| Summary: | While political scientists have traditionally examined social influence through social network or contextual studies, this dissertation argues for the use of spatial econometrics as an alternative approach. While spatial econometrics is not new to political science, the dissertation attempts to broaden its application by exploring spaces based on geography, demographic characteristics, and ideology. Social influence can be understood as a form of spatial interdependence among individuals in these spaces and can be analyzed as spatial autocorrelation.
In the dissertation, I discuss the dimensions of the three spaces, what might account for mutual influence in these spaces, how to measure distances in these spaces, and how to use these distances for estimating social influence in models of political attitudes using ANES data. By taking a broader approach to space, I show that spatial econometrics can offer many advantages over more conventional approaches. === text
|
|---|
