Simultaneous Estimation of Ideological Positions Using Favorability Ratings

• In Event: Advances in Measures of Political Attitudes (Virtual Session)

Abstract

Spatial models of politics produce geometric representations of political variables. There are a host of such models available that utilize various types of data, from roll call data to political donation data, to produce spatial representations of ideology. Where these representations have been available, as the seminal work of Poole and Rosenthal in the United States, they have been enormously helpful in understanding the political dynamics. In many parts of the world, however, we do not have access to the types of data that these models demand. I propose a method for simultaneous estimation of citizens’ ideologies and politicians’ perceived ideologies, using favorability data. I use this method on data gathered from Iran, a country about which we have little data and little understanding regarding the relative ideological positions of politicians. We show how citizens’ ideological positions are related to a host of political attitudes. Because the estimation is simultaneous and on a shared scale between politicians and citizens, we can also measure the average distance from all citizens to each politician. This gives interesting insights: for example, we can see that there are unpopular politicians located in favorable ideological positions.

Author

• Ali Sanaei, Shahid Beheshti University