Ali Kagalwala (Texas A&M University), Andrea Junqueira (Texas A&M University), Guy D. Whitten (Texas A&M University), Laron K. Williams (University of Missouri) and Cameron Wimpy (Arkansas State University)
Abstract: Models of time series cross sectional data have become the modal empirical tool for testing theories in many subfields of political science. Despite the prevalence of models with data along these lines, there is a lot of disagreement about how such models should be specified and estimated. Most papers that provide advice about how to deal with the time series aspects of such models largely ignore or treat the spatial aspects of them as nuisances. Similarly, most papers that provide advice about how to deal with the spatial aspects of such models largely ignore the temporal aspects. In this paper, we attempt to bridge the gap between these two areas of the literature by exploring the utility of different modeling strategies which treat both the temporal and spatial aspects of models of time series cross sectional data as phenomena of interest and offer practical advice for researchers about estimation and interpretation strategies.