The Multiple Matching Perspective on Value versus Identity: Investigating How Political Ideology and Party Identity Contribute to Citizens’ Support for Political Candidates

Abstract

Many past studies tested effects of political value (i.e., ideology) and identity (i.e., party identity) on support for political candidates. Little has compared their effects by considering them from the perspectives of citizens and the candidate. This paper takes them into consideration by introducing and using the multiple matching perspective. It compares how much the predictive power of citizens’ ideology and political identity are moderated by a candidate’s ideology as well as the candidate’s party affiliation. Therefore, four types of matches (i.e., statistical moderations) are compared under this perspective: A. citizens’ ideology-candidates’ ideology match, B. citizens’ party-candidates’ ideology cross-match, C. citizens’ ideology-candidates’ party cross-match, and D. citizens’ party-candidates’ party match. The following novel and nuanced patterns about the role of ideology and party identity are uncovered under this perspective using four studies (total N=41,986): 1. The effect of ideological match (A) is large, robust, and consistent. 2. The moderating effect of candidates’ ideology (A and B) is more powerful than candidates’ party affiliation (C and D) except during the final stage of a presidential race (when A and D are similar). 3. Citizens’ party identity may guide them to support a candidate whose ideology is congruent with the party that citizens identify with (B), but it is less so for the reverse of it (less evidence for C). Finally, a supplemental study shows that citizens’ ideology is more central to their self-concept and more moralized than their party identity, and this may explain the primacy of ideology over party identity.

Politicians’ Ideology Matters More Than Their Race in Determining the Association Between White Identity and Evaluation of the Politicians

“… 3. Although a detailed discussion of ideology is beyond the scope of this article, it should be noted that scholars continue to debate the utility of ideology and its measures, particularly vis-à-vis party identification (e.g., Ellis & Stimson, 2012; Kinder & Kalmoe, 2017; Mason, 2018). Nonetheless, emerging evidence suggests that, after correcting for potential omitted variable bias, ideology, measured on a 7-point scale like in the current studies, performs very well in predicting candidate evaluation (Bai, 2020). …”

“…The gender and party factors are unrelated to the hypotheses, so their effects are theoretically treated as random effects (see below). This procedure was demonstrated by Bai (2021, Appendix B) as effective.…”