“You Can Throw Them All in One Bag”

Outgroup Homogeneity & Negative Affect

Felix Grünewald

July 5, 2024

Introduction

„Are political opponents perceived as more homogeneous than political allies? What are the causes and consequences of such a homogeneity perception?“

Cross-national evidence for political outgroup homogeneity
Causal evidence for relationship between outgroup homogeneity & dislike

Examples

“Socialism is the mainstream of the Biden campaign.”

Donald Trump

“The obvious plan is to build a government with the Left party.” ¹

Paul Ziemiak

Why did I engage with this topic? I believe it’s an intuitive assumption that we see our opposition as a more monolithic block while differentiating much more on our own side, but I also came across a few examples of it in the real world
Trump is an example from a different context with two parties but it’s one that a lot of people know. His campaign tried to frame Joe Biden as a socialist and an accomplice of Bernie Sanders when neither Biden nor democratic socialists would probably agree with that statement.
There are also examples from multiparty contexts, like Paul Ziemiak, then General Secretary of the Christian Democrats in Germany, who here is implying a coalition between SPD, Greens and the Left party, when the socialdemocratic Olaf Scholz and the Left themselves wouldn’t have described each other to be on the best terms.
Both examples are implications of togetherness of political actors that they wouldn’t necessarily perceive themselves

Introduction

Homogeneity = less differentiated perception of the outgroup vs. ingroup
Psychology: Variability of stereotypes smaller in outgroup (Boldry et al., 2007; Ostrom & Sedikides, 1992)
Affective Polarisation: Like and dislike of political parties based on group identities
Assimilation & Divergence: Party perception dependent on favorability, distance, context (Giebler et al., 2021; Merrill et al., 2001)

So what is homogeneity? I understand it here as a lack of differentiation between different members of an outgroup
This is a familiar concept in Psychology where it has been shown that social outgroups are often perceived with less variability than the ingroup
This does not relate to the perception of stereotypes, but to how common they are in the outgroup
That’s the difference between: “Some Greens want to take away my car” vs. “All of them do”
Looking at political science, this fits right in with research on affective polarisation which is also strongly connected to ideas of group identities to explain affective relationships.
But it also builds on literature around assimilation and divergence which has been showing for a while that our party perceptions can be skewed and dependent on context and distance

Identities in Multiparty Systems

Party-identification lower (Harteveld, 2021), but spans multiple parties (Kekkonen et al., 2022)
Recent literature arguing for affect between political camps (Bantel, 2023)
Groups of parties treated as a single camp in political perception?

Polarisation and Identity

What is Affective Polarisation and (when) do we think it is a problem?
Disagreement and dislike unavoidable or even necessary (Kreiss & McGregor, 2023)
Homogeneity perception = evaluation on identity instead of positions

So why does that matter?
I believe this to be especially important in a context of affective polarisation, where we are trying to understand right now when polarisation becomes a problem and how it oculd be dangerous to society.
On the one hand, it is probably to be expected that we dislike others with opinions that are totally different or even dangerous to us.
In fact, we can argue that AP is necessary to achieve progress and equality in a democratic arena
However, what if we do not dislike them because of an opinion they hold, but because we see them as part of a group and because we ascribe characteristics to them simply because of this group affiliation?
If we see the oppositon as a more homogenous group than it is, we build our judgement of them on a basis of identity, not disagreement, and that is a very different divide to discuss and overcome.

Homogeneity and Negative Affect

Hypotheses

H1: Parties in opposing party camps are perceived as ideologically closer to each other than those in the allied camp.

H2: Perception of homogeneity in the opposite camp is negatively related to feelings towards that camp.

H3: Perception of homogeneity in the opposite camp is negatively related to vote switching.

Analysis

Data: CSES 1-5, 12 countries¹
Groups: Left-Right party camps
Homogeneity: Standard deviation of issue perception
Affect: Mean like-dislike rating of camp-parties

Results

Results - Outcamp Homogeneity

Results - Left/Right Effects

Figure 1: Stacked dataset regression on camp homogeneity perception.

H2: Homogeneity reduces Affect

Figure 2: Camp affect at different levels of homogeneity perception.

H3: Homogeneity vs. Vote Switching

Figure 3: Outcamp homogeneneity and the likelihood to switch votes.

Causality

Regression models on outcamp homogeneity and affect between waves.
	Outcamp Feeling	Outcamp SD
Previous wave: Outcamp Feeling	0.102***	0.446***
	(0.007)	(0.004)
Previous wave: Outcamp Variance	0.710***	0.049***
	(0.003)	(0.002)
(Intercept)	0.741***	0.764***
	(0.023)	(0.015)
Num.Obs.	42245	40349
R2	0.535	0.229
+ p < 0.1, * p < 0.05, p < 0.01, * p < 0.001

Robustness

Outcamp homogeneity holds for government/opposition camps (CSES) and for different issue dimensions (GLES)
Results remain robust when excluding populist parties
Controls for distance, inparty identification

Conclusion

Group homogeneity is an important concept for interparty relations
Reducing links between political camps, increasing negative affect
Possible feedback loops (Wilson et al., 2020), but important puzzle piece to explain negative affect
Judgement based on identity instead of positions

So, to conclude, these result confirm that political camps are a crucial dimension to consider when we look at multiparty systems
But more than that, the results show that homogeneity is a important concept for what we call polarisation but also for interparty relations in general
Seems to be reducing links between political camps, increasing negative affect
The effect direction here is not 100% clear and I believe that there will be somewhat of a reinforcing feedback loop (as others have suggested before) but nevertheless, outgroup homogeneity appears to be an important puzzle piece to explain affective polarisation
Especially, how does polarisation actually become a problem? And with homogeneity perception, we no longer fight about positions but about identities and those fights are not easy to settle

Next steps

Do people infer impressions of one party to the whole camp?
Can a reduction in homogeneity perception soften negative affect?

Contact:

felixgruenewald.net

@felixgruenewald

References

Bantel, I. (2023). Camps, not just parties. The dynamic foundations of affective polarization in multi-party systems. Electoral Studies, 83, 102614. https://doi.org/10.1016/j.electstud.2023.102614

Boldry, J. G., Gaertner, L., & Quinn, J. (2007). Measuring the Measures: A Meta-Analytic Investigation of the Measures of Outgroup Homogeneity. Group Processes & Intergroup Relations, 10(2), 157–178. https://doi.org/10.1177/1368430207075153

Giebler, H., Meyer, T. M., & Wagner, M. (2021). The changing meaning of left and right: Supply- and demand-side effects on the perception of party positions. Journal of Elections, Public Opinion and Parties, 31(2), 243–262. https://doi.org/10.1080/17457289.2019.1609001

Harteveld, E. (2021). Fragmented foes: Affective polarization in the multiparty context of the Netherlands. Electoral Studies, 71, 102332. https://doi.org/10.1016/j.electstud.2021.102332

Kekkonen, A., Suuronen, A., Kawecki, D., & Strandberg, K. (2022). Puzzles in affective polarization research: Party attitudes, partisan social distance, and multiple party identification. Frontiers in Political Science, 4, 920567. https://doi.org/10.3389/fpos.2022.920567

Kreiss, D., & McGregor, S. C. (2023). A review and provocation: On polarization and platforms. New Media & Society, 146144482311618. https://doi.org/10.1177/14614448231161880

Merrill, S., Grofman, B., & Adams, J. (2001). Assimilation and contrast effects in voter projections of party locations: Evidence from Norway, France, and the USA. European Journal of Political Research, 40(2), 199–221. https://doi.org/10.1111/1475-6765.00594

Ostrom, T. M., & Sedikides, C. (1992). Out-group homogeneity effects in natural and minimal groups. Psychological Bulletin, 112(3), 536–552. https://doi.org/10.1037/0033-2909.112.3.536

Sherman, A. (2020). Trump’s false claim that Biden is a socialist. In Politifact. https://www.politifact.com/factchecks/2020/oct/15/donald-trump/trumps-false-claim-biden-socialist/.

Törnberg, P. (2022). How digital media drive affective polarization through partisan sorting. Proceedings of the National Academy of Sciences, 119(42), e2207159119. https://doi.org/10.1073/pnas.2207159119

Wilson, A. E., Parker, V. A., & Feinberg, M. (2020). Polarization in the contemporary political and media landscape. Current Opinion in Behavioral Sciences, 34, 223–228. https://doi.org/10.1016/j.cobeha.2020.07.005

Appendix

Homogeneity Government/Opposition

Vote Switching GLES

Figure 4: Outcamp homogeneneity and the likelihood to switch votes between camps (GLES 2021, w15-20).

Party Camp Agreement

Follow-Up

Inference of Negativity

Generalization of negative experiences to whole group
Judgement of unrelated group members through ascribed identity instead of their positions
Conflict, not avoidance, causing affective polarization (Törnberg, 2022)

H1: Confrontation with individual members of the outgroup will have a stronger effect on the outgroup as a whole, the more homogenously that group is perceived as.

Design - Homogeneity

Please indicate where you would place [Party A] on a scale from 0 to 10, where 0 means “Extreme Left” and 10 means “Extreme Right”.

Left

Right

Design - Confrontation

Please read the following statement of a member of [Party C].

“I absolutely HATE [thing that Party A stands for]!”

Design - Effect

Please indicate where you would place [Party A] on a scale from 0 to 10, where 0 means “Extremely Favorable” and 10 means “Extreme Unfavorable”.

Favorable

Unfavorable