Article Summary
Introduction
One explanation for high levels of affective polarization in the U.S. comes from social identity theory, which suggests that as conflict between Republicans and Democrats has risen over the last several decades, people’s partisan identities become closer to the conception of self. The authors aim to test whether social identity theory is a good descriptor of political identities in the U.S. by measuring the magnitude of in-party love and out-party hostility in political and non-political settings. They also aim to test the relative magnitude of affective polarization in political and non-political settings.
Analytical Approach
The authors distributed a survey to 303 undergraduates. Embedded within were 10 items measuring trust in various political (e.g. “I trust ___ members of Congress to make decisions in a fair way”) and non-political groups (e.g. “I trust ___ business owners to pay their workers a fair wage”). They experimentally manipulated whether participants were asked about the in-party, out-party, or were given no cue about party.
Main Findings
First, they found that polarization (defined as the difference in trust between in-partisans and out-partisans) was greater for the political groups than the non-political groups. While there was polarization regarding the non-political groups (.15 difference between in-party and out-party on a 0-1 scale), the polarization for political groups was nearly twice as large (.26).
Next, in order to establish the relative importance of in-group love and out-group hostility, they compared trust in in-partisans and out-partisans to the control condition where no party information was given. With regard to the non-political groups, participants generally preferred in-partisans to the neutral control, and there was no strong difference between control and out-partisans. Meaning, they find consistent evidence of in-group love but not out-group hostility in non-political settings. In political settings, they again find similar evidence of in-group love (i.e. participants report more trust in in-party political actors than unidentified political actors). However, in the political domain, participants distrusted out-group actors even more than they trusted in-group members (both compared to control). Further, they find that how strong an individual’s party identity is moderates the effect of the party cue manipulation on trust.
Implications
The authors find consistent evidence of partisan gaps in trust of both political and non-political actors. This speaks to the widespread nature of polarization and its tendency to spillover into non-political domains. The authors find that in-group love contributes to polarization in both political and non-political domains, while out-group hate contributes only in the political domain, but is stronger than in-group love in that domain. They argue that the strength of out-group hate in the political domain warrants a reconsideration of social identity theory as it applies to partisanship, though this argument rests on a limited reading of social identity theory.
Questions left unanswered
The authors use an a priori distinction between political and non-political groups. Some questions for future analysis are: which groups/actors are considered political or non-political by laypeople, and how are ostensibly non-political groups (teachers, nurses, LGBTQ people) brought in or out of politics depending on the social context.
Methods and Analysis
Was the study and its analyses pre-registered?: No
Did the study rely on proxy variables to measure polarization?: Yes
They compared trust levels in inparty and outparty actors.
Were standard p-value thresholds used (p<.05 or 95% Confidence Intervals that don’t overlap zero)?: Yes
- Largest p-value presented as significant: 0.05
Were correlational results interpreted with causal language?: No
Limitations / Weaknesses
While the data presented are useful, the framing rests on a fundamental misunderstanding of social identity theory (SIT). The authors argue that the outsize strength of out-group hostility, compared to in-group love, in political settings warrants a re-examination of whether SIT applies in the political context, when the results are in fact precisely in line with SIT’s predictions. SIT posits that intergroup discrimination springs initially from in-group preference, rather than out-group derogation, but does not make the argument that out-group hostility can never exist alongside (and even surpass) in-group love in certain contexts. Indeed, the Brewer (1999) article the authors offer a limited quote from contains a lengthy section entitled “Ingroup Preference as a Platform for Outgroup Hate” in which she argues that in the contexts where intergroup divisions are moralized, threaten the in-group, have opposing goals, have opposing values, and/or are politicized, they are extremely likely to lead to high level of out-group derogation. Clearly, partisan identities in the U.S. fit all five of those criteria, and would be expected to produce both in-group love and out-group hostility - the precise pattern established in this paper.
Open Data & Analyses
Does the article make the replication data publicly available?: No
Does the article make the replication analysis scripts publicly available?: No
Article Citation
Rudolph, T. J., & Hetherington, M. J. (2021). Affective Polarization in Political and Nonpolitical Settings. International Journal of Public Opinion Research, 33(3), 591–606. https://doi.org/10.1093/ijpor/edaa040
Bibtex
1
2
3
4
5
6
7
8
9
10
11
12
@article{Rudolph2021,
author = {Rudolph, Thomas J. and Hetherington, Marc J.},
doi = {10.1093/ijpor/edaa040},
file = {:Users/xanni/Documents/Readings for Mendeley/edaa040.pdf:pdf},
issn = {14716909},
journal = {International Journal of Public Opinion Research},
number = {3},
pages = {591--606},
title = ,
volume = {33},
year = {2021}
}