Spencer Piston

Why state constitutions differ in their treatment of same-sex marriage

Some states treat a same-sex marriage as legally equal to a marriage between a man and a woman. Other states prohibit legal recognition of same-sex marriages in their constitutions. In every state that has a constitutional restriction against same-sex marriage, the amendment was passed by a popular vote. The conventional wisdom about allowing voter participation in such decisions is that they yield constitutional outcomes that reflect attitude differences across states. We reexamine the attitude-amendment relationship and find it to be weaker than expected. In particular, we show that states vary in the costs they impose on constituencies that desire constitutional change. Some states impose very low costs (i.e., a simple majority of voters is sufficient for change). Other states impose very high costs (i.e., substantial legislative and voter supermajoriries are requires). We find that variations in the legal status of same-sex marriage across US states is better explained by these variations in costs than they are by differences in public opinion. Our method yields an improved explanation of why states differ in their constitutional treatment of same-sex marriage today. Our findings have distinct implications for people who wish to understand and/or change the future status of same-sex couples in state constitutions.

Accentuating the Negative: Candidate Race and Campaign Strategy

This article examines the impact of candidate race and campaign negativity on candidate evaluations and turnout. Unlike previous research, we argue that candidate race and campaign negativity should be considered simultaneously. In order to test this argument, we conduct a survey experiment of a nationally representative sample of White adults and a replication study. While we find, consistent with previous research, that respondents unfavorably evaluate candidates who decide to sponsor a negative ad, there are two important exceptions to this pattern: When the ad sponsor is Black, among White respondents who view Blacks negatively, the penalty for going negative is disproportionately large, while among White respondents who view Blacks positively, the penalty for going negative is disproportionately small. More generally, our findings suggest that the effects of candidate attributes and campaign strategy on voter behavior should not be considered in isolation, as they are mutually reinforcing.

Clear as Black and White: The Effects of Ambiguous Rhetoric Depend on Candidate Race

Campaign advisors and political scientists have long acknowledged the benefits of ambiguous position taking. We argue, however, that these benefits do not extend to black candidates facing nonblack voters. When a white candidate makes vague statements, many of these voters project their own policy positions onto the candidate, increasing support for the candidate. But they are less likely to extend black candidates the same courtesy. We test these claims with an original two-wave survey experiment varying the race of male candidates on a national sample of nonblack voters. We find that ambiguity boosts support for white male candidates but not for black male candidates. In fact, black male candidates who make ambiguous statements are actually punished for doing so by racially prejudiced voters. These results clarify limits on the utility of the electoral strategy of ambiguity and identify a key condition under which prejudice shapes voter behavior.