The chapter wraps up by placing the ecological inference problem throughout the context of the literature on partial identification and by describing latest work generalizing using logical bounds to provide solutions which are “regions” as an alternative of point estimates for parameters. Often our research requires that we use more specifically defined methods to reply our analysis questions. In our operating example, we’ve up to now ignored the fact that our dependent variable is sometimes a dichotomous variable (as in Table 1.3 above), but there are good causes to imagine that we should take this into consideration. Discrete (p. 22)selection modeling (Chapter 22) by Glasgow and Alvarez presents methods for coping with dichotomous variables and with ordered and unordered selections.
The Department of Political Science on the University of Florida presents a Ph.D. subject in political methodology. Political methodology, broadly outlined, addresses the tools of inquiry which are applicable to … Read More