One size may not fit all: Exploring how the intersection of race and gender and stigma consciousness predict effective identity-safe cues for Black women

Featuring Black women scientists on science companies’ websites is one recruitment practice that can close gender and racial gaps within STEM fields by increasing feelings of trust and belonging for Black women job applicants.

Introduction

Black women are extremely underrepresented in STEM and make up only 2% of the entire US STEM workforce. Evidence suggests they must work much harder than others to not only prove their capabilities but also to discredit negative stereotypes about the performance of women and Black people in the sciences.  During job recruitment, Black women also face social identity threat, (i.e. when individuals worry they will be undervalued or discriminated against because of identifying with a stigmatized minority group), which may lead to poor recruitment and retention rates. Additionally, stigma consciousness, or sensitivity to the possibility of experiencing discrimination because of their race and/or gender, can increase vulnerability to social identity threat. Evidence suggests that individuals who are higher in stigma consciousness have heightened vigilance for signs that their group will be devalued in a given environment.

Interventions that signal welcoming environments could help companies alleviate this underrepresentation of Black women in STEM. Particularly, the use of identity-safe cues (i.e. images and statements suggesting that an identity will be valued) in recruitment materials may signal to Black women that they will be respected in a STEM company.

The authors of this study investigate the effectiveness of identity-safe cues in online recruitment materials to increase Black women’s sense of belonging and trust in a STEM company. Specifically, they looked at whether gender or race would be more important when developing inclusive recruitment materials for Black women. Additionally, this study explored whether the level of a black woman’s stigma consciousness (to either gender or race) would influence the effectiveness of a particular identity safe cue.

Two main theoretical approaches guided their investigation.

  1. The “ethnic prominence” theory asserts that black women are more sensitive to racial discrimination than gender discrimination, because of the significant history of racial violence and discrimination in the United States. According to this theory, Black men or Black women can equally serve as an identity safe cue for Black women.
  2. The “double jeopardy” theory states that the intersectional effects of gender and race discrimination for Black women cannot be separated and thus only Black women can serve as an effective identity safe cue for other Black women (compared to Black men or White women).

This study also investigated ways in which images of white female scientists who express allyship with Black women could signal a welcoming environment and promote diversity in recruitment.

Findings

In the first experiment, participants were shown images of either a Black woman scientist, a Black man scientist, a White woman scientist, or no scientist. When Black women viewed the profiles of Black women scientists on the fictional company website, they reported the most anticipated trust and belonging at the company.

Experiment 1

  • When shown the profile of a Black woman scientist on a science and technology company’s website, Black female respondents reported higher anticipated trust and belonging (0.21 average z-score) than when showed the profile of a Black man (0.11), White woman (-0.17), or no profile at all (-0.14)
  • When compared to viewing no profile at all, the Black woman’s profile was the only one to significantly increase the participants’ feelings of trust and belonging with the regards to the fictional company (mean difference of 0.35)
  • Participants felt significantly more similar to the Black woman scientist (mean difference of 0.5, based on averages from a 5-point Likert scale) and marginally more similar to the Black man scientist (mean difference of 0.26) when compared to the White woman scientist
  • Participants felt only marginally more similar to the Black woman scientist than the Black man scientist (mean difference 0.24)
  • Participants reported feeling significantly more race stigma conscious (mean score of 3.39) and gender-race stigma conscious (3.41) than simply gender stigma conscious (3.25)
  • Participants with higher overall stigma consciousness (one standard deviation above the mean) reported stronger feelings of trust and belonging for the Black profiles compared to the White woman profile, while those with lower stigma consciousness (one standard deviation below the mean) had similar reported feelings of trust and belonging among all profiles

Experiment 2

Since images of White women scientists elicited the least amount of trust, a follow-up experiment was conducted with a 5 profile condition design, which found that a Black woman scientist or White woman scientist who expressed allyship with Black women both functioned well as identity safe cues.

  • Participants reported the highest feelings of trust and belonging when shown the profile of the Black woman (mean score of 0.17) or the profile of the White woman ally (0.18)
  • Participants reported lower feelings of trust and belonging when shown the White woman who values diversity (-0.02), followed by the profile of a White woman with no mention of race (-0.16), followed by no profile at all (-0.17)
  • Participants reported higher feelings of similarity towards the Black woman profile than the White woman scientist (mean difference of 0.36), while the White ally and diversity profiles did not significantly differ from the White woman condition
  • Participants perceived higher allyship from the Black woman scientist and the White woman ally (mean difference of 0.81 and 0.79 respectively) than from the White woman scientist alone, while the White profile endorsing diversity did not endorse significantly different feelings of allyship
  • Participants who reported higher overall stigma consciousness (one standard deviation above the mean) reported stronger feelings of trust and belonging for the Black woman profile compared to the White ally
  • Participants with lower stigma consciousness (one standard deviation below the mean) reported stronger feelings of trust and belonging with the White Ally profile than the Black woman profile
  • Participants with higher gender-race stigma consciousness reported lower anticipated feelings of trust and belonging for all profiles except for the Black woman profile

Overall, profiles of Black women and Black men scientists were more effective identity-safe cues than White women, with profiles of Black women being the most effective. These findings confirm that images of a successful Black female scientist can help to alleviate social identity threat.  Experiment 2 further revealed that a White woman expressing a deliberate interest in Black women’s professional input signals a welcoming work environment more effectively than the presence of a White woman alone. These results were influenced by the subject’s level of race and/or gender stigma consciousness and White allyship was found most effective for subjects with lower stigma consciousness.

Methodology

This study consisted of two experiments in which the effectiveness of “identify safe cues” for black women scientists was estimated.

Experiment 1

This experiment recruited 393 black women from Amazon’s Mechanical Turk website (a paid crowdsourcing platform) advertising a study about first impressions of science companies. Participants were compensated $1.50 to visit the homepage of a fictional company’s (“LabTech”) website and read a staff profile. Researchers randomized participants to read one of four profiles: that of a Black woman, a White woman, a Black man, or no profile at all. The profiles had identical blurbs for a scientist (under the name Mark or Melissa Evans) who was a “head of a biomedical research team at LabTech” and described their research interests.

Participants who read profiles responded to statements on how similar they felt to the scientist (e.g. “this person seems similar to me”) on a 5-point Likert scale which was used to calculate an index of felt similarity. All participants then responded to statements about the work environment as though they worked at Labtech (e.g.  “I would feel respected”, “People at this company would like me”) on a 5-point Likert scale. These questions captured either the  predicted trust and comfort level or anticipated belonging of the participants. Since predicted trust and comfort scores were strongly correlated with anticipated belonging scores, the z scores of both were averaged into a composite measure indexing anticipated trust and belonging that was used in the final analysis. Z-scores here are the number of standard deviations away from the mean reported value for anticipated trust, comfort and belonging.       

Participants also responded to statements about race, gender and gender-race stigma (e.g. “When interacting with people, I feel like they interpret all of my behaviors in of my gender [race] [race and gender]”) on a 5-point Likert scale. The experiment ended with a brief demographic questionnaire and a debriefing with the participants.

Experiment 2

Similar to Experiment 1, 426 Black women recruited from Mechanical Turk were compensated $1.50 to view LabTech’s website, read a scientist’s profile, and respond to a questionnaire. Researchers then randomized participants to read one of 5 profiles: a Black women scientist, a White woman scientist, a White woman scientist who endorses diversity, a White woman scientist who acknowledged the value Black women bring to the sciences (the “ally”), or no profile at all.

Participants responded to the same surveys as Experiment 1 with some additional questions: two statements assessing the perceived “allyship” of the scientist with Black women (on a 5-point Likert scale) and statements measuring participants’ potential organizational commitment to the company. The latter’s responses were averaged to create an index of predicted affective organizational commitment (on a 5-point Likert scale). This index highly correlated to the anticipated trust and belonging score and was included into the score by averaging the compiled z-scores. Participants then responded to questions regarding stigma consciousness. The experiment concluded with a brief demographic questionnaire and debriefing.

Related GAP Studies