Key points are not available for this paper at this time.
This methodological paper uses a new conceptualization of attitudes, the Causal Attitude Network (CAN) model, to investigate populism and nativism. The CAN model conceptualizes attitudes as a complex system of interrelated factors and captures different attitude components' interplay and mutual dependence. Drawing on two previously collected, representative datasets (N1 = 8046, N2 = 2030) from 15 European countries, we estimated the networks structure of populist attitudes (PA), nativism, conspiracy mentality, threat appraisals, and demographics. Besides confirmatory factor analyses (CFA) and exploratory factor analyses (EFA), our general procedure included the estimation of Mixed Graphical Models (MGMs) with LASSO Regularization and the Extended Bayesian Information Criterion (EBIC), before we analyzed our network models with the Walktrap algorithm and Network Comparison Test (NCT). A cluster analysis based on different algorithms (hierarchical, clara, pam) divided the countries into two clusters, while the main difference between them was how the networks related to perceiving politicians as corrupt elites. To gain a deeper understanding of the PA scale, we investigated it from a network perspective. We found that the PA network of Turkey (a hybrid case of religious-conservative and nationalist-authoritarian) differed the most from all other countries under investigation.
Abadi et al. (Tue,) studied this question.