. As I argued in the last chapter, there are an infinite number of ways in which the citizens of a polity can be divided into latent groups. For example, the types of attributes that may be used to define groups include the quality and pitch of a person's voice; the speed and accuracy with which a person types; and a person's passion for scuba diving. And even given a particular type of attribute such as vocal pitch, there is no end to the ways in which people can be sorted into groups for the purposes of assessing heterogeneity, such as sopranos versus all others, on the one hand, and sopranos, altos, tenors, and basses, on the other.