Abstract Public opinion research has intensively examined mass ideological polarization, often linking it to outcomes that threaten key democratic ideals. Both theory and empirical evidence suggest that it comprises multiple manifestations, with their relationships, however, remaining largely unclear. Meanwhile, previous studies have seldom conducted cross-contextual comparisons, which are essential for understanding how macrostructural factors shape mass polarization. These limitations partly arise from the methodological challenges that hinder consistent measurement across diverse contexts. To fill these gaps, this study introduces a belief network approach to compare three core manifestations of mass ideological polarization: disagreement, symbolic-operational ideological alignment, and within-operational ideological alignment. Using data from 78 societies in the World Values Survey and the European Values Study (2017–2023), we employ improved network-based measures that capture ideological alignment without relying on context-specific assumptions. Surprisingly, our findings reveal substantial cross-national variation and contrasting correlations among these manifestations, indicating that mass ideological polarization is neither singular nor universal. We further apply the network approach to examine how modernization, a critical macrostructural factor, shapes ideological polarization, showing that it predicts lower disagreement but higher ideological alignment. By offering a scalable and context-adaptable tool for comparative public opinion research, this study proposes an expanded theoretical framework of mass polarization, highlighting the need to distinguish its diverse manifestations and their structural drivers across global contexts.
Guo et al. (Sat,) studied this question.