Published online by Cambridge University Press: 04 January 2017
This paper extends models of micro- and macropartisanship in two ways. It first develops a model of individual partisanship that accommodates changes in partisan utilities. Achen's Bayesian partisan updating is a special case of our model. This more general micromodel is then aggregated to create a model of macropartisanship. This macromodel is a more general version of the models of macropartisanship estimated by various authors. The less restricted version incorporates possible individual and temporal heterogeneity. We present an example using real data that offers a possible way to estimate the parameters in the full aggregate model.
Authors' note: This paper continues a project that began with a paper on partisanship and path dependence presented at the 2007 Meeting of the Society for Political Methodology, State College, PA, July 2007. We express our thanks to Chris Achen, Neal Beck, and John Freeman for their comments on that paper and for pushing us to explore this connection. Jude Hays offered suggestions that improved the model. Lastly, we want to thank the editors and reviewers for their requests and suggestions, which led us to do more that we thought possible. All errors are, of course, our own.