Almost all human beings wonder, at one time or another, whether they will be prejudicially misjudged based on some readily observable aspect of themselves – based on a group identity. The identity could be predicated on skin colour, ethnicity, gender, nationality, sexual preference, religion, height, weight – and we could be looked upon with disfavour based on a stereotype.
This concern can create anxiety and stress, and may lead to heightened feelings of persecution, failure of performance, and even the abandonment of dreams.[1]It is a threat based on negative stereotypes, and this stereotype threat can carry a harmful weight. In the modern-day availability cascade of identity politics, social justice, and diversity, equity and inclusion (DEI) discussions, stereotype threat can feel all-consuming. This pervasive identity malaise can act to further divide society.
In this article, we will break down the problem, discuss how to alleviate the problem, and warn about behaviour and actions that can make the issue worse. Our main goal is to help ease social divides, particularly regarding performance in careers, hobbies, or athletics, primarily through advice/coaching and, secondarily, through commentary on social policy.
What is stereotype threat?
The term stereotype threat, coined by psychologists in the 1990s, was defined by originator Claude Steele as a “Socially premised psychological threat that arises when one is in a situation or doing something for which a negative stereotype about one’s group applies.”[2]Steele’s definition touches on several interesting aspects of the concept that have persisted in studies.
First, this is a threat based on a feeling within persons that a negative stereotype is going to be applied to them. The threat requires that such a stereotype exists and is well known within society, but it does not require that the stereotype is being applied to those experiencing the threat. Stereotype threat is about a perception – of being targeted with a stereotype – and the effect that perception has on a person.
The first work on stereotype threat involved a study on African American college students, which we will examine in some detail later. Of course, the threat can apply to any group. White males, men, female engineers, women, ethnic groups – anyone can be part of the threatened group.
However, the threat most often applies when a group is in a minority situation.[3] [4] Consider female engineers in a male-dominated engineering program, and male teachers in female-dominated primary school teaching. Both groups have been shown to feel anxiety from stereotypes – either group can become the out-group depending on the situation.[5]
This social anxiety appears to create or increase the tendency for a wide variety of negative outcomes with groups experiencing it. These include stress, poorer than expected work or test performance, cognitive overload affecting working memory, the increase in the feeling of a negative bias against the group, and the potential disengagement or even abandonment of the activity altogether.[6] [7] [8] [9] [10]The poor outcomes are often referred to as the manifestation of a self-fulfilling prophecy, where anxiety and other responses to the threat lead to behaviours and feelings within the threatened group that makes the threat perpetuate the stereotype.[11]
Figure 1 summarizes key features of stereotype threat. We will next consider the general nature of stereotypes within society, followed by a discussion of the evidence supporting stereotype threat.
Figure 1 – Summary definition of stereotype threat and its effects.[12] [13] [14] [15] [16]
All humans generalize
Stereotypes are a kind of generalization; in this context an overgeneralization based on the application of assumptions about a person based on some group to which they belong.[17]Some readers may be tempted to believe that they do not engage in stereotyping, that this is an odious activity only undertaken by others. However, generalizing and stereotyping is part of the human condition. We all do it. Kahneman would describe our tendency to generalize as part of the fast-thinking part of our minds.[18]We spoke about decision making and nudging being related to fast thinking in Nudge and the uneasy pursuit of context and objectivity.
In this model, fast thinking and generalizations are a requirement for survival in daily life. These are the small, unremembered thoughts and decisions that allow us to cross a road safely, spot danger in a back alley or social setting, or to catch a baseball. We generalize as part of a categorization and learning process about the world.[19]In this sense, generalizing is akin to the transfer of previous observations and learning on new situations used in analogic reasoning. [20]This type of learning is widely used, including in the geosciences for carbon storage and skill transfer, and through race, gender, and age classifications used in medical studies, where it is driven by utility.[21] [22]
In our article on the Robbers Cave experiment and DEI (The Robbers Cave experiment’s important lesson for DEI), we explored how competition brought about a dark and pejorative set of stereotypes amongst a group of boys. This experiment showed how each group (the in-group) assigned attributes to the other group (the out-group) as they competed in various tasks at a summer camp. Negative stereotypes were assigned even though the two groups were essentially homogeneous. Robbers cave illustrated the propensity for unfounded negative generalizations when groups are in conflict. These stereotypes were not resolved until the groups were brought from competition to co-operation.[23]As limited and arguably flawed an experiment as Robbers Cave was, it warned about the poor social outcomes from conflict between groups.
What does the evidence look like?
Much of the evidence for stereotype threat comes through studies of academic, athletic, and work performance. We will reproduce and evaluate data from a representative and well-known set of experiments conducted by Cohen at al., at Stanford University.[24]
These experiments are concerned with the scholastic achievement of African American students and were conducted under a set of different mentoring conditions from the teaching team. This is a typical stereotype threat experiment in that it involves academic, medical, or athletics testing, and includes a teacher or coach role that unobtrusively provides a means to vary the test conditions. The negative stereotype being treated is the threat that African Americans are less intelligent and academically capable than the rest of the student body. The stereotype is certainly a damaging one and affects other minorities. We recently discussed a related issue in the article Social justice gone awry – morbid ideas that destroy us , where university testing standards were being lowered in association with minority enrollment quotas in a Canadian medical school.
The Stanford study involved 45 African American and 48 white students, each assigned with the task of writing a letter for potential publication in a technical journal.[25]The experimental designers contrived to make it apparent that the student’s skin colour would be known to the review board for the submitted letters through the inclusion of a picture of each student that would accompany each submission. The photo was to accompany the publication – which is sometimes done in journals. The students were also led to believe that the review board would be white.
The test introduced three feedback conditions, which – in the theoretical framework of stereotype threat – would provoke or potentially ease the level of threat to the students.
The first condition was that of criticism of the papers. This was applied consistently to all students, giving each student the same number of negative comments, stated in a rigorously similar manner. One-third of each group received only this criticism.
The second condition added a few general, positive comments about the work in addition to the criticism. In the experiment, this was termed as a positive buffer and was applied with rigorous similarity in the group that received it.
The third condition received the same criticism, but also included a discussion on the high standards of the journal to explain the reason for the criticism, together with the assurance that the review board were confident that the student could achieve those standards.
The experimental findings would be determined through statistical analysis of questionnaires given to the students grouped in all three feedback conditions.
The experimental hypotheses were the following:
- The African American students would feel that the criticism was biased against them, due to the stereotype.
- That the white students would be less affected by the criticism as they would not experience the stereotype threat.
- That the inclusion of assurance and an appeal to a reason for the criticism (standards) would ease the difference in perceived threat between the African American and white students.
Figures 2, 3, and 4 illustrate the results, adapted directly from the study.[26]These results support the hypotheses. Figure 2 illustrates the perceived bias, which is a direct proxy for perceived level of threat. Higher values of bias indicate higher threat levels, which are notably reduced in the third feedback condition, where assurances and an explanation of the high standards are given.
Figure 2 – Perceived bias as a function of feedback condition and race. Adapted from [27]. Higher bias ratings indicate elevated threat levels.
Figure 3 gives the same analysis regarding the students’ motivation levels to revise their letters. The motivation of the African American students was highly negative when the group was given unbuffered criticism, but somewhat exceeded that of the white students when both the assurance and appeal to high standards were provided.
Figure 3 – Task motivation as a function of feedback condition and race. Adapted from [28]. Higher motivation ratings indicate relaxing threat levels.
Figure 4 explores a more wide-ranging attitude: whether the student felt positively disposed to similar subjects and learnings after each control test. This has been interpreted in the light of later experimentation to be related to a feeling of belongingness with the subject matter and place in this subculture of society.[29]As in the previous results, the African American students reacted the most negatively to the unbuffered criticism, but became more positive with the additions of assurance and a rational appeal to the high standards of the journal. This attitude supports the notion that by easing this anxiety, potentially threatened groups are less likely to disengage from or abandon the activities.
Figure 4 – Belongingness as a function of feedback condition and race. Adapted from [30]. Higher belongingness ratings indicate relaxing threat levels.
A similar set of experiments was undertaken with female engineers in the male-dominated Faculty of Engineering at the University of Waterloo.[31]The feedback conditions, associated solutions, and results were found to be very similar to the previous study, with two remarkable observations: first that increasing grade point average (GPA) outcomes were proportional to easing threat levels and, secondly, a reinforcement of the notion that the women’s sense of social belonging was affected negatively by the threat.[32]
We discussed women in science, technology, engineering and mathematics (STEM) in Examining the gender lens and the Women in Cleantech Challenge . In that article, we noted that women graduating with STEM degrees have a much lower career uptake in STEM compared to their male counterparts at 23% versus 42%, respectively. This is congruent with our discussion of stereotype threat and feelings of belonging or social acceptance.
Other similarly designed experiments have been performed on different at-threat groups, largely with similar results.[33]
The strength of this evidence is limited. The STEM data illustrates a somewhat troubling outcome, but it does not directly prove that more women leave STEM than men solely because of stereotype threat. There are certainly other forces at work. Similar arguments could be – and have been – made against the stereotype threat experiments conducted at Stanford and other related tests. Cause and effect are difficult to pin down, different subjects experience the threat differently, and the experiments themselves may be difficult to apply without interfering with the results.[34]
Ultimately, the evidence is statistical, somewhat uncertain, but strong enough to take seriously.
How to reduce or eliminate stereotype threat
The experiments on stereotype threat considered ways of easing the problem – an action psychologists call intervention. These interventions revolve primarily around assurances, which include:[35]
- Coaching the threatened group to view the circumstances as non-threatening. This is a reclassification of the situation.
- Coach the threatened group in management techniques to reduce the cognitive load and underperformance mechanisms.
- Alter the circumstances so that they seem less threatening.
The assurances and explanations behind the standards that were used in the Stanford experiment we discussed followed the first intervention. The second intervention – management techniques – consists of mindfulness and self-affirmation techniques. The last intervention type involves making the circumstances seem safer using positive role models from the minority group or meeting and mixing with the majority group.[36]
How to make the problem worse
Exacerbating the feelings of stereotype threat largely involves doing the opposite of the three interventions that we use to reduce the threat.
Affirming and reinforcing the stereotype is clearly the wrong thing to do. In the Stanford study and similar experiments, things were done to bring the stereotype threat into play. Pictures of the students were taken, making the subjects aware that their skin colour would be known and could be the subject of bias. The review board was alluded to as white. These may be arguably subtle things, but people experiencing stereotype threat become aware of such cues.[37] [38]
In our article Social justice gone awry – morbid ideas that destroy us , we described how one of critical race theory’s (CRT) fundamental tenets is that society is racist. This is further followed by the argument that this racism is pervasive, deep, and cannot be undone without drastic, unequal measures. We interpreted this thinking as counterproductive and morbid, and argue that these tenets of CRT would increase threat levels for individuals of all groups. It seems the CRT approach is the opposite of what should be done to ease social unrest premised from the perception of stereotypes. This is escalation rather than de-escalation.
Clearly, ignoring stereotypes is not a solution, either. These issues, while unfortunately natural – `should be discussed and minimized. In our article The Robbers Cave experiment’s important lesson for DEI, we explored the different ways in which diversity can be approached. It was found that approaches that were open, positive, and affirming were effective in reducing feelings of stereotypes for all groups, whereas approaches that were prescriptive, suppressive, and negative resulted in increasing feelings of stereotyping. A heavy-handed approach that creates villains and victims – and focuses on shame and feelings of hopelessness – is harmful.
We want our social policies to reduce stereotyping and the feelings of threat from stereotypes, but the approach should be subtle, affirming, and result in a de-escalation of the anxiety. Policies focusing on identity must be crafted carefully and in a way that does not increase or perpetuate the threat, and does not foster division.
Returning to the discussion of a Canadian medical school reducing standards and introducing quotas to minority applicants in Social justice gone awry – morbid ideas that destroy us , we see that the policy had four problems. First, the approach is unequal and likely to prime social division. Second, by lowering standards it affirms the stereotype that the minority students are inferior. This is damaging to the minority students and reinforces not just the threat of the stereotype, but the stereotype itself. The third problem is that it is unnecessary. Properly coached students of all backgrounds can rise to meet high standards, as shown in the Stanford experiment. Rather than introducing such poorly thought policy, our efforts would be better invested in providing positive coaching and teaching. The fourth problem with this approach is that it erodes the quality of the medical school and society’s faith in our medical institutions.
Rolling up the threat
Stereotype threat is a damaging social anxiety. It is the weight of negative expectations, and it can fall on any group, depending on circumstances. The impact of stereotype threat may be statistical and challenging to measure, but it is certainly negative in effect to productivity, happiness, and social harmony.
We have seen that ignoring the threat – paying no attention to negative stereotypes that exist about all groups in society – does not work. This strategy is equivalent to the unbuffered criticism group in the psychological experiments we discussed, and those outcomes were poor.
At the same time, triggering or feeding the stereotypes is unhelpful. We interpret the stereotype-priming elements of identity politics and CRT to fall into this unhelpful category. Divisive movements such as CRT highlight and exacerbate the differences between groups. This is a formula for growing a threat, not reducing it, and worse, creating greater social isolation between groups.
The experiments themselves have pointed to numerous more positive approaches to the problem. Interestingly, there is no experimental evidence suggesting that lowering standards is helpful, but rather that better coaching to threatened groups is crucial. The best results come from de-escalating the threat, and assuring groups that they can succeed, and that tests are there through necessity, not punitive action, through identifying in-group role models and fostering co-operation. Figure 5 summarizes these findings.
Readers can judge for themselves whether the social policies within their jurisdictions are likely to help their citizens overcome this psychological threat.
Figure 5 – Summary observations of factors that may increase or reduce stereotype threat in work or athletic performance, and our interpretation of the effect of certain social policy interventions.
References
[1] Steele C., and J. Aronson, 1995, Stereotype threat and the intellectual test performance of African Americans. J Pers Soc Psychol. 69(5), 797-811
[2] Steele C., and J. Aronson, 1995, Stereotype threat and the intellectual test performance of African Americans. J Pers Soc Psychol. 69(5), 797-811
[3] Steele C., and J. Aronson, 1995, Stereotype threat and the intellectual test performance of African Americans. J Pers Soc Psychol. 69(5), 797-811
[4] Brooks, J., 2022, Defining Stereotype Threat and Why It Matters, JPOSNA, 5(S1), 1-5
[5] Kalokerinos, E., K. Kjelsaas, S. Bennetts, and C. von Hippel, 2017, Men in pink collars: Stereotype threat and disengagement among male teachers and child protection workers. European Journal of Social Psychology, 47(5), 553-565
[6] Steele C., and J. Aronson, 1995, Stereotype threat and the intellectual test performance of African Americans. J Pers Soc Psychol. 69(5), 797-811
[7] Brooks, J., 2022, Defining Stereotype Threat and Why It Matters, JPOSNA, 5(S1), 1-5
[8] Kalokerinos, E., K. Kjelsaas, S. Bennetts, and C. von Hippel, 2017, Men in pink collars: Stereotype threat and disengagement among male teachers and child protection workers. European Journal of Social Psychology, 47(5), 553-565
[9] Spencer, S., C. Logel, and P. Davies, 2016, Stereotype Threat, Annu Rev. Psychol., 67, 415-437
[10] Drew, C., July 10, 2023, Stereotype Threat: 15 examples, Definition, Criticisms, HelpfulProfessor.com
[11] Brooks, J., 2022, Defining Stereotype Threat and Why It Matters, JPOSNA, 5(S1), 1-5
[12] Steele C., and J. Aronson, 1995, Stereotype threat and the intellectual test performance of African Americans. J Pers Soc Psychol. 69(5), 797-811
[13] Brooks, J., 2022, Defining Stereotype Threat and Why It Matters, JPOSNA, 5(S1), 1-5
[14] Kalokerinos, E., K. Kjelsaas, S. Bennetts, and C. von Hippel, 2017, Men in pink collars: Stereotype threat and disengagement among male teachers and child protection workers. European Journal of Social Psychology, 47(5), 553-565
[15] Spencer, S., C. Logel, and P. Davies, 2016, Stereotype Threat, Annu Rev. Psychol., 67, 415-437
[16] Drew, C., July 10, 2023, Stereotype Threat: 15 examples, Definition, Criticisms, HelpfulProfessor.com
[17] Raskhoff, S., May 14, 2012, Understanding Generalization and Stereotypes, Everyday Sociology Blog
[18] Kahneman, D., 2011, Thinking, Fast and Slow: Farrar, Straus and Giroux
[19] Taylor, J., A. Cortese, H. Barron, X. Pan, M. Sakagami, and D. Zeithamova, How do we generalize? Neuron Behav. Data Anal. Theory, 1, 1-39
[20] Hunt, L., E. Street, and G. Hack, 2023, CCS expertise and analogic reasoning, CSEG RECORDER, 48(9), 1-21
[21] Hunt, L., E. Street, and G. Hack, 2023, CCS expertise and analogic reasoning, CSEG RECORDER, 48(9), 1-21
[22] Wasserman, D., 2010, The justifiability of racial classification and generalizations in contemporary clinical and research practice, Law, Probability & Risk, 9(3-4), 215-226
[23] Sherif, M. 1956, Experiments in Group Conflict, Scientific American, 195 (5), 54-59
[24] Cohen, G., C. Steele, and L. Ross, 2012, The Mentor’s Dilemma: Providing Critical Feedback Across the Racial Divide, Personality and Social Psychology Bulletin, 25(10), 1302-1318
[25] Cohen, G., C. Steele, and L. Ross, 2012, The Mentor’s Dilemma: Providing Critical Feedback Across the Racial Divide, Personality and Social Psychology Bulletin, 25(10), 1302-1318
[26] Cohen, G., C. Steele, and L. Ross, 2012, The Mentor’s Dilemma: Providing Critical Feedback Across the Racial Divide, Personality and Social Psychology Bulletin, 25(10), 1302-1318
[27] Cohen, G., C. Steele, and L. Ross, 2012, The Mentor’s Dilemma: Providing Critical Feedback Across the Racial Divide, Personality and Social Psychology Bulletin, 25(10), 1302-1318
[28] Cohen, G., C. Steele, and L. Ross, 2012, The Mentor’s Dilemma: Providing Critical Feedback Across the Racial Divide, Personality and Social Psychology Bulletin, 25(10), 1302-1318
[29] Walton, G., C. Logel, J. Peach, S. Spencer, and M. Zanna, 2014, Two Brief Interventions to Mitigate a “Chilly Climate” Transform Women’s Experience, Relationships, and Achievement in Engineering, American Psychological Association, 107(2), 468-485
[30] Cohen, G., C. Steele, and L. Ross, 2012, The Mentor’s Dilemma: Providing Critical Feedback Across the Racial Divide, Personality and Social Psychology Bulletin, 25(10), 1302-1318
[31] Walton, G., C. Logel, J. Peach, S. Spencer, and M. Zanna, 2014, Two Brief Interventions to Mitigate a “Chilly Climate” Transform Women’s Experience, Relationships, and Achievement in Engineering, American Psychological Association, 107(2), 468-485
[32] Walton, G., C. Logel, J. Peach, S. Spencer, and M. Zanna, 2014, Two Brief Interventions to Mitigate a “Chilly Climate” Transform Women’s Experience, Relationships, and Achievement in Engineering, American Psychological Association, 107(2), 468-485
[33] Pennington, C., D. Heim, A. Levy, and D. Larkin, 2016, Twenty Years of Stereotype Threat Research: A Review of Psychological Mediators. Plos One, 11(1)
[34] Pennington, C., D. Heim, A. Levy, and D. Larkin, 2016, Twenty Years of Stereotype Threat Research: A Review of Psychological Mediators. Plos One, 11(1)
[35] Spencer, S., C. Logel, and P. Davies, 2016, Stereotype Threat, Annu Rev. Psychol., 67, 415-437
[36] Spencer, S., C. Logel, and P. Davies, 2016, Stereotype Threat, Annu Rev. Psychol., 67, 415-437
[37] Spencer, S., C. Logel, and P. Davies, 2016, Stereotype Threat, Annu Rev. Psychol., 67, 415-437
[38] Aronson, J., D. Burgess, S. Phelan, and L Juarez, 2013, Unhealthy Interactions: The Role of Stereotype Threat in Health Disparities, American Journal of Public Health, 103(1), 50-56
