Voting in Four Elective Offices

Abstract

CPS survey data from 1952 to 1978 are employed in a comparative analysis of the effects of three predictors on voting for House, Senate, gubernatorial, and presidential candidates. Among the findings are: party identification is equally important in voting for all four offices, and despite a declining effect it is still the most important predictor of voting; presidential coattails especially serve as a vote guide for the less informed, and are more important in federal than state elections; incumbency is most important to the moderately aware voter, and only since 1966 has it been most important in House elections.

Explaining the Vote in Judicial Elections: the 1984 Ohio Supreme Court Elections

“…But even those who do vote, taken as a group, demonstrate only limited knowledge of the candidates (Judicature 1955;Johnson, Schaefer, and McKnight 1978). As studies of non-judicial elections suggest, the ways that voters make their decisions are likely to be affected a good deal by the limited information that judicial contests provide (Hinckley, Hofstetter, and Kessel 1974;Shaffer 1982).…”

Governance and U.S. Senate Elections: A Comparison of Presidenttal and Midterm Elections

“…Relevant only for presidential election years, a positive sign would signify the existence of presidential coattails. Instead of reflecting a strategic decision to promote a particular type of public policy, a presidential coattail vote often signifies a simple vote cue for constituents lacking information about nonpresidential candidates (Shaffer 1982).…”

“…Hence, increases in the number of Congress members from the president's party lead to significant increases in the success of his legislative program (Rivers andRose 1985jEdwards 1980). Sources of data include annual issues of the Statistical Abstractofthe Unitedstates;, 1982the 1982, and 1990, and 1982 editions of the Ahunac ofAmerican Politics; the Federal Election Commission, series #6 (April 1977); "1976 Senatorial Campaign Receipts and Expenditures;, 1982 3All of the adjusted R-squared values were below 55%, except for DPCUR in the presidential year analysis which yielded an adjusted R-squared value of 68%. This higher value was largely because of the correlation between DPCUR (state unemployment change) and both M P W V (presidential vote) and DPCRIPC (state income change).…”