They Don’t See a Lot of People My Color: A Mixed Methods Study of Racial/Ethnic Stereotype Threat Among Medical Students on Core Clerkships

Abstract

Purpose 

Stereotype threat is an important psychological phenomenon in which fear of fulfilling negative stereotypes about one’s group impairs performance. The effects of stereotype threat in medical education are poorly characterized. This study examined the prevalence of racial/ethnic stereotype threat amongst fourth-year medical students and explored its impact on students’ clinical experience.

Method 

This was an explanatory sequential mixed methods study at 2 institutions in 2019. First, the authors administered the quantitative Stereotype Vulnerability Scale (SVS) to fourth-year medical students. The authors then conducted semistructured interviews among a purposive sample of students with high SVS scores, using a qualitative phenomenographic approach to analyze experiences of stereotype threat. The research team considered reflexivity through group discussion and journaling.

Results 

Overall, 52% (184/353) of students responded to the survey. Collectively, 28% of students had high vulnerability to stereotype threat: 82% of Black, 45% of Asian, 43% of Latinx, and 4% of White students. Eighteen students participated in interviews. Stereotype threat was a dynamic, 3-stage process triggered when students experienced the workplace through the colored lens of race/ethnicity by standing out, reliving past experiences, and witnessing microaggressions. Next, students engaged in internal dialogue to navigate racially charged events and workplace power dynamics. These efforts depleted cognitive resources and interfered with learning. Finally, students responded and coped to withstand threats. Immediate and deferred interventions from allies reduced stereotype threat.

Conclusions 

Stereotype threat is common, particularly among non-White students, and interferes with learning. Increased minority representation and developing evidence-based strategies for allyship around microaggressions could mitigate effects of stereotype threat.

An equitable learning environment provides all students with opportunities to learn, demonstrate their learning, and succeed.¹ Students from racial/ethnic groups underrepresented in medicine (UIM) face inequities because they must simultaneously confront pressures universal to medical students—improving patient care skills and medical knowledge, adapting to new clinical settings, and deciphering spoken and unspoken expectations—while also navigating UIM-specific pressures.^2,3 UIM learners (African American, Latinx, and Native) and other nonmajority racial/ethnic groups may face additional pressures that disproportionately and negatively affect their performance, including supervisor biases, possibly poorer prior academic training, and stereotype threat.^3–9 Consequently, UIM students as a group receive lower performance ratings and grades on clerkships compared with non-UIM peers.^3,10,11 Similarly, non-White students receive less favorable Medical Student Performance Evaluation summary descriptors and lower clerkship grades than White students.^11,12 To create more equitable learning environments, it is important to understand the factors which disproportionately affect some students. Despite evidence that some medical students experience racial/ethnic stereotype threat, the ways in which stereotype threat may contribute to inequities has not been well explored.^8,13,14

Stereotype threat is a psychological phenomenon in which members of negatively stereotyped groups worry that they will conform to those stereotypes, a fear which impairs performance.^15–17 Impairment from stereotype threat arises as awareness of the stereotype influences learners’ affective, cognitive, and motivational states.^18,19 Stereotype threat adversely affects vulnerable individuals across contexts: it has been implicated in the underperformance of women in mathematics, African American and Latinx students on tests of intellectual ability, and the elderly undergoing cognitive tests.^15,20,21 Even subtle context-dependent cues around race/ethnicity can trigger stereotype threat and impair performance.²⁰ For instance, when told they would be compared with Asian men, White men underperformed on a math examination.²² Because stereotype threat is context dependent, it is important to understand which aspects of the clinical learning environment trigger stereotype threat in students.

Previous work described how African American and first-generation health professions students’ perceptions of racial stereotypes triggered negative emotions, but the implications for their learning and performance were not elaborated.¹³ In a recent multi-institutional survey of over 600 fourth-year medical students, UIM students had higher rates of racial/ethnic stereotype threat than non-UIM students (55.7% versus 10.9%). Students with higher stereotype threat earned fewer honors grades.⁸ This study did not detail rates or experiences of stereotype threat by specific race/ethnicity. Therefore, we designed this study to (1) determine the prevalence of racial stereotype threat stratified by medical student race/ethnicity and (2) explore student experiences of stereotype threat during clinical rotations. This information can inform efforts to optimize learning experiences for diverse students.

Method

Design

This mixed methods study employed an explanatory sequential design (quantitative survey followed by qualitative interviews) at 2 institutions in March through May 2019.²³ In Phase 1, we administered the quantitative Stereotype Vulnerability Scale (SVS) to assess the prevalence of racial/ethnic stereotype threat among medical students.²⁰ In Phase 2, viewing from an interpretivist ontological perspective, we used a qualitative phenomenographic approach to interview students with high SVS scores to explore their experiences around stereotype threat.²⁴ This methodology enabled us to understand quantitative survey results through individual interviews, explore why students scored high on this measure of vulnerability to stereotype threat, and probe the breadth of personal experiences to understand the phenomenon of stereotype threat in the medical training context.²³

This study was approved by the institutional review boards at the University of California, San Francisco (UCSF) and the University of Colorado Anschutz Medical Campus.

Setting and participants

Study institutions were UCSF and University of Colorado Schools of Medicine (CU), both large public institutions. In 2018, the UCSF student body was 33% White, 27% Asian, and 27% UIM; faculty were 59% White, 28% Asian, and 9% UIM (4% unknown).²⁵ The CU incoming class of 2018 was 52% White and 28% UIM (20% not reported); faculty were 65% White, 9% Asian, and 7% UIM (19% multiracial or unknown).²⁶ Eligible participants were all fourth-year medical students at both institutions.

Phase 1: Quantitative survey

Fourth-year students received individualized email invitations from the Qualtrics survey platform to complete a 12-item survey. Nonrespondents received up to 3 weekly reminders. Consistent with our previous study, we used an adapted, 5-item SVS tool to assess perceptions of stereotypes about one’s race/ethnicity on clerkships (Table 1).^8,20 We eliminated 3 original SVS items due to double-negative wording which confused students.⁸ Seven questions addressed demographic characteristics: gender (2), race/ethnicity using U.S. census categorizations (2), age (1), first-generation college status (1), and medical school (1).

Table 1: Adapted Stereotype Vulnerability Scale^a

We calculated descriptive statistics for all demographics. SVS item responses ranged from 1 to 5 (1 = strongly disagree, 3 = neither agree/disagree, 5 = strongly agree). One SVS item was reverse coded, so that a higher score meant more threat. The summed values from the 5 items created the SVS score (5–25). We dichotomized SVS scores with > 15 representing high student vulnerability to stereotype threat (“high threat”) and ≤ 15 representing low vulnerability (“low threat”).⁸ Quantitative data were analyzed using STATA 15.1 (Stata Corp, College Station, Texas).

Phase 2: Individual interviews

Individual in-depth interviews allowed us to explore students’ feelings and experiences with safety to discuss sensitive, racially/ethnically charged content.¹³ Students endorsing high threat and who provided their email address were eligible for interviews. We invited eligible students as they completed the survey. Students received up to 3 email reminders over 2 weeks.

Two investigators (J.L.B., K.E.H.) developed a semistructured interview guide based on the literature on stereotype threat.^16–18,20 J.L.B. conducted 2 pilot interviews to refine the interview guide for clarity and completeness. An inherent tension exists in categorizing students using race/ethnicity terminology due to variable interpretations.²⁷ Students may share the same racial/ethnic identity but differ in skin color or language(s) spoken. Because we were interested in self-perceptions, the interviewer solicited each participant’s self-reported race/ethnicity at the start of the interview and then used that terminology throughout the interview. Interview questions explored students’ consciousness of their race/ethnicity during clerkships, how they perceived stereotypes affecting their performance evaluations, and how they perceived their race affected their clerkship experience and performance (see Supplemental Digital Appendix 1 at https://links.lww.com/ACADMED/A999). Three trained African American investigators (J.L.B., R.R., A.d.P-.J.) interviewed students either in-person or via videoconference platform. Interviewees received a $20 electronic gift card. We continued interviews until the research team identified sufficiency with respect to a diverse sample of respondents and variability in experience of stereotype threat.^24,28

We audio-recorded and transcribed all interviews verbatim and deidentified transcripts before analysis. Our analysis used phenomenography, a qualitative methodology which allows researchers to characterize the multiple ways that individuals understand and experience a phenomenon.^24,29 Phenomenography focuses on how these variations in experience relate to one another and involves 7 steps of analysis: familiarization, condensation, comparison, grouping, articulating, labeling, and contrasting.^24,29

Two investigators (J.L.B., K.E.H.) reviewed 6 transcripts to develop a codebook by first generating preliminary codes and then met to discuss and refine codes. Each coding team member (J.L.B., T.L., A.d.P.-J., A.T., K.E.H.) used the draft codebook to code a new transcript individually. After discussion, J.L.B. and K.E.H. again revised the codebook for clarity and to minimize redundancy. J.L.B. read all transcripts before finalizing the codebook (familiarization). Two investigators coded each transcript; J.L.B. coded all transcripts (condensation). We reconciled differences in coding through discussion. Interviews were coded using Dedoose Version 8.0.35 (Dedoose, Los Angeles, California).

Consistent with a phenomenographic approach, we conducted comparison, grouping, articulating, and labeling steps iteratively until the entire research team felt that we sufficiently captured the essence in variation of stereotype threat from our interviews. The contrasting step occurred as we synthesized data into themes (below).

We employed several strategies to ensure trustworthiness. Considering reflexivity, the study team included 1 man and 5 women of diverse professional roles; 3 investigators identified as African American, 2 White, and 1 Middle Eastern.³⁰ The coding investigators kept reflexivity journals to record their thoughts, potential biases, and emotions prompted by the data; they discussed these reactions with the full study team.³¹ After analysis, we emailed synthesized interview results to all interview participants for member checking.³² Eight participants (44% of interviewees) gave feedback via phone or email. All respondents said that the results accurately reflected their experience. One participant recommended increasing the number of quotations related to a particular topic and one corrected the role of a supervisor whom we had mischaracterized.

Results

Phase 1: Quantitative survey assessing stereotype threat vulnerability

Overall, 52% (184/353) of students responded to the survey (Table 2). Respondents’ average age was 26.9 (SD = 2.6); 52% (96/184) were women and 26% (48/184) UIM. Demographics and means by school are listed in Supplemental Digital Appendix 2 (at https://links.lww.com/ACADMED/A999). Collectively, 28% of respondents had high vulnerability to stereotype threat; 82% of Black, 45% of Asian, 43% of Latinx, and 4% of White students. On average, Black students scored 17.6 (SD = 2.6) on the SVS, Asians 14.9 (3.3), Latinx 14.3 (4.9), and Whites 8.8 (3.4; Table 2). The mean SVS score for all survey respondents with high threat was 18.3 (1.9).

Table 2: Demographic Data for Survey Respondents (n = 353) and Mean Stereotype Vulnerability Scale Scores by Group for Fourth-Year Medical Student Survey Respondents at 2 Medical Schools in 2019^{8 , 20}

Phase 2: Qualitative interviews exploring the meaning of stereotype threat vulnerability

We invited 24 students and interviewed all 18 who responded to interview invitations. Interviewees had an average SVS score of 18.6 (SD = 1.7); 11 (61%) were women. Nine self-identified as Black, 2 Latinx, 3 Middle Eastern, 3 Asian, and 1 multiracial. Interviews lasted an average of 40 minutes (range, 29–50 minutes). Interview findings were consistent across institutions.

Interviewees described stereotype threat as a dynamic process influenced by their internal and external environment, rather than a static fear of stereotypes. We developed the Clerkship Student Stereotype Threat Model which characterizes the phenomenon of stereotype threat into 3 stages (Figure 1). (1) Triggering: participants described how standing out because of their race/ethnicity, previous experiences, or microaggressions caused them to experience the workplace through a colored lens of race/ethnicity, triggering stereotype threat. (2) Internal dialogue: students spent substantial energy processing these triggering events. Their internal dialogue around how to navigate racially/ethnically charged events and power dynamics in the environment interfered with clinical learning. (3) Response: students described how they responded and coped to withstand threats during their clerkships. While responses to these experiences varied, students rarely confronted triggers directly. At times, patients and providers served as allies and, through their actions, helped to decrease students’ stereotype threat. We describe findings below with participant number and race/ethnicity in parentheses. We conclude with findings regarding allies.

Figure 1: Clerkship Student Stereotype Threat Model: Stereotype threat was a 3-stage process in which a student was first triggered by race, next spent substantial cognitive resources to understand and then decide how to best respond to the trigger, and finally responded to that trigger. Interventions from allies helped to mitigate stereotype threat.

A colored lens.

Race/ethnicity was omnipresent in day-to-day activity of participants and served as a colored lens through which students viewed their experiences and others seemed to view them. Students were continually prompted to consider their race/ethnicity because they stood out due to the absence of others who looked like them, recalling their past experiences, intersectionality of race/ethnicity with other identities, and frequent microaggressions.

From a clerkship’s start, students received cues from the learning environment that they stood out, cues which triggered feelings of stereotype threat. One staff told a student: “They don’t see a lot of people my color in this area” (18-Black). One student described a patient fixating on her identity by asking, “‘What kind of a name is that? Where are you from?’ I was like, ‘Well, I grew up here.’ ‘No, but what are you?’” (11-Middle Eastern). Students felt this lack of representation negatively affecting them: “If I notice that I’m one of the few people of my race or ethnicity … I start feeling self-conscious about myself and how I present to others” (3-Asian). For some students, the salience of race and ethnicity engendered an internal pressure to represent their group well. They felt compelled to work harder than other students and be more thorough. Students felt burdened to not only showcase their capabilities but also combat others’ biases. Students acknowledged that their experiences around race/ethnicity differed based on their skin tone. While students with lighter skin avoided some race-based interactions, they still experienced substantial racial/ethnic stereotype threat.

Past experiences around race/ethnicity could trigger stereotype threat. One student described how he suffered from stereotype threat despite an overall pleasant clerkship experience because he previously had negative educational experiences relating to his ethnicity: “You’re getting in because of affirmative action … I’ve internalized those stereotypes of intellect…. But fortunately in medical school I haven’t had those experiences be brought up in a threatening or demeaning way” (10-Latinx). Despite this rationalization, the student continued to believe that others thought less of him as a medical student.

The coexistence of race/ethnicity with other identities held by students highlighted the phenomenon of intersectionality: how distinct marginalized identities interact. Intersectionality of race/ethnicity with gender, class, and sexual orientation was most commonly mentioned. Most women participants described experiencing the intersectionality of race/ethnicity with gender. Recalling written feedback describing her as combative, one said, “I can only attribute it from the trope of an angry Black woman” (1-Black). Many participants revealed how even seemingly innocuous conversation about weekend plans could trigger feelings of misbelonging due to intersectionality of race/ethnicity and class: “Attendings talk about skiing, golfing, all these things … I just can’t relate to it, because I didn’t grow up doing any of that” (15-Black). Intersectionality prompted questions: “Which part of me are they responding to today?” (4-Asian).

Microaggressions.

Microaggressions featured prominently in the experience of stereotype threat and typically came from supervisors or patients. Microaggressions were ubiquitous and often subtle, in the context of complex and rapid interactions. Multiple students shared that some supervisors used feedback discussions to convey racist opinions. One resident’s feedback was that the student was not fitting into his team’s culture. The student interpreted that it was “White professional culture that [the resident] was talking about” (9-Asian). Another student interpreted an attending’s feedback about professionalism as coded language for needing to talk “less ethnic” (3-Asian). Students felt activated around race when patients made comments addressing student identity or tried to prevent minority trainees from caring for them. At times, racism felt salient to students, but they felt it was overlooked by others around them. Ultimately, as one participant acknowledged, “You’re dealing with all these passive-aggressive microaggressions that are just literally everywhere” (15-Black).

Sometimes, supervisors made egregious comments. Referring to a Black patient who had survived multiple complications, one student quoted a supervisor saying, “‘Our patient is like a cat, they must have 9 lives or something. Or no, maybe more like a cockroach.’…. And I don’t think it’s a far leap to say ‘if you view a patient that way. Like, what do you think of me?’” (8-Black). Students were negatively affected by vicarious threats while overhearing aggressions committed against other minority students. Referring to her Middle Eastern peer, one student said, “Even though I’m Black and there’s a whole bunch of perceptions around that, I’ve never been called the ‘n’ word, but she gets called a terrorist” (16-Black).

Internal dialogue.

Students engaged in extensive internal dialogue to interpret motivation behind offensive interactions. Even when offended by an interaction, many students doubted their emotional reaction: “There’s always that, like, was it a microaggression?…. Am I just being too sensitive? Am I just too tired?” (8-Black). Students pondered whether they were simply projecting their own feelings onto others: “Sometimes I wonder if they think I’m not as capable, or not as smart. I don’t know if that’s me projecting on myself too” (15-Black). Processing whether and how to respond, participants considering power dynamics, assessment and grading, and the personal emotional impact of experiences.

Power dynamics featured prominently as students perceived themselves as both minorities and students positioned at the bottom of the medical hierarchy. Hierarchy influenced students’ decision to respond to threats from supervisors and patients. Students often expected their supervisors to respond in their defense; this support usually did not manifest. Sometimes, supervisors responded unfavorably. One student recalled how a patient said to her, “‘I had these Iranian neighbors that, Lord knows, maybe they’re spies.’ And I remember my resident laughed at that comment” (12-Middle Eastern). Students observed many events triggered by, or witnessed by, attending physicians. Attendings’ elevated position in the hierarchy diminished students’ empowerment to respond. One decided whether to respond by assessing whether a threat arose “from a top person…. If I need something from them, then I’m hesitant to respond” (13-Black).

Students debated how to respond to microaggressions while simultaneously juggling pressures of assessment and grading, and their responsibilities for patient care and learning. After weighing the consequences of responding, many concluded that the most expeditious resolution was silence: “I just wanted to be quiet and get out” (18-Black). Students commonly refrained from responding to avoid jeopardizing their evaluations or grades: “I didn’t say anything. And I wish I had, but like at the time I was like, ‘Oh no, your career’s on the line’” (8-Black). Interviewees believed that threats detracted from their educational experience and performance by adding an extra load onto already taxing clerkships. One student explained their thought process:

It was so stressful to get these comments…. Do I want to give an answer that’s going to appease them and maintain a good relationship? … it’s a big loaded question when I’m trying to think of the differential for altered mental status. (2-Latinx)

Students felt disadvantaged on clerkships compared with White peers who did not have to think about these issues. One student compared his experience to “a White student who doesn’t have to constantly survey. And they can probably think about what they’re supposed to be thinking about” (13-Black).

The emotional impact of negative interactions became apparent as multiple students cried during study interviews. Students used charged words to describe their feelings: “shocked,” “dehumanized,” “bothered,” “isolated,” “powerless,” “sad,” “disrespected,” and “singled out.” Reflecting on a conversation in which some classmates implied that more Black men did not deserve to be in their medical school, “It kind of sometimes makes me have that imposter syndrome … [begins to cry] … the feelings of just ‘do I deserve to be here?’ were really profound” (15-Black). One participant described an attending mischaracterizing the culture of the student’s Middle Eastern country; when the student attempted to correct him, the attending persisted. “I couldn’t stop thinking about it even though … I kept telling myself it’s not a big deal” (14-Middle Eastern).

Responding and coping.

Students showed resilience using multiple coping mechanisms to mitigate stereotype threat. They often tried to work within the system to navigate microaggressions, while also staying true to themselves and their purpose. There was large heterogeneity in how students managed threats. Some who came from undergraduate institutions with predominantly White students felt armed with preexisting coping strategies. Participants used a variety of techniques, including avoidance, prevention, deferral, and confrontation to manage threats (Table 3).

Table 3: Students’ Response to a Perceived Threat: Strategies or Behaviors Used by Students to Respond to or Cope With an Event That Triggered Stereotype Threat as Described by Students in Qualitative Interviews at 2 Schools in 2019

When students sensed that minority patients were receiving poorer quality care than others, they described shouldering the burden to provide care surreptitiously or reassure patients independently. These efforts entailed spending more time with patients, speaking with them in their native language, or helping to coordinate appointments. Students felt empowered and inspired to improve patient care through these unique contributions. Some cited these interactions as the reason they went into medicine and a primary driver to persevere.

Allies.

When asked where they felt safe, some students shared instances in which supervisors responded in ways that reduced their stereotype threat. Some supervisors served as allies who created a safe environment by reducing the threat after a negative race-related incident. Table 4 shows examples of supervisor responses that students found effective: drawing the line for patients, reassuring patients, correcting misconceptions, creating a teaching moment, and reflecting afterward. Each of these techniques promoted students’ feelings of safety.

Table 4: Supervisors Creating a Safer Clinical Learning Environment Through Allyship: Scenarios Described by Students During Qualitative Interviews at 2 Schools in 2019 About Stereotype Threat in Which Supervisors Helped to Promote a Safe Learning Environment by Appropriately Intervening Upon a Threatening Situation Either During or After the Trigger

Minority providers and patients served as allies who provided strength and positivity for many participants to continue to push forward despite stereotype threat. Multiple students emphasized the importance of members of their race to increase the number of potential allies. “Residents that are of color, they get the struggle. They take that extra time to just see how you’re doing or give that extra hand of encouragement” (15-Black). Minority patients also affirmed students. One patient said to a student, “You got to keep doing this, you can’t fall off the path” (8-Black). Students felt their own resolve buoyed by allies’ support.

Discussion

Our study found that racial and ethnic stereotype threat is a widespread, dynamic, and consequential process for minority clerkship students. This study corroborates and expands upon previous work exploring stereotype threat among African American doctoral and health professions students.^13,33 Many racial and ethnic minority students, not just UIM students, suffer from racial/ethnic stereotype threat. Strikingly, over 80% of our Black respondents and almost half of Latinx and Asian respondents were highly vulnerable to stereotype threat. Middle Eastern students, considered White by U.S. census definitions and not specifically categorized in our quantitative survey, also emerged as highly vulnerable to stereotype threat.

Our interviews revealed the numerous and pervasive ways that stereotype threat negatively affects medical students’ learning experience and performance. Below, we use cognitive load theory (CLT) and critical race theory (CRT) to examine students’ experience with race/ethnicity and the impact of stereotype threat. Effective interventions from allies subjectively reduced stereotype threat among our participants.

CLT highlights how stereotype threat can impair student performance. According to CLT, learners have finite working memory capacity which must accommodate the intrinsic load (task difficulty), extraneous load (distractors which consume working memory but do not help accomplish the task), and germane load (working memory to process a challenging task).^34,35 Consistent with prior studies in fields outside medical education, our results suggest that stereotype threat depletes students’ working memory by increasing extraneous load.^17,36,37 Participants commented on their increased extraneous load due to stereotype threat—that is, thinking about race at the expense of clinical reasoning—and frequently felt disadvantaged because White students did not have to contemplate these same issues. This finding can contribute to the performance degradation previously observed with stereotype threat in clinical medical students.⁸ For some interviewees, increased cognitive load persisted long after the initial threat ended. Learning is optimized when trainees perceive psychological well-being; our trainees felt emotionally depleted by stereotype threat.³⁸

Students described many key tenants of CRT as they described the omnipresent, racially/ethnically colored lens that triggered stereotype threat.³⁹ CRT, a critical pedagogy, argues that racial disparities exist because society is fundamentally racist and is organized to perpetuate those disparities.⁴⁰ CRT views racism as a structural and endemic problem in education. CRT, and our students’ stories, describe how microaggressions, intersectionality, and Whiteness as property (e.g., White professional culture as the desired medical culture) perpetuate racial disparities in performance.⁴⁰ Microaggressions are important initiators of, but distinct from, stereotype threat. Any reminder of race/ethnicity can trigger stereotype threat.¹⁷ Racial/ethnic microaggressions by definition are directed to one’s racial/ethnic group and therefore cause racial/ethnic salience.⁴¹ However, standing out due to lack of representation and students’ previous life experiences also trigger stereotype threat without microaggressions. The interrelationship between microaggressions and stereotype threat highlights a need to explore stereotype threat among other student groups which commonly face microaggressions such as women and lesbian, gay, bisexual, transgender, and queer (LGBTQ) students.⁴²

CRT explores the way that power impacts educational disparities. This prompts the question of how medical educators can empower trainees to respond to triggers to mitigate their stereotype threat and also how they can be better allies.⁴⁰ Because the medical hierarchy weighed so heavily, students often avoided conflict in the face of a triggering event and relied on others to speak up. When allies intervened on a microaggression, students perceived less threat—they no longer had to question whether something was a microaggression or how to respond, nor did they continue to feel isolated. Allies decreased the salience of negative stereotypes and students’ pressure to disprove them. This finding highlights the need to move beyond simply identifying microaggressions: we must generate evidence-based solutions to respond to microaggressions and then train faculty and residents on how to do so.^43–45

This study has limitations. We used self-report data with a survey response rate of 52%: it is possible that students more affected by stereotype threat were more likely to complete the survey and therefore we may overestimate the prevalence of stereotype threat. Conversely, because stereotype threat can be unconscious, we may underestimate its prevalence.²⁰ This study was conducted at 2 medical schools, and results do not represent the experience of all medical students who experience racial/ethnic stereotype threat. It is unclear how well our results generalize or transfer to other schools.

Conclusions

This study highlights a prevalent and concerning phenomenon amongst minority medical students. Respondents shared many ways in which stereotype treat distracts from their clinical learning and also showcased their strength, perseverance, and coping skills. To mitigate the negative effects of stereotype threat, there is a critical need to increase minority representation at all levels of the medical pipeline, equip supervisors to respond to microaggressions and avoid perpetrating them, and train all students and faculty as allies.

Acknowledgments:

The authors wish to thank Alicia Fernandez, MD, and the University of California, San Francisco Educational Scholarship Conference for their thoughtful and expert feedback on earlier versions of this article, and Victoria Ruddick for her skillful artistic support in helping to design the figure.