Political trust among European youth: Evaluating multi-dimensionality and cross-national measurement comparability

Abstract

Educational research has focused increasingly on comparing student civic outcomes across various educational systems. However, the study of non-cognitive, attitudinal measures has assumed a subsidiary role to the study of cognitive measures, despite its potential to shape policy and practice. This article investigates the validity and cross-country measurement equivalence or invariance (ME/I) of political trust measures among adolescents. Drawing on student data from ICCS 2016 covering 15 European educational systems (N = 52,788), the standard test for exact ME/I (i.e., Multi-Group Confirmatory Factor Analysis) is juxtaposed with a novel method for approximate ME/I (i.e., Alignment Optimization). The results underscore a two-dimensional structure of political trust, separating trust in order institutions from trust in representative institutions. While MGCFA results suggest configural invariance, rendering correlational analyses and country mean comparisons meaningless, Alignment Optimization yields valid latent means with an acceptable degree of non-invariance (i.e., 16%). Implications for comparative research on political trust among adolescents are discussed.