Researchers measured conventionalism via matched incumbent House election margins across threat levels

Evidence

Conventionalism, the “rigid adherence to conventional, middle-class values” (Adorno et al., 1950), is difficult to measure directly. Drawing on voting statistics, we developed a related measure of politico-economic conservatism, which is arguably one component of conventionalism. To avoid confounding by presidential charisma and the staggered Senate election cycle, we examined House of Representatives election returns (World Almanac and Book of Facts, 1972–1990) for 1978 (when threat levels were on the upswing) and 1986 (when threat levels had greatly receded). Did the higher and increasing threat of 1978 induce greater support for conservative candidates, compared with the low-threat election of 1986? In 1978, the Democrats won 277 seats in the House of Representatives, as compared with 258 in 1986, which argues against our hypothesis. However, these crude totals are affected by extraneous factors such as the “coattails” effect, incumbency, and differences in the candidates themselves. Moreover, party labels are not a very reliable guide to liberal-versus-conservative ideology. To control for these factors, we considered only cases in which the same incumbent stood for reelection in both 1978 and 1986. We chose incumbents from the two extremes of the ideological spectrum on the basis of overall ratings (from 1978–1986) by the liberal Americans for Democratic Action (ADA), the conservative Americans for Constitutional Action (ACA), and the American Conservative Union (ACU; Barone & Ujifusa, 1981, 1983–1987). Our final sample included the 30 most conservative (having the highest ACA or ACU minus ADA rating) and the 30 most liberal (having the highest ADA minus ACA or ACU rating) incumbent House candidates. From the election data reported in the same sources, we calculated the percentage margin of victory over the principal opponent in both 1978 and 1986 and then calculated the change in this margin between the two elections. — John T. Jost and Jim Sidanius, Political Psychology, 808

Categorization

• DEFINE: Illustrates a careful methodological design: using matched same-incumbent race margins to isolate ideological voting effects from noise.
• XREF: connects to: Field’s titling study results may reflect political context rather than property effects — Philip Mirowski & Dieter Plehwe (eds.), The Road from Mont Pèlerin: The Making…, 1929; Symbolic threat outperformed realistic threat in predicting mayoral vote choice — Blake M. Riek, Eric W. Mania, and Samuel L. Gaertner, Intergroup Threat and Outgroup…, 34