The apolitical side of affective polarization: Examining the role of perceived non-political differences and similarities among voters

Previous research has shown that affective polarization is higher when citizens are more ideologically and politically different: citizens who are further away from each other on the political spectrum dislike each other more (Harteveld, 2021a; van Erkel & Turkenburg, 2022). While political factors like these are extremely important to understand affective polarization among citizens, we examine if other, non-political, factors are also related to citizens’ negative affect towards political others. In particular, we ask: are citizens more (or less) affectively polarized the more (or less) they perceive political others as different from themselves with regards to non-political features, such as their lifestyle, personality, or norms and values? Put differently, is disliking political others not only connected to differences in political views, but also to the perception that these citizens are different human beings altogether? Recent research indicates substantial links between citizens’ actual political and non-political preferences (e.g., lifestyle choices, DellaPosta et al., 2015) and shows that strong actual alignment between one’s political and other social identities (i.e., social sorting) increases affective polarization (Harteveld, 2021b; Mason, 2016). In this paper, we shift focus to the still more scarcely studied role of perceptions. Following Tsoulou-Malakoudi et al. (2023, p. 1), we study and define perceptions of differentness and similarity (PoDS) as "citizens’ perception that politically other- or like-minded citizens are different and/or similar from themselves with regard to attributes that are not obviously political in nature", and examine its link with affective polarization. We first expect that the stronger citizens’ perception of differentness, the more they are affectively polarized (H1). Second, we expect this relationship to be stronger for citizens with a strong positive partisan identity (H2a), a strong negative partisan identity (H2b), and for more extreme party voters (H2c). Third, we compare the importance of perceived non-political differences to perceived political differences and expect political differences to have a bigger ‘net effect’, especially among stronger partisans (H3). Finally, we examine if certain non-political characteristics are associated more strongly with affective polarization than others, thereby comparing the role of lifestyle, personality, norms and values, and socio-economic position (RQ1). To study this, we conduct a survey among ca. 7000 Belgian citizens (quota-based sample for gender, age, education level, region). The data collection starts in February 2024, but preliminary findings of a pretest (N=2307) already indicate that affective polarization is significantly higher the more citizens perceive political others to be different from themselves with regards to various non-political features. Conversely, citizens who perceive themselves to be more similar to political others are less polarized. Citizens moreover seem to have little difficulty with describing political others in non-political ways, implying that citizens easily make these inferences from the political to the non-political. In the next steps we dive deeper into these main findings, gathering more insight on the moderating role of partisan identities and on the various types of non-political features and how they may (or may not) relate to affective polarization. Doing so, this study contributes to a better understanding of horizontal affective polarization.