GOP Vote. Brad Jones 1. August 7, University of California, Davis. Bradford S. Jones, UC-Davis, Dept. of Political Science

Transcription

1 Bradford S., UC-Davis, Dept. of Political Science GOP Vote Brad 1 1 Department of Political Science University of California, Davis August 7, 2009

2 Bradford S., UC-Davis, Dept. of Political Science

3 Bradford S., UC-Davis, Dept. of Political Science Vote Confirmation vote on August 6, 2009 for Judge Sonia. Vote: All Democrats voted for confirmation (Kennedy was not present) 9 of 40 Republicans voted for confirmation

4 Bradford S., UC-Davis, Dept. of Political Science Vote: GOP I treat the probability of support as a function of three variables: % Latino population (logged), in Latino population, and McCain s vote margin in the 2008 Presidential Election. t-tests for log of population and McCain margin: t-test for log(population): X for Supporters:.78 X for Opposers: 1.65 p=.02 t-test for McCain Margin: X for Supporters: X for Opposers: -3 p=.0001 Correlation between population variables is -.39.

5 Bradford S., UC-Davis, Dept. of Political Science Vote: GOP In general, those GOP in opposition come from states with a significantly larger Latino population. Similarly, opposition Senators tend to come from states where McCain did very well in the 2008 presidential election. No clear pattern is found for population change among supporters vs. opposers. Simple logit model treating support/opposition as a function of these three variables:

6 Bradford S., UC-Davis, Dept. of Political Science Support for : GOP Senators Estimate Std. Error Pr(> z ) (Intercept) log(population) (Population) McCain Margin

7 Bradford S., UC-Davis, Dept. of Political Science Vote: GOP Size of Latino population negatively related to log-odds of support. Change in population from 1990 to 2000 negatively related to log-odds of support, though p-value is large. (I rescaled this change score by dividing through by 100; therefore a 2 means 200 percent change). Size of McCain 2008 vote margin negatively related to log-odds of support. States evincing high levels of support for McCain but also having large Latino population are associated with Senators voting no on the confirmation. Predicted probability intervals for each covariate:

8 Bradford S., UC-Davis, Dept. of Political Science

9 Bradford S., UC-Davis, Dept. of Political Science Questions? To what extent will GOP rhetoric against anger and/or mobilize Latino voters? The question seems relevant given that states where no votes were prevalent are also states with large Latino populations. Of course, the extent to which these GOP Senators rely on the Latino vote is an issue. Put bluntly, does it/will it matter that these Senators who represent states with a large Latino population voted against a Latina judge for the Supreme Court? Put in perspective, this outcome is another example where GOP strategy runs squarely opposed to the preferences of many (not all) Latinos. Rhetoric on immigration in the middle/late 2000s is another example (IMHO).

10 Bradford S., UC-Davis, Dept. of Political Science Questions? Further, to view Latinos as a monolithic bloc is incorrect. There is substantial variability in partisan affiliation (though Democratic identification is most prevalent). A natural question to ask is how (or if) GOP strategy will push identification toward the Democratic side (I m working on this issue currently, as are several other researchers)? Given expansion of this population (by 2050, Latinos will comprise 25 percent of the US population), short term victory within the GOP may yield long-term difficulties. By victory I mean successful appeals to GOP base voters, a base that does not seem as prevalent as it did in the middle 2000s.

11 Bradford S., UC-Davis, Dept. of Political Science Sources Vote data and McCain margin data come from New York Times (8/6/09). Population data comes from U.S. Census Bureau. NB: analysis merely descriptive... n very small and is not a sample obviously.