Mid Sweden University

How does elite communication influence affective polarization between partisan groups? Drawing on the literature on partisan source cues, we expect that communication from in- or outgroup party representatives will increase affective polarization. We argue that polarized social identities are reinforced by partisan source cues, which bias perceptions of elite communication and result in increased intergroup differentiation. Further, we expect that the effect of such source cues is greater for voters with stronger partisan affinities. To evaluate our hypotheses, we performed a survey experiment among about 1300 voters in Sweden. Our analyses show that individuals who received a factual political message with a source cue from an in- or outgroup representative exhibited higher affective polarization, especially when they already held strong partisan affinities. This suggests that political elites can increase affective polarization by reinforcing existing group identities, and that this occurs in conjunction with biased interpretation of elite communication. The results improve our understanding of how political elites can influence affective polarization and add to previous research on party cues and attitude formation by demonstrating that such source cues can also increase intergroup differentiation.

After a period in which affective polarization—defined here as the difference between positive feelings toward in-parties and negative out-party animus—has mostly focused on the single US case, there has recently been an increase in large-N comparative analyses and single case studies in other countries, including in the Nordic region. This study adds to this by studying and comparing affective polarization in the Nordic countries. In line with what previous comparative and single case studies have already indicated, the results show that affective polarization has tended to be higher in Sweden and Denmark than in Norway, Iceland, and Finland. The article also tracks time trends for the association between ideological distance from parties and affective party evaluations. As expected, placing parties further from oneself on the left-right scale has been more strongly associated with party affect in Denmark and Sweden. Furthermore, the results show that there are some variations between the countries in terms of how distance from parties on other ideological dimensions than left-right placement is associated with out-party affect. 

Political polarization has become one of the buzzwords of the 21st century political vocabulary. Several political and societal developments, such as the success of populist parties, negative discord on social media, and the smearing of political opponents, to name just a few, all get lumped together under the umbrella term of increased political polarization. However, in contrast to the traditional conceptualization of political polarization as policy disagreement, these developments are arguably better captured by what researchers have labeled affective political polarization, i.e., the difference between positive feelings toward ingroup parties/partisans and negative feelings toward outgroup parties/partisans. Affective polarization is often described as both normatively and empirically distinct from polarization in terms of ideology and policy differences. Moreover, it has been ascribed several important consequences, ranging from decreased trust in political institutions and condoning violations of democratic norms when the in-party is in government to increased political activism and discrimination in nonpolitical settings.

Yet, despite its apparent relevance, it was not until recently that research on affective polarization started gaining traction outside of the US. The aim of this thesis is to study affective polarization in Nordic countries that are often viewed as consensual, egalitarian and compromise-oriented political systems. This has been done with single case studies of Sweden (two articles), Norway (one article), and one article that focuses on all five countries together. Thus, the first research question compares how affective polarization has changed over time in Nordic countries. Is the perceived increase in polarization mentioned above also visible in terms of affective polarization? What differences have there been between the Nordic countries that are otherwise thought of as similar political systems?

Next, expanding affective polarization research from US two-party systems to other multiparty systems presents methodological challenges. A second overarching research question therefore explores how patterns of affective polarization can be understood by breaking down the electorate into supporters of different party blocs and by examining how the importance of ideology for affective party evaluations has changed over time. While one objective of the thesis consists of outlining patterns of affective polarization, another aim has been to explore some of its causes and consequences. More specifically, the research questions examine factors that go together with higher levels of affective polarization and how negative out-party affect impacted attitudes about the political response to the coronavirus in Norway.

The thesis relies on survey data of representative samples of citizens from the respective countries, spanning the course of several decades. Furthermore, to provide causal leverage when exploring the causes and consequences of affective polarization, I also make use of survey experiments and panel data. The results suggest that affective polarization has increased in the Nordic region as a whole during the last election years, thus pointing to the relevance of the subject. However, in contrast to the typical view of Nordic countries as similar systems, it has tended to be higher in Sweden and Denmark than in the other three countries, at least if we focus on the 2010s. In addition, the articles illustrate several important correlates of affective polarization and, in particular, that a subtle priming of people’s partisanship can be enough to trigger it. Finally, with a design based on previous research in the US, the results show, among other things, that dislike toward out-parties influenced attitudes about the response to the coronavirus crisis in Norway, although this effect depended on whether people evaluated the response of the government or Norway more generally.

Research on affective polarization and negative partisanship toward disliked out-parties has increased significantly in recent years. However, there are surprisingly few studies that actually examine its political consequences, especially outside of the US. This study relies on two survey experiments to examine how dislike toward out-parties affected how Norwegian citizens evaluated the country's response to the coronavirus crisis. The first experiment follows the example of previous research on the US case and tests how out-party dislike measured before the coronavirus outbreak affected subsequent attitudes about how Norway and the conservative government had managed the crisis. The second experiment then randomly assigns party cues to a policy proposal included in the country's economic rescue package and tests whether like-dislike party evaluations moderate the effect of receiving the party cues. Overall, the results show that out-party dislike predicted attitudes to the government's response, but, contrary to studies focusing on the US case, this effect was either nonexistent or weaker for those who rated the country's response. Additionally, while out-party cues polarized opinions to the proposal, the moderating effect of out-party dislike was only more consistently found for those who received party cues from the populist-right party.

Currently, very little is known about the individual level foundations of affective polarization outside the US context. This paper addresses this research gap by exploring the predictors of affective polarization in Sweden, using nationally representative survey data. From 2010 till the 2018 elections, the Swedish multiparty system was divided into two affectively converged mainstream party blocs that were moderately polarized between each other, and an asymmetrically polarized populist right party that invoked significantly more hostility from the supporters of mainstream parties than vice versa. To unpack the foundations of affective polarization in such a tripolar configuration, we employ a methodological approach that distinguishes not just between in- and outparties but also in- and out-blocs, and separates different conflict dimensions. We find that: (1) voters with stronger partisan identities and more extreme political attitudes exhibit higher levels of affective polarization. The latter, however, is a better predictor of direct dislike towards political opponents; (2) the effects of political attitudes correspond to the tripolar nature of the party system: while affective polarization between the centre-right and -left blocs is mostly driven by socioeconomic positions, the hostility towards Sweden Democrats links predominantly to cultural issues, most notably immigration; (3) institutional trust has a two-edged relationship with affective polarization: populist right voters that trust the country’s central institutions more, are less polarized towards mainstream blocs, whereas among centre-right voters, higher trust associates with stronger animosity towards Sweden Democrats. We believe that these findings could have broad implications for affective polarization research in multiparty systems. 

I denna antologi presenterar 13 forskare studier om svenska riksdagsledamöter. Vi får bland annat veta hur väl riksdagsledamöternas åsikter överensstämmer med väljarnas – både när det gäller politiska sakfrågor och hur de demokratiska spelreglerna bör se ut. Några kapitel handlar om social representation, särskilt med betoning på kvinnor och män i politiken och feministerna i riksdagen undersöks. En studie berör hur politiska erfarenheter från lokal och regional nivå påverkar riksdagsledamöterna. Två kapitel handlar om hur olika partierna är åsiktsmässigt. Kontakter mellan medborgare och riksdagsledamöter, social tillit och upplevelser av politisk makt är andra teman som behandlas. Bidragen bygger på resultat från Riksdagsundersökningen (RDU, en enkät riktad till samtliga riksdagsledamöter som genomförts vid Göteborgs universitet sedan 1960-talet. Antologins redaktör, David Karlsson, är undersökningsledare för RDU och verksam vid Förvaltningshögskolan, Göteborgs universitet. Medverkande författare från Statsvetenskapliga institutionen, Göteborgs universitet, är: Dennis Andersson, Peter Esaiasson, Sören Holmberg, Elin Naurin, Henrik Ekengren Oscarsson, Helena Olofsdotter Stensöta, Lena Wängnerud och Patrik Öhberg. Därtill medverkar Magnus Hagevi (Linnéuniversitetet), Lukas Lindstrand (tidigare SOM-institutet, Göteborgs universitet), Alexander Reiljan (European University Institute) samt Alexander Ryan (Mittuniversitetet).