Political Tribalism in Today’s America

Guest Post by Michael Bernstein

This is a second guest post by Michael Bernstein, an assistant professor at Brown University. You can find out more about him at his first guest post, Who Agrees with Hitler?

I am especially pleased to present this one because it is a summary of the first article to appear in The Journal of Open Inquiry in the Behavioral Sciences (JOIBS) an initiative of The Society for Open Inquiry in the Behavioral Sciences (SOIBS). SOIBS was created as a refuge from the ongoing political corruption of academia and JOIBS is part of that initiative. I am the founding editor and current Editor in Chief. You can find the full article that is the topic of this guest essay here.

---

“All of my folks hate all of your folks. It’s as American as apple pie.” – Tom Lehrer, National Brotherhood Week

Michael H. Bernstein, Scott D. Martin, Nicholas S. Zambrotta .

This past summer, the FBI raided Donald Trump’s Mar-a-Lago residence and seized what were likely classified materials he illegally possessed.  This seemingly set off a bizarre string of events where Joe Biden and Mike Pence then disclosed that they also possessed classified documents.

To psychologists, what was noteworthy about these events was not the possession of the documents, per se. Rather, it was the constant finger-pointing from each political side. Partisans were quick to accuse their political adversary of acting improperly but were reluctant to admit that a politician on their own side was guilty of the same.

While this is one of the more salient recent examples, partisan double-standards seem to be everywhere. Differential treatment of candidates lied at the heart of the recently published Durham Report. Also, consider that Republicans blocked Merrick Garland from joining the Supreme Court because his appointment came in the waning months of Obama’s presidency, yet were quick to confirm Amy Coney Barrett near the end of Trump’s term. Democrats have been eager to point to Trump’s racially insensitive remarks, but appear dismissive of such statements by Biden. I could keep going, but you get the point.

We have been interested in these sorts of double-standards for a few years. We recently published our research (Bernstein et al., 2023) in the Journal of Open Inquiry in the Behavioral Sciences.

In our first study[1], we surveyed a group of college students (n=139) and a group of online volunteers (n=192) who identified as either Republican or Democrat[2]. We assembled 10 political policies/actions that, in reality, occurred under both President Obama and President Trump. Participants were asked how much they agreed with each. For example, during their presidential term, both Trump and Biden ordered a series of drone strikes, and participants were presented with “It has been reported that president Trump/Obama [participants only saw one of these names] ordered a drone strike in the Middle East that resulted in the death of suspected terrorists. There were likely also civilian casualties. To what extent to you support drone strikes that kill terrorists but probably also kill civilians?” Response options were on a 5-point scale and whether participants were told it was Trump or Obama was determined randomly, though again, the statement is truthful in both circumstances. This means that about half the time participants rated how much they supported the policy from a President on their side of the aisle and the other half of the time participants rated how much they supported a policy from a president on the other side. The items covered a wide range of topics including: unemployment, immigration, Iranian sanctions, hate crimes, meeting with a controversial rapper, China tariffs, holding immigrants in cells, the cost of prescription drugs, and the opioid crisis.

By looking at the interaction between which president the policy was attributed to (Trump v. Obama) and the participant’s political affiliation (Republican vs. Democrat), we could test the existence of political tribalism. This interaction effect was significant for 9/10 items in the college sample and 6/10 items in the online volunteer sample. In other words, Democrats generally support the same policy more when attributed to Obama versus Trump while Republicans generally support the same policy more when attributed to Trump versus Obama. The college student results are shown in Figure 1 below where scores greater than 0 indicate pro-Obama bias and scores lower than 0 indicate pro-Trump bias.  For instance, in the case of drone strikes, Republicans gave it an agreement rating of 0.55 higher in the Trump versus Obama frame, and Democrats gave it an agreement rating of 0.30 higher in the Obama versus Trump frame.

Figure 1. Political Tribalism Across 10 Policies

We also wanted to examine whether perceptions of bigotry would depend on the speaker. In the most dramatic example from our study, using the same college student sample, we took a real quotation from Dr. Martin Luther King Jr. The statement could be read as an All Lives Matter slogan. It began with “Black supremacy is as dangerous as white supremacy” and ended with “God is interested in the freedom of the whole human race.” We told participants that the statement was either made by MLK or by Donald Trump, and they were asked how racist the speech was from “not at all” to “extremely.” The interaction effect between speaker frame (MLK vs. Trump) and participant political affiliation (Republican vs. Democrat) was an astounding F=309 (p<.001). As shown in Figure 2, Republicans didn’t think the speech was racist when uttered by either MLK or Trump. Democrats agreed the speech was not racist when uttered by MLK, but rated it much higher in racism when uttered by Trump. (In a follow-up study [Bernstein, Brown, Zambrotta, & Bleske-Rechek, in prep], participants were shown 2 real Biden quotes and 2 real Trump quotes, randomly attributed to either, and asked to indicate how bigoted the statement was. Here again we see clear evidence that attitudes depend on the speaker, although unlike the MLK/Trump experiment, this was more so the case for Republicans than Democrats).

Figure 2. Perceived Racism of Speech

Next, we turned our attention to what, at that time, was the upcoming 2020 presidential election. As you may remember, even before election day, there was speculation about election interference, and whether this would be a “clean election.” Some readers may be inclined to roll their eyes at this as we have yet to see compelling evidence that the election was stolen from Trump despite many claims (including from the former president himself) that it was.  However, there are all sorts of genuinely ambiguous situations. Consider this question we asked: “Imagine Biden/Trump wins the election and there is evidence that the Chinese (if Biden frame)/Russian (if Trump frame) government attempted to influence the election.” Is Biden/Trump the legitimate president? Response options were simply “yes,” “no,” and “unsure.”

Figure 3a below shows what we found among the 467 online volunteers who participated. Again, we see evidence for tribalism, with Republicans viewing Trump as more legitimate when he wins despite Russian influence compared to Biden winning despite Chinese influence. And Democrats had the opposite trend, with Biden being seen as more legitimate than Trump when there is foul play from our adversaries abroad (note that lower values on the figure indicate more perceived legitimacy). Similar findings emerged when we asked about whether Biden or Trump would be the legitimate president if there is influence from both China and Russia (Figure 3b) and if there were uncounted ballots for the losing candidate (Figure 3c)

Figure 3. Perceived legitimacy of Hypothetical Election Results

Which side is more biased?

We designed this research to examine whether partisans are biased. We were not especially interested in asking which side is more biased, although this question is hard to avoid. It is also the source of debate. Jost (2021) has argued that conservatives are more dogmatic and rigid than Democrats. But in a recent meta-analysis (Ditto et al., 2019), conservatives and liberals seemed to be equally bias. In our research, we looked at bias from a variety of angles. But at its core, we considered bias to entail rating some action or statement as more favorable (or less unfavorable) when attributed to one’s own side versus the opposing side. In general, we found that bias was somewhat higher among Democrats than Republicans across most of our tests, though there was plenty of bias from each side. Interested readers should take a look at the paper for a more complete description.

What do these findings mean?

We live in an era of partisanship that is becoming increasingly more extreme. Everything appears to be politicized, with differences even emerging in response to the COVID-19 pandemic (Conway et al., 2020; Micalizzi et al., 2021). More than 1 in 3 people say they would be somewhat or very unhappy if their child married someone from the opposing political party (Najle & Jones, 2019). The existence of tribalism suggests that people are not objectively judging policy and statements. This poses a serious concern for our ability to effectively communicate and reach consensus about political and social issues. How can we forthrightly debate the merits of presidential actions, the bigotry of statements, and the legitimacy of elections when attitudes are, to some degree, a function of the interaction between the speaker/politician and one’s own political party? [3]

References

Bernstein,  M.H.,  Zambrotta,  N.S.,  Martin,  S.D.,  Micalizzi,  L.  (2023). Tribalism  in  American  Politics:  Are Partisans  Guilty  of  Double-Standards? Journal  of  Open  Inquiry  in  the  Behavioral  Sciences

Conway III, L. G., Woodard, S. R., Zubrod, A., & Chan, L. (2020). Why are conservatives less concerned about the coronavirus (COVID-19) than liberals? Testing experiential versus political explanations. PsyArXiv Preprints.

Ditto, P. H., Liu, B. S., Clark, C. J., Wojcik, S. P., Chen, E. E., Grady, R. H., . . . Zinger, J. F. (2019). At Least Bias Is Bipartisan: A Meta-Analytic Comparison of Partisan Bias in Liberals and Conservatives. Perspectives on Psychological Science, 14(2), 273-291.

Jost, J. T. (2021). Left and Right: The Psychological Significance of a Political Distinction: Oxford University Press

Micalizzi, L., Zambrotta, N. S., & Bernstein, M. H. (2021). Stockpiling in the time of COVID-19. British Journal of Health Psychology, 26(2), 535-543.

Najle, M., & Jones, R. P. (2019). Pluralism in a Divided Nation. Retrieved from https://www.prri.org/research/american-democracy-in-crisis-the-fate-of-pluralism-in-adivided-nation/

---

Footnotes

[1] This was our first chronological study, though it’s presented as Study 2 in the paper

[2] This was determined by who the participant planned to vote for in the 2020 election

[3] The final few sentences of the post are taken (with permission) nearly verbatim from Bernstein et al., 2023.

---