Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
The marginal abstention hypothesis is that once at the polls, individuals are more likely to vote in the closer races on the ballot and to abstain from voting in the safer races. This hypothesis is a straightforward extension of the rational voting model. In previous empirical applications of the rational voter theory, voter turnout and voting in a given election are treated as one in the same. These studies have produced mixed results. When applied to the problem of explaining marginal voting behavior, the theory works well. Data on a recent sample of U.S. House and Senate elections is used to test the marginal abstention hypothesis. In the average Congressional District, about three percent more people vote in the Senate race than in the House race. This difference varies in a fashion predictable by the theory. Copyright Martinus Nijhoff Publishers 1987