Please note: This course will be taught in hybrid mode. Hybrid delivery of courses will include synchronous live sessions during which on campus and online students will be taught simultaneously.
Martin C. Steinwand is a Senior Lecturer in the Department of Government at the University of Essex. His research interests include spatial econometrics, especially as a tool to capture strategic decision making. Substantively he is interested in foreign aid, international institutions, coalition dynamics, and domestic sources of foreign policies. His works have appeared in journals such as International Organization, Journal of Conflict Resolution, Review of International Organizations, and other peer-reviewed journals.
Spatial dependencies are a universal feature in the social sciences. Phenomena as diverse as the occurrence and outcomes of violent mass protests, policy learning and position taking in party competition, or the competitive setting of tax rates to attract foreign direct investment across neighboring jurisdictions, all share a similar feature: actions taken by one actor are shaped in a theoretically meaningful way by the actions of one or more other actors. Spatial econometrics allows us to detect, model and estimate such interdependencies, and to work towards a causal interpretation of such relationships. The theoretical substance lies in the nature of interconnectedness between units, which can be geographic, economic, cultural, strategic etc., thus covering a wide ground of social science applications. This course begins with a data-oriented view of spatial patterns and dependencies in the data, then introduces a theory guided approach to building, estimating, and evaluating spatial and spatiotemporal regression models, and ends with a critical evaluation of the spatial approaches in the context of causal analysis.
The course starts from the premise that interconnectedness is an important and theoretically meaningful feature of a broad range of phenomena in the social sciences. The main aim is therefore to enable students to identify and incorporate interconnected features in the study of their own data and areas of interest. This will involve learning how to detect spatial patterns, bringing data into a suitable format for spatial analysis, the estimation of structural parameters of spatial and spatiotemporal models and the presentation of effects. The materials provided in the labs will enable the students to undertake their own applied spatial project.
All necessary background materials will be covered (in brief), though students will benefit most from the course if they have some understanding of regression analysis and a basic knowledge of matrix algebra and maximum likelihood, as well some familiarity with either R or Stata.
Representative Background Reading
Ward, Michael D and Kristian S Gleditsch. 2018. Spatial Regression Models: Second Edition. Quantitative Applications in the Social Sciences 155. Sage. – This book will be provided by ESS.
Le Sage, James and R. Kelley Pace. 2009. Introduction to Spatial Econometrics. CRC Press.
Franzese, Robert J and Jude C. Hays. 2008. “Empirical Models of Spatial Interdependence”. In: Oxford Handbook of Political Metholodogy. Eds. Janet Box-Steffensmeier, Henry Brady, and David Collier. Oxford.
Beck, N., Gleditsch, K.S. and Beardsley, K., 2006. “Space is more than geography: Using spatial econometrics in the study of political economy.” International studies quarterly, 50(1), pp.27-44.
Zhukov, Yuri M. and Brandon M Stewart. 2013. “Choosing your neighbors: Networks of diffusion in international relations.” International Studies Quarterly, 57(2), pp.271-287.
Gibbons, Stephen and Henery G. Overman. 2012. “Mostly pointless spatial econometrics?” Journal of regional Science, 52(2), pp.172-191.
Background Knowledge Required
Linear regression = moderate
Matrix algebra = elementary
Maximum likelihood = elementary
OLS = moderate
Stata = elementary OR
R = elementary