Putting Voters and Politicians on a Common Ideological Scale

In class we were talking about the Bafumi and Herron work on simultaneously estimating ideal points of voters and congressmembers. One of the challenges in interpreting this sort of work is that survey responses are different from vote choices.

Anyway, we were discussing all this, and Lucas mentioned this paper where they addressed some of those issues in a study in Switzerland:

Little is known about the ideological relationship between the Swiss political elite and the general public. Based on the SELECTS 2007 candidate and voter surveys, we compare the value orientations of both groups by applying ordinal factor analysis. First, we test whether political leaders or their supporters are more ideologically polarized. Second, we investigate whether ideological congruency between the electorate and representatives varies from party to party. Third, we examine whether winning candidates are ideologically more remote from their party supporters than unsuccessful candidates. We find that ideological polarization is larger within the political elite than within the general public. As a consequence, representatives of parties with rather extreme value orientations represent the moderate electorate rather poorly. Similarly, successful candidates are found to be more distant from their party supporters than unsuccessful candidates. These findings challenge traditional spatial voting theory but accord nicely with the directional model of voting behavior.

One difference between U.S. and Swiss politics is that we have a two-party system with close connections to political ideology, whereas their system is more multidimensional.

[Cross-posted at The Monkey Cage]

Andrew Gelman

Andrew Gelman is a professor of statistics and political science and director of the Applied Statistics Center at Columbia University.