Establishing the construct and predictive validity of brief measures of affective polarization

Abstract

Measuring affective polarization, defined as the liking for one's political ingroup and the dislike for political outgroups, poses methodological challenges in multiparty systems: evaluations of seven, 13 or even more parties in a survey are costly, time-consuming and demanding. Some studies therefore use subsets of parties to create brief affective polarization measures. However, it is unclear how this affects the construct and predictive validity of these brief measures, potentially causing problematic inferences. Across 39 countries ($N=66,880$), we demonstrate that brief measures that include ratings of only three to five parties can maintain acceptable validity, as illustrated by strong correlations with full measures and consistent associations with political correlates. The construct and predictive validity of brief measures are best when selecting a set of large, ideologically diverse parties. We provide specific recommendations for the effective measurement of affective polarization in different multiparty systems.