Article Summary
Introduction
The purpose of this article is to introduce and describe the construct of affective polarization, which had emerged as a distinct area of research at the time of the article’s publication in 2019. The authors review the connections between social identities, group sorting, and affective polarization and they discuss how affective polarization is measured, how the current ecosystem has bred such polarization, effects of affective polarization in political and nonpolitical domains, and how it might be addressed. Throughout the paper, they identify areas where current research is lacking, conflicting, or still developing.
Analytical Approach
This theoretical review article references and describes relevant research about the concept of affective polarization, including both empirical work and other theoretical work. It is a selective review article that aims to make an argument rather than systematically reviewing the state of the literature. The authors draw on key references about affective polarization to present a coherent and informative summary for the reader to understand affective polarization, where it comes from, and how it works.
Main Findings
The authors comprehensively review multiple aspects of affective polarization. In the first section, “Affective Polarization: An Outgrowth of Partisan Social Identity,” they describe how affective polarization results from social identities; from the sorting of oneself and others into political ingroups and outgroups as a way to make sense of one’s social world. This naturally leads to positive feelings toward one’s own political party and negative feelings toward the opposing party.
Following, in “How Do We Measure Affective Polarization,” the authors describe three measurement methods. Self-report measures are common, with the most central being the feeling thermometer from the American National Election Study (ANES), in which respondents rate how they feel toward the ingroup and outgroup on a scale from 0-100, and the difference between the two feeling scores represents affective polarization. This measure shows significant increases in affective polarization in recent years, nearly doubling. Implicit measures such as the Implicit Association Test (IAT) are also used and show less polarization than survey measures do. Finally, behavioral measures include economic games or studies in which partisanship of targets of judgment is manipulated. Survey measures provide the advantage of longitudinal data available for the past several decades, but also come with the drawback of potential survey demand effects. Implicit measures avoid demand effects, but do not correlate as strongly with actual behavior. Both implicit and behavioral measures do not have longitudinal data.
The authors next describe features of the contemporary environment that have created and exacerbate affective polarization. Liberal versus conservative ideologies are now sorted by Democratic and Republican party lines. The authors note that other research has shown that increasing ideological extremity increases partisan affect. Political campaigns have been shown to increase affective polarization by making partisanship more salient. The authors also conduct a deep review of research showing that the growth of partisan news outlets and one-sidedness of the content is associated with hostility toward the opposing party, though the mechanism is unclear. The authors note that the review of research reveals that it is impossible to say that partisan news actually causes affective polarization; some studies show that individuals who are already extreme are choosing partisan news sources while others find that affective polarization increased more among those who did not have access to as much partisan content. Given that social networks and family units are also becoming increasingly homogeneous along partisan lines, it’s possible that individuals are receiving partisan content through their interactions with others, not solely through news outlets. The authors conclude their review of the research on the origins of affective polarization by noting that more work needs to be done to fully understand the role of partisan echo chambers–both among social networks and news outlets–on affective polarization.
The section entitled “The Nonpolitical Consequences of Affective Polarization” describes research findings that affective polarization impacts people’s choices of friendly and romantic social partners, as well as perceptions of the economy and economic behavior, and hiring decisions. Affective polarization also encourages political participation. The economic consequences of affective polarization include perceiving that the economy will do better when one’s own party is in control, economically punishing counterpartisans, and being more likely to hire copartisans. However, most of this literature about social preferences and effects on the economy is still mixed.
In “Decreasing Affective Polarization,” the authors present several evidence-based intervention strategies, including explicitly correcting misperceived facts about opposing parties and making partisan identities less salient than other non-polarized and overarching identities such as being Americans. The authors stress, however, that changing attitudes and feelings is not easy. Further research is needed in this area.
The authors conclude the piece with further questions on affective polarization in which they identify the need for research on the psychological underpinnings of affective polarization. Sorting into political parties along ideological lines may also contribute to or explain affective polarization, and should be explored. Future research should look at when and whether people are more driven by ingroup love or outgroup hate. Lastly, how the process of affective polarization plays out among political elites, and how polarization among elites and the public relate, are open questions as well.
Implications
This paper encourages and provides the basis for a shift in the field of polarization from focusing on ideological or issue polarization to affective polarization. They demonstrate that affective polarization is pervasive, with quite strong effects on attitudes and behavior within and beyond the political realm. They also identify several key areas of further research on affective polarization, as described above. They demonstrate the need for development and testing of interventions to ameliorate affective polarization, as it can impede the political process and detract from other domains of American life as well.
Questions left unanswered
The paper acknowledges, but only very briefly discusses, the relationship between issue polarization and affective polarization. The authors describe toward the beginning of the article that issue polarization may contribute to affective polarization, and touch on this point again later in the piece, but do not thoroughly examine the potential bidirectionality or orthogonality of the relationship between the two.
The authors do not discuss how political independents do or do not experience affective polarization, or how partisans feel toward them. The paper focuses on people who identify with political parties, primarily the Democratic or Republican parties in the United States.
The authors also acknowledge themselves that their paper lacks a thorough review of comparative literature examining how affective polarization looks in countries besides the United States. They focus on Democrats and Republicans in the United States, which may be a particularly polarized environment compared to the political state of other countries.
Methods and Analysis
Was the study and its analyses pre-registered?: No
Did the study rely on proxy variables to measure polarization?: N/A
Were standard p-value thresholds used (p<.05 or 95% Confidence Intervals that don’t overlap zero)?: NA
- Largest p-value presented as significant: NA
Were correlational results interpreted with causal language?: NA
Limitations / Weaknesses
Finally, the authors’ treatment of implicit measures of affective polarization does not align with contemporary debates in the social psychology research about implicit attitudes. Scholars have recently argued that implicit attitudes do not exist as stable, underlying individual differences (Payne et al., 2017; Petty et al., 2007); a more appropriate distinction to be drawn is between implicit and explicit types of measures, rather than implicit and explicit attitudes. The authors of this piece write as if implicit attitudes exist as a stable, underlying, formulated object in people’s minds, and claim that implicit measures are more valid than explicit measures. This could be contradicted by some of the recent social psychology attitudes literature. As one scholar wrote: “measures of implicit bias are meaningful, valid, and reliable. Contrary to most assumptions, however, they are meaningful, valid, and reliable measures of situations rather than persons” (Payne et al., 2017, pg. 236). Payne, B. K., Vuletich, H. A., & Lundberg, K. B. (2017). The bias of crowds: How implicit bias bridges personal and systemic prejudice. Psychological Inquiry, 28(4), 233-248. Petty, R. E., Briñol, P., & DeMarree, K. G. (2007). The Meta–Cognitive Model (MCM) of attitudes: Implications for attitude measurement, change, and strength. Social Cognition, 25(5), 657-686.
Open Data & Analyses
Does the article make the replication data publicly available?: NA
Does the article make the replication analysis scripts publicly available?: NA
Article Citation
Iyengar, S., Lelkes, Y., Levendusky, M., Malhotra, N., Westwood, S.J. (2019). The Origins and Consequences of Affective Polarization in the United States. Annual Review of Political Science, Vol. 22:129-146, https://doi.org/10.1146/annurev-polisci-051117-073034.
Bibtex
1
2
3
4
5
6
7
8
9
10
@article{doi:10.1146/annurev-polisci-051117-073034,
author = {Iyengar, Shanto and Lelkes, Yphtach and Levendusky, Matthew and Malhotra, Neil and Westwood, Sean J.},
title = {The Origins and Consequences of Affective Polarization in the United States},
journal = {Annual Review of Political Science},
volume = {22},
number = {1},
pages = {129-146},
year = {2019},
doi = {10.1146/annurev-polisci-051117-073034}