Search

In a recent APSR article, Hajnal (2009) explores the question of who loses in American democracy, and finds that African Americans are more likely than whites to vote for a losing candidate. This finding is extremely troublesome, as it suggests that African Americans' voices are systematically being shut out of the political process. We revisit this question and show that Hajnal's analyses are too broad in their scope, and that they do not account for the sampling procedures unique to each national exit poll. When we disaggregate his results by year, weight the data according to each exit poll's probability sampling weight, cluster standard errors to take account of the first-stage precinct-wide sampling frame, and examine a longer time period, we arrive at different conclusions. We find a surprising number of elections where minority voters are more likely than whites to vote for a winning candidate. However, we also find many elections where minority voters lose, and we find that Hajnal overestimates the extent to which Latinos and Asian Americans win in some U.S. elections. The picture of "who loses" in American democracy may not be as bleak as Hajnal suggests for African Americans, but is somewhat less positive than he suggests for Latinos and Asian Americans. [ABSTRACT FROM AUTHOR]

Can inconsistency in legislators' voting records be explained by the uncertainty they have about their constituency? Why does a two-dimensional model of roll-call voting explain the votes of some legislators better than others? In this paper, we present an uncertainty-based theory of legislative voting and inconsistency. We test this theory by examining whether U.S. senators' uncertainty about their constituency leads to higher levels of inconsistent voting in Congress. Specifically we compare southern senators who faced a large influx of black voters in the 1960s and 1970s with southern senators from more stable constituencies. We also examine other factors that can cause legislator uncertainty, such as legislators' backgrounds. Our results suggest that what some call "maverick" behavior by legislators can be better characterized as best responses by incumbents to a changing environment. ..PAT.-Unpublished Manuscript [ABSTRACT FROM AUTHOR]

This paper answers the following question: Under what conditions are states more likely to repeal disfranchisement provisions pertaining to the voting rights of persons convicted of a felony? We also address two broader questions: (1) Do parties seek electoral gain by changing the electoral system and voting laws? And (2) do parties override the ideological preferences of constituents in order to advance their electoral agenda? The answer to both questions is ‘yes’. We find that disfranchisement provisions are more likely to be repealed under a unified Democratic state government, but that states where electoral support for Democrats is generally weak are also more likely to repeal such provisions. We find no evidence that citizen ideology effects changes in voting rights. The implications of our analysis for students of parties and voting rights are clear: Rules are changed by those who stand to benefit, and parties matter at the state level. [ABSTRACT FROM AUTHOR]

Some of the scholarly conventional wisdom regarding voter turnout and participation, especially in the U.S. South, has been that race and ethnic background can be important predictors of turnout. However, the disproportional effect of felon and ex-felon disfranchisement laws on racial/ethnic minorities in some states suggests that the effects of race on aggregate voter turnout will be underestimated unless these laws are also considered. This electoral institution-the disfranchisement of felons and ex-felons-is an important omitted variable in models of state-level aggregate voter turnout, which might result in biased inferences pertaining to minority citizens' level of participation, especially in recent elections. Looking at statewide elections in the U.S. South from 1984-2000, we hypothesize that this electoral institution helps explain lower levels of voter turnout in states with high rates of felon disfranchisement. We also demonstrate, once felon disfranchisement is controlled for, that southern states with high minority populations will have higher levels of turnout in recent elections where minority voter mobilization has been on the rise. [ABSTRACT FROM AUTHOR]