Bias
Is Discrimination the Rule or the Exception?
Three recent publications show very low levels of discrimination.
Posted June 25, 2021 Reviewed by Devon Frye
Key points
- Three papers report that discrimination is the exception rather than the rule.
- This finding occurred despite using very different methods to study discrimination, increasing confidence in the validity of the conclusion.
- Low levels of discrimination might explain why diversity and implicit bias trainings have so little effect.
How widespread is racial discrimination? As the culture wars have heated up again, popular analyses, such as Kendi’s How to Be an Anti-Racist and Di’Angelo’s White Fragility, and the Pulitzer Prize-Winning 1619 Project are part of a wider social discourse implying or stating that racial discrimination is massively widespread because it is deeply ingrained in the soul of America as a society and Americans as a people.
The social science scholarship on discrimination is massive, and I won’t be reviewing it all here. Instead, however, I will be summarizing key results regarding discrimination from three papers that have all been recently published or accepted for publication.
![Lee Jussim Lee Jussim](https://cdn.psychologytoday.com/sites/default/files/styles/article-inline-half-caption/public/field_blog_entry_images/2021-06/karl_popper_street_1.jpg?itok=MCHZBjmR)
Although none were framed in exactly this way, one can think of all three as testing three competing hypotheses:
- There is no racial discrimination.
- There is massive racial discrimination; it occurs in most interracial interactions.
- There is some racial discrimination, but it is a relatively rare exception.
These are mutually exclusive hypotheses.
Publication #1: Discrimination Occurs 1.3 Percent of the Time
Peyton and Huber, publishing in the Journal of Politics, found anti-Black discrimination 1.3 percent of the time, which is the same as saying they found no anti-Black discrimination the other 98.7 percent of the time. In the study, they had over 700 people play the ultimatum game with either Black or White partners. This is a game often used in experimental studies. The first player proposes to the second how to divide some money. For example, the first player may be given a dollar to divide, and offers 30 cents to the second. If the second player accepts, then the first gets 70 cents and the second gets 30 cents. If the second rejects this division, neither gets anything.
The game is interesting, in part, because it pits naked self-interest against fairness. From a naked self-interest standpoint, second players should accept any non-zero offer, because they end up with more money accepting than rejecting the offer. From a “fairness” perspective, anything other than 50-50 is unfair. In the ultimatum game, the two people interact anonymously and only once, so there are no opportunities to reap benefits from either punishment or cooperation.
Over 700 people in this study played the ultimatum game 25 times with either Black or White partners, so the total number of offers accepted or refused was over 18,000. The authors wrote, “Racial discrimination occurs when a white individual rejects an offer from a Black individual that would be accepted if offered by a white individual.” This happened 1.3 percent of the time.
The participants in this study were Mechanical Turk workers, which is important because they are not a representative sample of Americans. Whether the 1.3 percent figure would generalize to “Americans” is unknowable from this study. Also, whereas the 98.7 percent nondiscrimination is very high, it was not a real-world context. Although this renders its implications for real-world discrimination unclear, it is consistent only with the third hypothesis, “There is some racial discrimination, but it is the relatively rare exception.”
![Lee Jussim Lee Jussim](https://cdn.psychologytoday.com/sites/default/files/styles/article-inline-half-caption/public/field_blog_entry_images/2021-06/1._mushroom_henge_82.jpg?itok=m1G2RQ-G)
Publication #2: Discrimination in 5 to 20 Percent of Responses
Campbell and Brauer (2021), writing in the Journal of Experimental Psychology: General, addressed many of the limitations of the Peyton and Huber study. Rather than study discrimination in an interesting but ultimately artificial lab setting, they studied it in the real world—on their college campus. They performed a climate study, which, among other things, asked white students and students of color (their terms) how often they felt welcomed or respected. Their samples included over 1,400 students of color and over 5,700 white students.
The climate study found that nearly two-thirds of students of color reported feeling welcomed and respected very often or extremely often. This proportion was below that of white students (about 85 percent of whom felt welcomed and respected very or extremely often), so the findings did not support hypothesis 1, “No racial discrimination.” However, when nearly two-thirds of students of color feel generally welcome and respected, nor did it support hypothesis 2, “Massive discrimination.” Again, the main result supported hypothesis 3, “There is some racial discrimination, but it is the relatively rare exception.”
Campbell and Brauer also conducted five additional studies assessing actual discrimination unobtrusively throughout their campus. For example, in one study, they had either a Black or White actor follow right behind a student entering a door. The key discrimination result was whether the student held the door open more often for the White than for the Black student.
In a follow-up experiment, they had actors who appeared to be either White, Asian, or Muslim ask for directions (discrimination outcome: Did they receive them?). In yet additional studies, they had actors who appeared White or Muslim, or White or Asian drop cards in an elevator (discrimination outcome: Did they receive help?). In yet another, they had a woman either wearing or not wearing a hijab (Muslim headgear) sit on a bus (discrimination outcome: Did people sit next to her). In yet an additional pair of studies, they submitted student resumes for on-campus jobs; in one case the students had either white or Arab-sounding names; in another, they had either White or Black-sounding names (discrimination outcomes: Did they receive a response or interview?).
Across this entire set of studies, discrimination ranged from 5 to 20 percent, which is the same as saying 80 to 95 percent of responses did not discriminate. Again, this disconfirms hypothesis 1: There is no discrimination. It also disconfirms hypothesis 2: There is massive discrimination. Instead, it supports hypothesis 3: There is some racial discrimination but it is the relatively rare exception.
Publication #3: 9.3 Percent Discriminate
In the Journal of Experimental Psychology: Applied, Notveldt et. al. examined discrimination in the selection of Airbnb listings. The host was either identified as ethnically Norwegian or ethnically Somali. Among a nationally representative sample of 801 Norwegians, there was a 9.3 percent preference for the listing by the Norwegian ethnic. This is the same as saying 90.7 percent of Norwegians in the sample did not discriminate.
The results were actually a bit more complicated, in that discrimination was as high as 15.9 percent for listings that had mediocre ratings, and as low as 1.5 percent for listings that had excellent ratings.
Regardless, as with the other studies, these findings, overall:
- Disconfirm hypothesis 1: There is no discrimination.
- Disconfirm hypothesis 2: There is massive discrimination.
- Confirm hypothesis 3: There is some discrimination, but it is the exception rather than the rule.
![Lee Jussim Lee Jussim](https://cdn.psychologytoday.com/sites/default/files/styles/article-inline-half-caption/public/field_blog_entry_images/2021-06/img_20200229_111300.jpg?itok=kSt755Lu)
Conclusions
These papers are all published in very high-quality scientific peer-reviewed outlets. All found support for Hypothesis 3: Discrimination exists, but it is not massive; instead, it is the exception rather than the rule. This finding is remarkable, given the highly varied methods used to assess discrimination. Of course, none of these studies assessed how a history of poverty, oppression, or disadvantage influence modern life outcomes. Nonetheless, they should give anyone claiming that discrimination is powerful and pervasive, an endemic feature of modern life that infects all or most interactions, serious pause.
It also may help explain why diversity and implicit bias trainings are not as effective as many proponents suggest: They involve subjecting massive numbers of people to “education” that 80 to 90 percent, or more, do not need.
References
Campbell, M. R. & Brauer, M. (2021). Is discrimination widespread? Testing assumptions about bias on a university campus. Journal of Experimental Psychology: General, 150(4), 756–777. https://doi.org/10.1037/xge0000983
Nodtveldt, K. B., Sjastad, H., Skard, S. R., Thorbjornsen, H., & Van Bavel, J. J. (2021). Racial bias in the sharing economy and the role of trust and self-congruence. Journal of Experimental Psychology: Applied. Advance online publication. https://doi.org/10.1037/xap0000355
Peyton, K. & Huber, G. A. (2021). Racial resentment, prejudice, and Discrimination. Journal of Politics. https://doi.org/10.1086/711558