Measuring party system responsiveness in contexts where voters have weakly structured belief systems is a challenge. Here is the result of some thought I have put into this over the past years, included in a symposium in European Political Science
Contexts outside the advanced developed democracies present a challenge to assessing how well party systems reflect voter preferences across over-arching policy dimensions because not all electorates readily interpret political conflict in dimensional terms. In this contribution, I advocate an approach suited for such contexts that combines deductive and inductive elements: It starts out with what observers consider the most important dividing lines in a party system, and then goes on to operationalize these dimensions in an inductive fashion by drawing on all theoretically relevant items that are available in mass and elite surveys. I devise a relative-fit measure of responsiveness that can be compared across space and time, even if positions at the elite and mass levels are measured on different scales. To illustrate the usefulness of the strategy, I show how it leads to novel contrasts in terms of programmatic responsiveness among four Latin American countries, namely, Chile, Brazil, Venezuela, and Bolivia.