Ally endorsement: Exploring allyship cues to promote perceptions of allyship and positive STEM beliefs among White female students

Abstract

Many science, technology, engineering, and math (STEM) domains are White-male-dominated; yet, investigations exploring how White men scientists can counter women’s underrepresentation are lacking. We examined whether exposure to a White male scientist endorsed as an ally by a gender ingroup member (i.e., an allyship cue) encouraged identity-safety and positive STEM beliefs among White female students. Relative to a White male scientist without an allyship cue, participants that viewed an endorsed scientist reported greater perceptions of allyship, identity-safety (Experiment 1–3), interest, and self-efficacy in computer science (Experiments 2 and 3). In Experiment 3, we recruited White women in STEM and found that ally endorsement from a White or Black female, but not a White male, promoted identity-safety. Moreover, the endorsed scientist encouraged perceptions that they possessed a shared reality with participants and recognized the bias that women in STEM face. Allyship cues may help White men attract White female students to STEM.

Keywords

allyship gender identity-safety cues STEM

Women are underrepresented in many science, technology, engineering, and math (STEM) domains (U.S. Department of Education, 2019). That is, the majority of STEM areas are male-dominated (National Science Foundation, 2019), which can threaten women’s identity-safety (i.e., perceptions that your identity will be valued in a particular setting; Steele et al., 2002) and discourage the pursuit of STEM classes and majors (Lewis et al., 2017; Martiny & Nikitin, 2019). Fortunately, exposure to a successful role model can counter such concerns among women (Dasgupta, 2011; Dennehy & Dasgupta, 2017). Indeed, even short-term exposure to a successful ingroup scientist can encourage positive STEM beliefs (Dasgupta & Asgari, 2004), such as promoting interest (Pietri, Johnson, et al., 2020) and self-efficacy in STEM (Stout et al., 2011).

Given women’s low representation in STEM, access to ingroup role models for women in STEM fields is limited. Moreover, research examining how men—who comprise the majority in STEM fields—can potentially attract women to STEM is lacking (Akcinar et al., 2011). The present experiments examine a novel approach to encourage identity-safety and positive STEM beliefs among White women students, drawing from research examining the efficacy of an allyship cue (i.e., an explicit signal that an outgroup member acts to support the success of your ingroup; Johnson & Pietri, 2020). Specifically, we examined whether a gender ingroup member endorsing a White male scientist as an ally acted as an effective allyship cue among White female students, promoted perceptions the outgroup scientist was an ally, and encouraged positive STEM beliefs. Recognizing that White women have multiple identities—a positively stereotyped racial identity and a negatively stereotyped gender identity in STEM (Eaton et al., 2020; Fiske et al., 2002)—we also tested whether an ally endorsement from a racial ingroup member (i.e., a White woman) would be equally effective as an endorsement from a racial outgroup member (i.e., a Black woman). Finally, extending past work (Johnson & Pietri, 2020), we also explored the processes underlying ally endorsement among a sample of White female students with STEM backgrounds.

Identity-Safety Cues in STEM Environments

White men comprise about one-third of the U.S. population but make up almost half of the U.S. STEM workforce (National Science Foundation, 2019). Although White women earn college degrees in greater numbers than their male counterparts (U.S. Department of Labor, 2019), their greater presence in higher education has yet to translate to representation in all STEM fields. Notably, the underrepresentation of women in STEM fields is not due to a lack of ability (Lindberg et al., 2010). For example, past work examining female underrepresentation in computer science has found that male and female computer science students do equally well in their computer science courses (Beyer, 1999, 2008; Tam & Bassett, 2006). The current demand for STEM professionals exceeds those available (U.S. Department of Labor, 2019), and colleges and universities are in a unique position to potentially spark interest in STEM courses among female students to ultimately increase their representation in STEM fields.

Women’s smaller representation in STEM may pose a challenge to attracting White female students to STEM classrooms because lack of numerical representation has been found to undermine feelings of identity-safety (Murphy et al., 2007). Concerns about identity-safety are often associated with decrements in anticipated belonging and trust, or in expectations that one will be treated fairly and welcome in STEM environments (Murphy & Taylor, 2012; Steele et al., 2002). Critically, an increased sense of belonging and trust among women has been found to significantly predict positive STEM beliefs, or interest in STEM classes (Cheryan et al., 2011, 2013) and self-efficacy in STEM (i.e., belief one can succeed; Findley-van Nostrand & Pollenz, 2017). Likewise, women that report greater belonging and trust, as well as more positive STEM beliefs, persist in STEM and ultimately earn STEM degrees (Good et al., 2012; Lewis et al., 2016, 2017). Consequently, encouraging White female students’ identity-safety and promoting positive STEM beliefs is critical for increasing their presence in STEM.

Exposing female students to an identity-safety cue (i.e., a cue signaling one’s identity is valued; Avery et al., 2004; Davies et al., 2005; Walton et al., 2015) is one reliable and robust strategy to attract women to STEM settings. Though some identity-safety cues (e.g., organizational diversity statements) may promote identity-safety by signaling an environment (or the people in it) will treat others fairly (Chaney et al., 2018; Johnson et al., 2021; Kaiser et al., 2013), numerous studies have examined successful ingroup scientists as effective identity-safety cues (see Dasgupta, 2011; MMorgenroth et al., 2015). The presence of a successful ingroup scientist can signal one belongs in STEM, and both meaningful and brief contact with relatable ingroup scientists are associated with increased identity-safety among female students in STEM settings (Dasgupta et al., 2015; Dennehy & Dasgupta, 2017; Johnson et al., 2019; O’Brien et al., 2017). Notably, although exposure to both women and men role model scientists has been found to encourage women’s belonging and interest in STEM (Cheryan et al., 2011, 2013), most research examining identity-safety cues has focused on women as ingroup role models (Dasgupta et al., 2015; Morgenroth et al., 2015). Fewer investigations have examined strategies to help White male scientists signal identity-safetyfor White women (for one exception, see Pietri, Drawbaugh, et al., 2020). Given the limited available ingroup role models due to women’s underrepresentation in STEM, identifying effective strategies to help White men promote identity-safety among White women may help attract women to STEM.

Promoting Positive STEM Beliefs Via Allyship Cues

One strategy to help White male scientists promote identity-safety may be via ally endorsement, an allyship cue in which an ingroup member shares their positive experiences with an outgroup member and labels them an ally (Johnson & Pietri, 2020). A growing number of studies have found that perceiving an outgroup scientist as an ally is positively associated with greater identity-safety in STEM environments (Chaney et al., 2018; Johnson et al., 2019; Moser & Branscombe, 2021; Pietri, Johnson, & Ozgumus, 2018). Relevant to the present research, Johnson and Pietri (2020) investigated whether ally endorsement was an effective allyship cue among Black women. Specifically, Black female participants read the profile of a successful White female scientist working in a fictitious biotechnology company, and the profile either did or did not contain an endorsement allyship cue. To operationalize endorsement, participants read a testimonial from a Black woman (i.e., a racial ingroup member) that described her positive experiences with a White female scientist and endorsed them as an ally. Relative to those who learned about a White female scientist with no allyship cue, those in the endorsed ally condition reported greater perceptions that the scientist was an ally, as well as greater anticipated belonging and trust. Taken together, this previous work demonstrates the utility of an allyship cue and suggests that promoting perceptions that a White male scientist is an ally may facilitate feelings of belonging and trust, and positive STEM beliefs among White women undergraduates.

Because previous research has only tested an endorsement allyship cue among Black women and not among a student sample, the current research builds upon past work and tests the efficacy of different endorsement allyship cues among White female students. Like Black women, White women also have multiple identities (i.e., White and female); however, their racial identity is high status and positively stereotyped in U.S. society (Eaton et al., 2020; Fiske et al., 2002). Thus, among White women, a gender-relevant identity-safety cue, or a cue relevant to their negatively stereotyped identity, has been found to promote identity-safety (e.g., Dennehy & Dasgupta, 2017; Stout et al., 2011). Furthermore, previous work has found that the presence of similarly stereotyped individuals (e.g., racial minorities) can encourage identity-safety among White women (Chaney et al., 2018), suggesting an endorsement allyship cue from a White woman or a Black woman (i.e., a racial outgroup but gender ingroup member) may both encourage identity-safety for White women.

An additional goal of the present work was to explore why ally endorsement encourages identity-safety. Ally endorsement is comprised of two key processes—a gender ingroup member expressing their positive experiences with an outgroup scientist, and explicitly labeling the outgroup scientist an ally. Although previous work found ally endorsement encouraged perceptions of allyship, and these perceptions positively related to identity-safety (Johnson & Pietri, 2020), it is unclear from this past work which aspect of ally endorsement is most important for encouraging identity safety. Consequently, in the present work, we also manipulated each aspect of ally endorsement to examine its independent effect on identity-safety and perceptions of allyship.

Drawing from recent literature adopting a multiple perspective approach to identify predictors of allyship (Kutlaca et al., 2020; Radke et al. 2020), as well as emerging scholarship examining the role of shared reality (i.e., having a similar inner state, or feelings, thoughts, and concerns; Echterhoff et al., 2009) in fostering social connections (Higgins et al., 2021; Rossignac-Milon et al., 2021), we also explored whether ally endorsement shaped other perceptions of the scientist beyond perceived allyship. Specifically, we examined whether the presence of an allyship cue encouraged perceptions the White male scientist recognizes bias women in STEM face. Recognition of the bias marginalized groups face has been identified as an essential component of allyship by persons belonging to high-status groups (Drury & Kaiser, 2014; LaCosse et al., 2021), as well as by members of marginalized groups (Ashburn-Nardo, 2018; Brown & Ostrove, 2013). Critically, awareness of the discrimination marginalized groups face has been found to encourage perceptions that White individuals are allies of racial and ethnic minorities (Droogendyk et al., 2016; Ostrove & Brown, 2018; Thorne et al., 2021), as well as perceptions that men are gender allies of women (Drury & Kaiser, 2014; Radke et al., 2018). Consequently, we also explored if ally endorsement altered perceptions the White male scientist recognized the bias women in STEM face.

Relatedly, perceiving a White male scientist as recognizing the bias women in STEM face may also signal the scientist shared reality with participants generally. A shared reality can facilitate more positive interpersonal interactions in close relationships (Higgins et al., 2021) and even among strangers (Rossignac-Milon et al., 2021). However, research examining shared reality as it relates to allyship and intergroup perceptions is scant. Of relevance to the present work, sharing a reality with another person can encourage feelings of trust (Echterhoff et al., 2009; Rossignac-Milon et al., 2021). Thus, we also explored whether the allyship cue encouraged perceptions the White male scientist shared participants’ reality, and whether these perceptions positively related to perceptions of allyship.

Overview and Hypotheses

The goal of the present experiments was to explore whether ally endorsement encouraged perceptions that a White male scientist was an ally and promoted identity-safety in STEM and positive STEM beliefs among White female students. In Experiment 1, participants viewed the profile of a White male scientist with no mention of allyship, or a profile where the White male scientist was endorsed as an ally by a White female. Participants then reported perceptions of whether the scientist was an ally, and their anticipated belonging and trust in STEM fields (i.e., identity-safety). In Experiment 2, participants viewed the profile of a computer science professor, presumably employed at their institution, in which the presence of an allyship cue was manipulated. Specifically, participants read a profile that contained no allyship cue, a profile in which the White male scientist was endorsed as an ally by a White female student, or a profile in which the White male scientist was endorsed as an ally by a Black female student. Thus, in Experiment 2, we tested whether endorsements from both White and Black female students would function as efficacious allyship cues. In addition to reporting their anticipated belonging and trust in computer science, we also had participants report their interest and self-efficacy in that field (i.e., STEM beliefs).

As a secondary goal, in Experiment 3, we also aimed to better understand the processes underlying ally endorsement. As noted previously, ally endorsement is comprised of two components: (a) the presence of a gender ingroup member sharing their positive experiences with an outgroup scientist and (b) explicitly endorsing the outgroup scientist as an ally of women in STEM. To identify which aspect of ally endorsement encouraged identity-safety, in Experiment 3, we manipulated each of these components. Specifically, in Experiment 3, we recruited White women in STEM fields and explored whether ally endorsement encouraged identity-safety when the endorsement comes from a White male student (i.e., gender outgroup member), and whether the presence of a gender ingroup member requires the explicit mention of allyship. As in Experiment 2, participants read information about a White male computer scientist in which we manipulated the presence of an allyship cue. In addition to receiving an ally endorsement by a White female or a Black female, we also added a condition where the scientist was endorsed by a White male student. We also included a condition where a White female student shared her positive experiences with the scientist but made no mention of allyship.

Across all experiments, we predicted that relative to the no allyship cue condition, the ally endorsement conditions featuring a gender ingroup member would encourage greater perceptions the White male scientist was an ally and promote greater identity-safety (i.e., Experiments 1–3) and positive STEM beliefs (i.e., Experiments 2 and 3). In Experiment 3, we also explored whether the presence of the allyship cue encouraged perceptions the White male scientist recognized bias and shared a view of reality, and whether these perceptions positively related to perceived allyship.

Experiment 1

Participants viewed the profile of a White male scientist with no mention of allyship, or a profile where the White male scientist was endorsed as an ally by a White female (i.e., the allyship cue). We anticipated that, relative to the no allyship cue condition, the allyship cue condition would result in greater perceptions the White male scientist was an ally and in greater anticipated identity-safety.

Method

Participants

Participants were students at a private university that participated in exchange for partial course credit. Because the majority of students at the university identified as White women, we intentionally advertised the study to students of all gender identities and races. Only results for White female participants (N = 97) were included for data analysis (results do not change when including the full sample; see supplemental material.

Due to resource constraints, in Experiments 1 and 2, we adopted a time-based stopping rule, and no data were analyzed prior to the end of data collection. A post hoc power sensitivity analysis with G*Power (Faul et al., 2007) found that with an independent samples t test, at 80% power, we were adequately powered to find an effect size of d = 0.51 at p < .05.

Controlling for demographic variables did not change our results (see Table 1 for demographic information for Experiments 1–3); thus, across all experiments, we did not include these variables in the analyses.

Table 1.

Demographic information: Experiments 1–3.

	Age	School year	STEM major
Experiment 1 (N = 97)	18–23 (M = 18.79, SD = 1.02)	62 (63.9%), 1st year; 22 (22.7%), 2nd year; 6 (6.2%), 3rd year; 7 (7.2%), 4th year.	15 (15.5%) reported STEM field or considering STEM field as major.
Experiment 2 (N = 152)	18–22 (M = 18.70, SD = 0.93)	102 (67.1%), 1st year; 29 (19.1%), 2nd year; 12 (7.9%), 3rd year; 9 (5.9%), 4th year.	27 (17.8%) reported STEM field or considering STEM field as major.
Experiment 3 (N = 289)	18–71 (M = 23.76, SD = 6.63)	3 (1.0%), 11th grade in high school; 36 (12.5%), 12th grade in high school; 22 (11.4%), 1st year college; 48 (16.6%), 2nd year college; 54 (18.7%), 3rd year college; 42 (14.5%), 4th year college; 50 (17.3%), 5+ years of college; 23 (8.0%), reported “Other.”	49 (17.0%), computer sciences; 36 (12.5%), computing (IT); 89 (30.8%), engineering; 14 (4.8%), physics; 38 (13.1%), science; 30 (10.4%), mathematics; 33 (11.4%), reported a field or major that combined more than one of the above.

Procedure

Across all experiments, all materials were adapted from Johnson and Pietri (2020), and participants began the online study by providing informed consent and were debriefed and thanked upon completion of the study.

The current experiment had a two-condition design (no allyship cue [n = 53] vs. endorsed ally-White female [n = 44]) and was advertisedas a study examining “impressions of organizations”. Participants viewed a web page for a fictitious biotechnology company and were asked to imagine they were employed at the company. Next, participants were randomly assigned to read the employee profile of a successful White male scientist that either did (i.e., endorsed ally-White female) or did not include an allyship cue (i.e., no allyship cue).

Participants in both conditions saw a picture of a White male scientist, Dr. Evans, and a profile describing his latest research projects. The endorsed ally-White female profile, however, also featured a testimonial from a community partner, Amber Pratcher, a principal at a local school. The testimonial included a photo of Amber (a professional headshot of a smiling White woman). Importantly, the testimonial noted Amber viewed Dr. Evans as an ally, “he recognizes that women are underrepresented, and actively works to support and recruit women into the sciences” (see the Appendix for the full manipulation, and Figure 1 for an example of the ally endorsement manipulation across experiments). Participants then reported perceived allyship and anticipated belonging and trust.

Figure 1.

Example of the endorsed ally manipulation.

Measures

Across all experiments, we measured continuous variables using a 5-point, Likert-type scale ranging from 1 (strongly disagree) to 5 (strongly agree). For each measure, we averaged participants’ responses to the items, with higher scores indicating more of each measured construct (see the Appendix for exact items for all measures in Experiments 1–3).

Participants rated their agreement with two items assessing perceived allyship (e.g., “Most likely this person wants to help women succeed in the sciences”); r(97) = .85, p < .001; M = 3.69, SD = 0.96 (Pietri, Johnson, & Ozgumus, 2018), and 12 items assessing anticipated belonging and trust¹ (e.g., “People in the company would like me”): five items from Good et al. (2012), four items from Purdie-Vaughns et al. (2008), three items from Walton and Cohen (2007); M = 3.33, SD = 0.62, α = .88.

Results and Discussion

Data and syntax for all experiments can be found at the Open Science Framework (https://osf.io/xgzv4/?view_only=a420005cc61843eaa5ba9d4f1a940897); see Table 2 for correlations among variables for Experiments 1–3. We first conducted an independent samples t test on perceived allyship and found that participants in the endorsed ally-White female condition (M = 4.47, SD = 0.63) reported significantly greater allyship than those in the no allyship cue condition (M = 3.01, SD = 0.55); t(95) = 12.17, p < .001, d = 2.47; mean difference: 1.49, 95% CI [1.25, 1.73]. An independent samples t test on anticipated belonging and trust found that participants in the endorsed ally-White female condition (M = 3.47, SD = 0.56) also reported significantly greater belonging and trust than participants in the no allyship cue condition (M = 3.22, SD = 0.93); t(95) = 2.00, p = .048, d = 0.41; mean difference: 0.25, 95% CI [0.01, 0.50]. Thus, Experiment 1 found that a White woman endorsing a White male scientist as an ally effectively promoted perceptions the scientist was an ally as well as greater anticipated belonging and trust, relative to participants in the no allyship cue condition.

Table 2.

Correlations among variables: Experiments 1–3.

	Experiment 1	Experiment 2				Experiment 3
Variable	1	1	2	3	4	1	2	3	4	5
1. Allyship	-	-				-
2. Belonging and trust	.32**	.26*	-			.35*	-
3. Interest	-	.20**	.63*	-		.40*	.46*	-
4. Self-efficacy	-	.20**	.60*	.54*	-	.16*	.54*	.27*	-
5. Recognizes bias	-	-	-	-	-	.68*	.21*	.37*	.11***	-
6. Shared reality	-	-	-	-	-	.47*	.46*	.44*	.27*	.40*

Note. In Experiment 2, interest refers to interest in computer science courses broadly, and in Experiment 3, it refers to interest in courses with the computer science professor.

p < .01. **p < .05. ***p < .10.

Experiment 2

In Experiment 2, we aimed to examine whether our allyship manipulation also promoted positive STEM beliefs as well as belonging and trust in computer science generally (i.e., not at a specific company). Consequently, participants reported their anticipated belonging and trust as well as their interest and self-efficacy in computer science broadly. New to Experiment 2, all participants learned about a computer science professor ostensibly working at their university, who either was endorsed by a White woman or a Black woman. Adding the Black woman condition allowed us to explore whether an endorsement was equally effective when it came from a racial and gender ingroup member (i.e., a White woman) or from only a gender ingroup member (i.e., a Black woman).²

Method

Participants

Participants took part in the study in exchange for partial course credit, and students who participated in Experiment 1 were excluded from accessing the study. In a departure from Experiment 1, the study was only advertised to students that identified as women; we recruited 178 participants across the semester. We again present the findings only for White female students (N = 152; see the supplemental material for results with the entire sample). A post hoc power sensitivity analysis using G*Power (Faul et al., 2007) found that for a between-subjects ANOVA with three groups, at 80% power, we were adequately powered to find an effect size of f = .25 at p < .05.

Procedure

The current experiment had a three-condition design (no allyship cue [n = 55] vs. endorsed ally-White female [n = 50] vs. endorsed ally-Black female [n = 47]) and was advertised as an online study examining “impressions of professor profiles” at the participants’ institution. All materials were modeled after Experiment 1. Participants viewed the web page for the Department of Computing Sciences at their institution, which was altered by the researchers to remove photos of faculty and students. However, all other information was consistent with the actual departmental website. Next, participants were randomly assigned to read either the profile of a successful White male computer science professor that included no allyship cues (i.e., no allyship cue), an identical profile that also included a testimonial from one of the professor’s White female students (i.e., endorsed ally-White female), or an identical profile that also included a testimonial from one of the professor’s Black female students (i.e., endorsed ally-Black female).

Across all conditions, participants saw a picture of a White male professor, Dr. Evans, who was described as a professor of computing science. The profile, made to match the profiles of professors actually employed at the institution, described how Dr. Evans became interested in computer science as well as his latest research involving students. The endorsed ally conditions, however, also featured a testimonial from a student currently working in Dr. Evans’s lab, Amber Pratcher. The testimonial included a photo of Amber; however, for those in the endorsed ally-White female condition, the photo featured a White female student, while those in the endorsed ally-Black female condition saw a photo featuring a Black female student. Besides the different photos, the testimonial was identical in the two endorsed ally conditions. The testimonial was primarily consistent with Experiment 1, although we adapted the language for our manipulations. Specifically, in both of the endorsed ally conditions, the student stated Dr. Evans advocates for both women and women of color (see the Appendix and Figure 1).

Participants next reported perceived allyship, anticipated belonging and trust, interest, and self-efficacy in computer science (the latter two in random order).

Measures

Participants rated their agreement with the same two items from Experiment 1 assessing perceived allyship (M = 3.99, SD = 0.70); r(152) = .69, p < .001, and their agreement with 12 items assessing anticipated belonging and trust (e.g., “People in the Department of Computing Sciences would like me”; M = 3.44, SD = 0.65, α = .90), adapted to focus on computer science.

Participants also indicated their agreement with three items assessing interest in computer science (e.g., “I would be interested in taking classes in computing sciences as a major; M = 3.37, SD = 1.05, α = .92; Pietri, Johnson, et al., 2020) and a single item assessing self-efficacy in computer science (“I feel confident I would have the ability to do well in computing sciences”; M = 3.51, SD = 0.98; adapted from Lent et al., 2005).

Results and Discussion

For Experiments 2 and 3, across all measures, we utilized a between-subjects analysis of variance (ANOVA) and least significant difference (LSD) post hoc tests to compare mean differences across conditions. Full ANOVA results, means, and standard deviations across conditions as well as mean differences and effect sizes relative to the no allyship condition can be found in Table 3. As shown in Table 3, we found a significant effect of condition across all measures (all p values ⩽ .030).

Table 3.

ANOVA, means, and standard deviations across conditions, and mean differences and effect sizes relative to the no allyship cue condition for Experiment 2.

Measure	F (2, 152)	p	η_p²	No allyship cue(n = 55)	Endorsed ally-White female(n = 50)	Endorsed ally-Black female(n = 47)
				M (SD)	M (SD)	M (SD)
1. Allyship	42.61	< .001	.36	3.44 (0.66)_b	4.35 (0.45)_a	4.26 (0.54)_a
2. Belonging and trust	6.01	.003	.08	3.23 (0.64)_b	3.66 (0.61)_a	3.43 (0.63)_ab
3. Interest	3.58	.030	.05	3.08 (1.04)_b	3.50 (1.02)_a	3.58 (1.04)_a
4. Self-efficacy	6.43	.002	.08	3.16 (1.03)_b	3.82 (0.85)_a	3.57 (0.92)_a
Mean differences and effect sizes relative to the no allyship cue condition
Measures		Endorsed ally-Black female(n = 50)		Endorsed ally-White female(n = 47)
		MD	d	MD	d
Allyship		0.82	1.36	0.92	1.62
Belonging and trust		0.19	0.28	0.43	0.65
Interest		0.48	0.10	0.42	0.41
Self-efficacy		0.41	0.11	0.70	0.01

Note. Within each row, shared subscripts indicate no significant difference.

Participants in both the endorsed ally-White female condition and the endorsed ally-Black female condition reported greater perceptions the White male professor was an ally, relative to the no allyship cue condition (both p values < .001). In contrast, participants in the two endorsed ally conditions did not differ (p = .406). Similarly, for anticipated belonging and trust, participants in the endorsed ally-White female condition (p = .001) reported greater anticipated belonging and trust, relative to the no allyship cue condition. Participants in the endorsed ally-Black female condition also tended to report greater belonging and trust, relative to the no allyship cue condition (p = .072); however, this difference was only marginally significant. The two endorsed ally conditions did not differ (p = .121).

Regarding interest in computer science, relative to the no allyship cue condition, participants in the endorsed ally-Black female condition (p = .015) and the endorsed ally-White female condition (p = .038) reported greater interest in computer science. The two endorsed ally conditions did not significantly differ (p = .920). Finally, turning to self-efficacy in computer science, relative to the no allyship cue condition, participants in the endorsed ally-Black female condition (p = .031) and the endorsed ally-White female condition (p = .001) reported greater self-efficacy in computer science. In contrast, participants in the two endorsed ally conditions did not differ (p = .205).

In summary, in Experiment 2, we found that our two allyship cue conditions were equally effective at encouraging perceptions of allyship, identity-safety, and positive STEM beliefs.Specifically, we found that, relative to the no allyship cue condition, endorsement as an ally by a White female or a Black female student served as an effective allyship cue, and resulted in greater perceptions that the White male computer scientist was an ally, greater belonging and trust as well as interest and self-efficacy in computer science broadly.

Experiment 3

In Experiment 3, we aimed to replicate our previous findings, this time with a predominantly student sample of White women established in STEM fields, as well as to better understand the mechanisms underlying the effectiveness of endorsement. Specifically, we tested whether simply learning that a gender ingroup member has had positive experiences with the White male scientist encouraged identity-safety, and whether an ally endorsement was effective when it came from a gender outgroup member (i.e., a White male student). Thus, we manipulated two important components of ingroup ally endorsement—the explicit mention of an allyship and the gender of the endorser (i.e., gender ingroup endorsement). To explore the generalizability of our findings, we recruited White women in STEM majors and fields, and added two new conditions: one in which the White male scientist was endorsed by a White male student and another where a White woman shared her positive experiences but did not endorse the scientist as an ally. Finally, to further understand the mechanisms involved in ally endorsement, we also explored whether the allyship cue promoted perceptions the scientist recognized bias and shared reality with participants, and whether these perceptions positively related to perceived allyship.

Method

Participants

We recruited 330 White women in STEM fields via the data acquisition platform Prolific; participants were paid $2.55 (i.e., approximately $10.23/hour). Our sample size for Experiment 3 was informed by Experiment 2. Specifically, in Experiment 2, the difference between the no allyship cue condition and the endorsed ally-White female condition yielded an effect size of d = 1.58 for allyship, d = 0.67 for belonging and trust, d = 0.39 for interest in computer science, and d = 0.62 for self-efficacy in computer science. We selected the smallest effect size of the four and conducted an a priori analysis using G*Power (Faul et al., 2007) for a one-way, between-subjects ANOVA with five groups, at 80% power. The results indicated we would need approximately 305 participants. To account for potential exclusions, we oversampled and recruited 330 participants.

To ensure our sample was comprised of only White women, we excluded 41 participants that reported their gender and/or race as otherwise, leaving a final sample of N = 289. A post hoc power sensitivity analysis using G*Power (Faul et al., 2007) indicated that for a between-subjects ANOVA with five groups, at 80% power, we were adequately powered to find an effect size of f = .21 at p < .05.

Procedure

The current experiment had a five-condition design (no allyship cue [n = 60] vs. positive experiences-White female [n = 54] vs. endorsed ally-White male [n = 60] vs. endorsed ally-Black female [n = 59] vs. endorsed ally-White female [n = 56]). Consistent with Experiment 2, the experiment was advertised as a study examining “impressions of professor profiles”. Participants viewed the web page for a fictitious department of computing sciences. Next, participants were randomly assigned to read one of five profiles featuring a successful White male computer scientist, three of which were the same conditions as in Experiment 2 (i.e., no allyship cue, endorsed ally-White female, endorsed ally-Black female). New to Experiment 3, the fourth condition featured a testimonial from one of the professor’s White female students, but only noted the student’s positive experience with Dr. Evans and made no mention of allyship (positive experiences-White female). The fifth condition was identical to the two endorsed ally conditions; however, it featured a White male student as the endorser (i.e., endorsed ally-White male).

Participants saw the same profile information for Dr. Evans as in Experiment 2. Those in the no allyship cue condition only saw the profile of Dr. Evans, whereas those in the positive experiences-White female condition saw this information in addition to a testimonial from a featured student, Amber Pratcher. The testimonial was the same as that used in Experiment 2 but only included Amber’s photo alongside the first paragraph describing Amber’s positive experiences with Dr. Evans, without mention of allyship. For the remaining three endorsed ally conditions, the testimonial information was identical to that of Experiment 2. The endorsed ally conditions each featured a photo³ of a student matching the race and gender of the profile in each respective assigned condition. The student was named Amber Pratcher in the endorsed ally-Black female and endorsed ally-White female conditions, and Aaron Pratcher in the endorsed ally-White male condition. All remaining information was identical across the endorsed ally conditions.

Participants next reported perceived allyship, shared reality, and recognition of bias (all in random order). Participants then reported anticipated belonging and trust, interest in classes with Dr. Evans, and self-efficacy in computer science (in random order).

Measures

All measures were scored as in Experiment 2. After reading the professor profile, participants reported perceived allyship (M = 4.00, SD = 0.76); r(289) = .70, p < .001, anticipated belonging and trust (M = 3.58, SD = 0.64, α = .89), and self-efficacy in computer science (M = 3.81, SD = 0.96). Because our sample consisted of individuals already established in STEM fields, in a deviation from Experiment 2, we asked participants to respond to two items gauging their interest in taking classes with Dr. Evans rather than computer science generally (e.g., “I would be interested in learning more about how computer science applies to my major/intended major with Dr. Evans”; M = 4.07, SD = 0.78), r(289) = .72, p < .001.

New to Experiment 3, participants also completed three items to index perceptions that Dr. Evans recognizes bias (e.g., “Most likely, Dr. Evans is aware of discrimination facing women in STEM”; modified from Pietri, Johnson, Ozgumus, & Young, 2018; M = 3.89, SD = 0.80, α = .88). Participants also completed four items to assess perceptions of shared reality (e.g., “I think Dr. Evans and I share the same thoughts and feelings about things”; modified from Rossignac-Milon et al., 2021; M = 3.58, SD = 0.80, α = .88).

Results

Full ANOVA results with means and standard deviations across conditions for all outcome measures can be found in Table 4; mean differences and effect sizes relative to the no allyship cue and positive experiences-White female conditions can be found in Table 5. We found a significant effect of condition across all measures (all p values ⩽ .027), with the exception of self-efficacy (p = .074).

Table 4.

ANOVA, means, and standard deviations across conditions: Experiment 3.

Measures	F (4, 284)	p	η_p²	M (SD) across conditions
				No allyship cue(n = 60)	Positive experiences-White female(n = 54)	Endorsed ally-White male(n = 60)	Endorsed ally-Black female(n = 59)	Endorsed ally-White female(n = 56)
Allyship	26.52	< .001	.27	3.40 (0.73)_a	3.68 (0.71)_b	4.32 (0.66)_c	4.31 (0.53)_c	4.32 (0.61)_c
Belonging and trust	3.52	.008	.05	3.34 (0.78)_a	3.65 (0.51)_b	3.52 (0.62)_ab	3.69 (0.61)_b	3.70 (0.58)_b
Interest	3.31	.011	.05	3.91 (0.82)_a	4.07 (0.76)_a	3.99 (0.84)_a	4.38 (0.53)_b	3.98 (0.85)_a
Self-efficacy	2.15	.074	.03	3.67 (1.12)_a	3.80 (0.94)_ab	3.78 (0.92)_ab	3.68 (1.06)_a	4.13 (0.66)_b
Recognizes bias	16.36	< .001	.19	3.41 (0.94)_a	3.53 (0.71)_a	4.22 (0.66)_b	4.17 (0.59)_b	4.09 (0.65)_b
Shared reality	4.63	.001	.06	3.27 (0.86)_a	3.52 (0.78)_a	3.60 (0.82)_b	3.88 (0.65)_bc	3.65 (0.77)_bc

Note. Within each row, shared subscripts indicate no significant difference.

Table 5.

Mean differences and effect sizes relative to the no allyship cue and positive experiences conditions: Experiment 3.

Mean differences and effect sizes relative to the no allyship cue condition
Measures	Positive experiences-White female(n = 54)		Endorsed ally-White male(n = 60)		Endorsed ally-Black female(n = 59)		Endorsed ally-White female(n = 56)
	MD	d	MD	d	MD	d	MD	d
Allyship	0.28	0.38	0.92	1.32	0.91	1.43	0.91	1.37
Belonging and trust	0.32	0.48	0.19	0.27	0.35	0.51	0.37	0.53
Interest	0.17	0.21	0.08	0.10	0.47	0.68	0.07	0.09
Self-efficacy	0.13	0.13	0.12	0.11	0.01	0.01	0.46	0.50
Recognizes bias	0.13	0.15	0.81	1.00	0.76	0.97	0.68	0.84
Shared reality	0.25	0.31	0.33	0.39	0.60	0.79	0.38	0.46
Mean differences and effect sizes relative to the positive experiences-White female condition
Measures		Endorsed ally-White male(n = 60)		Endorsed ally-Black female(n = 59)		Endorsed ally-White female(n = 56)
		MD	d	MD	d	MD	d
Allyship		0.64	0.94	0.63	1.02	0.65	0.98
Belonging and trust		−0.13	0.22	0.04	0.07	0.05	0.09
Interest		−0.08	0.10	0.30	0.46	−0.09	0.11
Self-efficacy		−0.01	0.02	−0.12	0.12	0.33	0.41
Recognizes bias		0.69	1.00	0.64	0.97	0.56	0.82
Shared reality		0.08	0.09	0.35	0.46	0.12	0.15

Note. To avoid redundancy, we omitted the no allyship cue condition in the bottom table; see the top table for comparison between this condition and the positive experiences-White female condition.

Regarding perceived allyship, participants in the endorsed ally-White female condition, the endorsed ally-Black female condition, and the endorsed ally-White male condition reported greater perceptions the White male professor was an ally, relative to both the no allyship cue condition (all p values < .001) and the positive experiences-White female condition (all p values < .001). Interestingly, the positive experiences-White female condition also resulted in significantly greater allyship, relative to the no allyship cue condition (p = .024). In contrast, participants in the three endorsed ally conditions did not differ on their level of perceived allyship (all p values ⩾ .943). Thus, the three ally endorsed conditions resulted in the highest perceptions of allyship. Interestingly, learning about the positive experiences of a White female student in the scientist’s lab also encouraged allyship perceptions, but to a significantly lesser extent than an explicit ally endorsement.

Next, turning to anticipated belonging and trust, the endorsed ally-White female (p = .003), the endorsed ally-Black female (p = .006), and the positive experiences-White female (p = .006) conditions each resulted in greater belonging and trust, relative to the no allyship cue condition. In contrast, the endorsed ally-White male condition did not significantly differ from the no allyship cue (p = .216), the positive experiences-White female (p = .122), or the two remaining endorsed ally condition (p values ⩾ .077). The endorsed ally-Black female and endorsed ally-White female conditions did not significantly differ from each other (p = .802), nor did they differ from the positive experiences-White female condition (p = .966 and p = .839, respectively). Unlike perceptions of allyship, only the two gender ingroup endorsement conditions encouraged identity-safety, while the endorsed ally-White male condition failed to significantly encourage identity-safety. Moreover, learning about a female student without an explicit mention of allyship also significantly promoted identity-safety.

For interest in classes with Dr. Evans, relative to the no allyship cue condition, participants in the endorsed ally-Black female condition (p = .001) reported significantly greater interest. Although participants in the other three conditions indicated higher interest, they did not differ significantly from the no allyship cue condition (all p values ⩾ .250; see Table 5). For self-efficacy in computer science, only the endorsed ally-White female condition resulted in significantly greater self-efficacy, relative to the no allyship cue condition (p = .010; all remaining p values ⩾ .470). Similar to interest, although participants in other three conditions indicated higher self-efficacy, these three conditions did not differ significantly from the no allyship cue condition (all p values ⩾ .470). Thus, we found some evidence that our (gender ingroup) endorsement manipulations influenced interest and self-efficacy; however, these effects were relatively weak among our sample of students with STEM backgrounds.

For recognizes bias, participants in the endorsed ally-White male condition, the endorsed ally-White female condition, and the endorsed ally-Black female condition reported significantly greater perceptions the White male professor recognized bias, relative to both the no allyship cue condition (all p values < .001) and the positive experiences-White female condition (all p values < .001). The positive experiences-White female condition did not significantly differ from the no allyship cue condition (p = .357), and participants in the three endorsed ally conditions did not differ on perceptions that Dr. Evans recognized bias (all p values ⩾ .344).

Turning to shared reality, the endorsed ally-White female (p = .010), the endorsed ally-Black female (p < .001), and the endorsed ally-White male (p = .022) conditions each resulted in greater shared reality, relative to the no allyship cue condition. In contrast, the positive experiences-White female condition did not significantly differ from the no allyship cue condition (p = .086). The endorsed ally-Black female condition also resulted in significantly higher shared reality than the positive experiences-White female condition (p = .018). Thus, all three ally endorsed conditions promoted perceptions that the scientist recognized bias against women in STEM and shared a reality with participants.

Finally, as shown in Table 2, both recognizes bias and shared reality positively and significantly correlated with perceptions of allyship as well as with belonging and trust.⁴

Discussion

In Experiment 3, we recruited White women studying in STEM fields and manipulated both components of ally endorsement—the presence of a gender ingroup member sharing their positive experiences with the White male scientist and the explicit mention that the White male scientist was an ally of female students in STEM. We found that the three allyship endorsement conditions were each beneficial for promoting perceptions of allyship. However, while the endorsed ally-White male condition promoted allyship, this condition did not significantly differ from the no allyship cue condition on belonging and trust. In contrast, replicating Experiments 1 and 2, we found that both the endorsed ally-White female and endorsed ally-Black female conditions promoted belonging and trust, relative to the no allyship cue condition. Additionally, having the White female student convey her positive experiences with the scientist (without mentioning allyship) promoted significantly higher belonging and trust in computer science, relative to the no allyship cue control condition.

Taken together, these findings suggest that explicitly mentioning allyship is beneficial for promoting allyship perceptions, but may not be sufficient to promote identity-safety when the message comes from a gender outgroup member. Having a female student convey her positive experiences with a White male scientist without calling him an ally did encourage belonging and trust, indicating that this may be the more critical facet of ally endorsement. Of importance, having both components may be most beneficial. We found that relative to the no allyship cue condition, only the endorsed ally-Black female and endorsed ally-White female conditions significantly encouraged positive STEM beliefs. Thus, Experiment 3 finds that endorsement from a White or Black female—a gender ingroup member—is the most effective allyship cue to encourage perceptions a White male scientist is an ally, and also to promote belonging and trust, interest, and self-efficacy in computer science.

Experiment 3 also found that each of the endorsed ally conditions not only encouraged perceptions the White male scientist was an ally, but also promoted perceptions the White male scientist recognized bias and shared a reality with participants. Of note, although the endorsed ally-White male condition promoted shared reality and recognition of bias, this condition did little to promote belonging and trust, relative to the no allyship cue condition. In contrast, although the positive experiences-White female condition did not promote perceptions of shared reality or recognition of bias, this condition did encourage perceptions the scientist was an ally. One explanation for this finding is that although a gender ingroup member in the White male scientist’s lab can encourage allyship, an allyship cue is needed to encourage perceptions the scientist recognizes bias and shares a reality with women in STEM. Experiment 3 found that even among White women in STEM fields, ally endorsement from an ingroup member effectively encouraged perceived allyship, and this positively related to the perception the scientist recognized bias and shared a reality with women in STEM.

General Discussion

We examined whether an allyship cue—a gender ingroup member endorsing a White male scientist as an ally—could encourage positive STEM beliefs among White female undergraduates. In Experiment 1, we found that relative to viewing a White male scientist without an allyship cue, participants perceived greater allyship and reported greater anticipated belonging and trust when learning that the White male scientist had been endorsed as an ally by a White woman. In Experiment 2, we extended the findings of Experiment 1 and found that participants reported greater beliefs the computer science professor was an ally, greater belonging and trust as well as interest and self-efficacy in computer science when the professor was endorsed as an ally by one of his White female or Black female students, compared to a profile without an allyship cue.

In Experiment 3, we recruited White women in STEM fields and again found that an endorsement from a Black or White gender ingroup member encouraged perceptions of allyship and promoted positive STEM beliefs. Moreover, although an endorsement from a White male promoted perceptions the scientist was an ally, endorsement from a gender outgroup member failed to encourage identity-safety or positive STEM beliefs. At the same time, we also found that simply learning about a gender ingroup member’s positive experiences with the scientist also encouraged perceptions the scientist was an ally and promoted identity-safety, suggesting this component is the most important aspect of ally endorsement. Finally, in Experiment 3, we also found that the presence of the allyship cue encouraged perceptions the White male scientist recognized bias and shared a reality with participants, and these perceptions positively related to perceived allyship. Extending previous work examining allyship cues among Black women (Johnson & Pietri, 2020; Pietri, Johnson, & Ozgumus, 2018), the current experiments offer initial evidence that a White male scientist endorsed as an ally (by a White or Black woman) not only promotes perceptions of allyship among White female students, but also promotes identity-safety and positive STEM beliefs.

Practical and Theoretical Implications

The present experiments have important implications for the role of White men in STEM in promoting positive STEM beliefs among White women. White men are overrepresented in many STEM domains (National Science Foundation, 2019), and the requirement that students of all majors complete computer science courses has increased significantly in recent years (Kafka, 2020). Of note, the majority of participants in Experiments 1 and 2 were not established in STEM fields, and the present work highlights one practical tool White male computer scientists can utilize to spark interest in computer science classes and majors. The present investigation also offers ally endorsement as a potential alternative strategy for White women that may struggle to trust White men who self-espouse allyship as genuine and sincere (Cheng et al., 2019; Drury & Kaiser, 2014; Park et al., 2020; Wiley & Dunne, 2019). Moreover, both Black (Eaton et al., 2020) and White women (Moss-Racusin et al., 2012) are negatively stereotyped in STEM, and the presence of other similarly stereotyped individuals can shield White women from threats to identity-safety (Chaney et al., 2018). In line with past work, we found that allyship endorsement cues from a White or Black woman were equally effective.

The present work also has practical implications for Black women, who are also underrepresented in STEM environments (National Science Foundation, 2019). Several investigations have found that relative to race-relevant identity-safety cues, gender-relevant identity-safety cues are less effective for Black women (Johnson et al., 2019; Pietri, Johnson, & Ozgumus, 2018), and past work found that an ally endorsement from a Black woman encouraged Black women’s identity-safety (Johnson & Pietri, 2020). Thus, one implication of the present work is having a White male scientist receive an endorsement from a Black female student may potentially benefit both Black women and White women, although future investigations should test this hypothesis directly. Of importance, we were able to make and test predictions about the benefit of a Black female student’s endorsement because our female sample identified as White. Psychological research has been criticized for failing to acknowledge the consequences of relying on primarily White samples and treating White as the default racial identity (DeJesus et al., 2019; Flores Niemann, 2020). Consistent with recommendations to counter such shortcomings (see Roberts et al., 2020), the present work provides a theoretically driven rationale for our selected sample. Specifically, we recognized that our White female sample possessed both an identity that is high status and positively stereotyped in STEM (Eaton et al., 2020) as well as an identity that is devalued in STEM environments (Fiske et al., 2002), and predicted that White women would benefit from any gender ingroup member ally endorsement, even when the endorser was a racial outgroup member.

The current research also contributes to a growing body of literature demonstrating the benefits of allyship in environments where women are underrepresented. The present experiments imply that it may be useful for White women to identify evidence that an outgroup scientist has had positive contact with a gender ingroup member, to potentially promote perceptions of allyship and positive STEM beliefs. Critically, the present findings are based on the assumption that gender ingroup members can accurately identify sincere allies (Dovidio et al., 2002). Past work examining marginalized groups’ perceptions of allyship suggests that effective allyship not only acknowledges the challenges and injustices marginalized groups face, but also takes action to combat these challenges (Brown & Ostrove, 2013; Droogendyk et al., 2016; Ostrove & Brown, 2018). Moreover, research examining organizational diversity messages has found that expressed commitments in the absence of evidence or actions supporting them can backfire and be perceived as dishonest or inauthentic (Kroeper et al., 2020; Wilton et al., 2020; Windscheid et al., 2016). Thus, White male scientists that fail to acknowledge and act on women’s behalf will be unlikely to receive endorsements from women in STEM settings. Moreover, in real-world STEM environments, White male scientists that solicit or receive endorsements in the absence of compelling evidence that they support women in STEM will likely be perceived as engaging in performative or inauthentic allyship, and will only undermine feelings of identity-safety.

Limitations and Future Directions

Because the present work only recruited White women, it remains an empirical question to what extent allyship endorsement from a White woman serves as an effective allyship cue among women of other races and ethnicities. As previously noted, past work examining ally endorsement among Black women only examined ally endorsement from an individual matching both participants’ race and gender (Johnson & Pietri, 2020), making it unclear under what conditions an endorser sharing only their gender would cue allyship for Black women. However, although Black women do not readily perceive White women as allies of their group (Johnson & Pietri, 2020; Pietri, Johnson, & Ozgumus, 2018), research on the identity-threat perspective highlights the importance of context in signaling identity-safety (Major & O’Brien, 2005). Future work should examine if making salient that women (of all racial identities) face gender bias in STEM (O’Brien et al., 2020) can serve as one potential context where White women can act as effective ally endorsers for Black women.

Interestingly, in Experiment 3, although endorsement from the White male student encouraged perceptions the scientist was an ally, this endorsement failed to translate into positive STEM beliefs. One explanation for this finding is that even though the scientist was perceived as an ally, endorsement from a White male might have also signaled women’s underrepresentation in STEM (Murphy et al., 2007), undermining the efficacy of the allyship cue. Thus, an allyship cue from a gender ingroup member may be effective, in part, because it also signals the ingroup is represented in the domain (Milless et al., 2021). Supporting this explanation, we also found that learning about a White female’s positive experiences with the White male scientist (without any mention of allyship) promoted allyship as well as belonging and trust. Though the present experiment suggests that ally endorsement from a White man may be ineffective among White women, an endorsement from a male possessing a marginalized identity may be effective. Similarly stereotyped exemplars function as identity-safety cues for White women (Chaney et al., 2018); thus, Black or Latino men, who are similarly stereotyped in STEM, may also serve as effective endorsers for White women. Future work should manipulate the race and gender of the endorser to identify under what conditions ally endorsement can encourage positive STEM beliefs for White women.

Identifying strategies to aid White men in attracting female students to STEM fields is imperative for the recruitment and retention of women in STEM majors. The current research represents a critical first step by demonstrating that ally endorsement from either a White or Black woman is a beneficial strategy to cue allyship and help White male scientists promote identity-safety and positive STEM beliefs among White female students.

Supplemental Material

sj-docx-1-gpi-10.1177_13684302221080467 – Supplemental material for Ally endorsement: Exploring allyship cues to promote perceptions of allyship and positive STEM beliefs among White female students

Supplemental material, sj-docx-1-gpi-10.1177_13684302221080467 for Ally endorsement: Exploring allyship cues to promote perceptions of allyship and positive STEM beliefs among White female students by India R. Johnson and Evava S. Pietri in Group Processes & Intergroup Relations

Footnotes

Appendix

Acknowledgements

The first author would like to acknowledge her team of undergraduates—members of the SAP Lab—for their contributions to the present work.

Author note

Dr. Pietri is currently an Associate Professor at the University of Colorado – Boulder, and was at IUPUI when the present work was conducted.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iDs

India R. Johnson

Evava S. Pietri

Supplemental material

Supplemental material for this article is available online.

Notes

References

Akcinar

E. N.

Carr

P. B.

Walton

G. M.

(2011). Interactions with men and Whites matter too. Psychological Inquiry, 22, 247–251. https://doi.org/10.1080/1047840X.2011.625606

Ashburn-Nardo

(2018). What can allies do? In Colella

King

(Eds.), The Oxford handbook of workplace discrimination. (pp. 373–386) Oxford University Press.

Avery

D. R.

Hernandez

Hebl

M. R.

(2004). Who’s watching the race? Racial salience in recruitment advertising. Journal of Applied Social Psychology, 34, 146–161. https://doi.org/10.1111/j.1559-1816.2004.tb02541.x

Beyer

(1999). The accuracy of academic gender stereotypes. Sex Roles, 40, 787–813. https://doi.org/10.1023/A:1018864803330

Beyer

(2008). Predictors of female and male computer science students’ grades. Journal of Women and Minorities in Science and Engineering, 14, 377–409. https://doi.org/10.1615/JWomenMinorScienEng.v14.i4.30

Brown

K. T.

Ostrove

J. M.

(2013). What does it mean to be an ally?: The perception of allies from the perspective of people of color. Journal of Applied Social Psychology, 43, 2211–2222. https://doi.org/10.1111/jasp.12172

Chaney

K. E.

Sanchez

D. T.

Remedios

J. D.

(2018). We are in this together: How the presence of similarly stereotyped allies buffer against identity threat. Journal of Experimental Social Psychology, 79, 410–422. https://doi.org/10.1016/j.jesp.2018.09.005

Cheng

Traylor

King

(2019). Helping or hurting?: Understanding women’s perceptions of male allies. Personnel Assessment and Decisions, 5, Article 6. https://doi.org/10.25035/pad.2019.02.006

Cheryan

Drury

B. J.

Vichayapai

(2013). Enduring influence of stereotypical computer science role models on women’s academic aspirations. Psychology of Women Quarterly, 37, 72–79. https://doi.org/10.1177/0361684312459328

10.

Cheryan

Siy

J. O.

Vichayapai

Drury

B. J.

Kim

(2011). Do female and male role models who embody STEM stereotypes hinder women’s anticipated success in STEM? Social Psychological and Personality Science, 2, 656–664. https://doi.org/10.1177/1948550611405218

11.

Dasgupta

(2011). Ingroup experts and peers as social vaccines who inoculate the self-concept: The stereotype inoculation model. Psychological Inquiry, 22, 231–246. https://doi.org/10.1080/1047840X.2011.607313

12.

Dasgupta

Asgari

(2004). Seeing is believing: Exposure to counterstereotypic women leaders and its effect on the malleability of automatic gender stereotyping. Journal of Experimental Social Psychology, 40, 642–658. https://doi.org/10.1016/j.jesp.2004.02.003

13.

Dasgupta

Scircle

M. M.

Hunsinger

(2015). Female peers in small work groups enhance women’s motivation, verbal participation, and career aspirations in engineering. Proceedings of the National Academy of Sciences of the USA, 112, 4988–4993. https://doi.org/10.1073/pnas.1422822112

14.

Davies

P. G.

Spencer

S. J.

Steele

C. M.

(2005). Clearing the air: Identity safety moderates the effects of stereotype threat on women’s leadership aspirations. Journal of Personality and Social Psychology, 88, 276–287. https://doi.org/10.1037/0022-3514.88.2.276

15.

DeJesus

J. M.

Callanan

M. A.

Solis

Gelman

S. A.

(2019). Generic language in scientific communication. Proceedings of the National Academy of Sciences of the USA, 116, 18370–18377. https://doi.org/10.1073/pnas.1817706116

16.

Dennehy

T. C.

Dasgupta

(2017). Female peer mentors early in college increase women’s positive academic experiences and retention in engineering. Proceedings of the National Academy of Sciences of the USA, 114, 5964–5969. https://doi.org/10.1073/pnas.1613117114

17.

Dovidio

J. F.

Kawakami

Gaertner

S. L.

(2002). Implicit and explicit prejudice and interracial interaction. Journal of Personality and Social Psychology, 82, 62–68. https://doi.org/10.1037/0022-3514.82.1.62

18.

Droogendyk

Wright

S. C.

Lubensky

Louis

W. R.

(2016). Acting in solidarity: Cross-group contact between disadvantaged group members and advantaged group allies. Journal of Social Issues, 72, 315–334. https://doi.org/10.1111/josi.12168

19.

Drury

B. J.

Kaiser

C. R.

(2014). Allies against sexism: The role of men in confronting sexism. Journal of Social Issues, 70, 637–652. https://doi.org/10.1111/josi.12083

20.

Eaton

A. A.

Saunders

J. F.

Jacobson

R. K.

West

(2020). How gender and race stereotypes impact the advancement of scholars in STEM: Professors’ biased evaluations of physics and biology post-doctoral candidates. Sex Roles, 82, 127–141. https://doi.org/10.1007/s11199-019-01052-w

21.

Echterhoff

Higgins

E. T.

Levine

J. M.

(2009). Shared reality: Experiencing commonality with others’ inner states about the world. Perspectives on Psychological Science, 4, 496–521. https://doi.org/10.1111/j.1745-6924.2009.01161.x

22.

Faul

Erdfelder

Lang

A.-G.

Buchner

(2007). G*Power 3: A flexible statistical power analysis program for the social, behavioral, and biomedical sciences. Behavior Research Methods, 39, 175–191. https://doi.org/10.3758/BF03193146

23.

Findley-van Nostrand

Pollenz

R. S

. (2017). Evaluating psychosocial mechanisms underlying STEM persistence in undergraduates: Evidence of impact from a six-day pre-college engagement STEM academy program. CBE – Life Sciences Education, 16, 1–15. https://doi.org/10.1187/cbe.16-10-0294

24.

Fiske

S. T.

Cuddy

A. J. C.

Glick

. (2002). A model of (often mixed) stereotype content: Competence and warmth respectively follow from perceived status and competition. Journal of Personality and Social Psychology, 82(6), 878–902. https://doi.org/10.1037//0022-3514.82.6.878.

25.

Flores Niemann

. (2020). The obfuscation of the realities of women of color due to the false dichotomy phrasing of “women and minorities.” Women & Therapy, 43, 278–286. https://doi.org/10.1080/02703149.2020.1729475

26.

Good

Rattan

Dweck

C. S.

(2012). Why do women opt out? Sense of belonging and women’s representation in mathematics. Journal of Personality and Social Psychology, 102, 700–717. https://doi.org/10.1037/a0026659

27.

Higgins

E. T.

Rossignac-Milon

Echterhoff

(2021). Shared reality: From sharing-is-believing to merging minds. Current Directions in Psychological Science, 30, 103–110. https://doi.org/10.1177/0963721421992027

28.

Johnson

I. R.

Pietri

E. S.

(2020). An ally you say? Endorsing White women as allies to encourage perceptions of allyship and organizational identity-safety among Black women. Group Processes & Intergroup Relations. Advance online publication. https://doi.org/10.1177/1368430220975482

29.

Johnson

I. R.

Pietri

E. S.

Buck

D. M.

Daas

(2021). What’s in a pronoun: Exploring gender pronouns as an organizational identity-safety cue among sexual and gender minorities. Journal of Experimental Social Psychology, 97, Article 104194. https://doi.org/10.1016/j.jesp.2021.104194

30.

Johnson

I. R.

Pietri

E. S.

Fullilove

Mowrer

(2019). Exploring identity-safety cues and allyship among Black women students in STEM environments. Psychology of Women Quarterly, 43, 131–150. https://doi.org/10.1177/0361684319830926

31.

Kafka

A. C.

(2020, April 15). The discipline that is transforming higher ed: The computer-science boom is straining colleges. But it could save some, too. The Chronicle of Higher Education. https://www.chronicle.com/article/The-Discipline-That-Is/248536

32.

Kaiser

C. R.

Major

Jurcevic

Dover

T. L.

Brady

L. M.

Shapiro

J. R.

(2013). Presumed fair: Ironic effects of organizational diversity structures. Journal of Personality and Social Psychology, 104, 504–519. https://doi.org/10.1037/a0030838

33.

Kroeper

K. M.

Williams

H. E.

Murphy

M. C.

(2020). Counterfeit diversity: How strategically misrepresenting gender diversity dampens organizations’ perceived sincerity and elevates women’s identity threat concerns. Journal of Personality and Social Psychology. Advance online publication. https://doi.org/10.1037/pspi0000348

34.

Kutlaca

Radke

H. R.

Iyer

Becker

J. C.

(2020). Understanding allies’ participation in social change: A multiple perspectives approach. European Journal of Social Psychology, 50, 1248–1258. https://doi.org/10.1002/ejsp.2720

35.

LaCosse

Krusemark

Foltz

Plant

E. A.

(2021). Antiracism: Development and validation of a measure designed to identify White Americans who proactively fight to end discrimination toward Black Americans. Group Processes & Intergroup Relations. Advance online publication. https://doi.org/10.1177/13684302211048886

36.

Lent

R. W.

Brown

S. D.

Sheu

H.-B.

Schmidt

Brenner

B. R.

Gloster

C. S.

Wilkins

Schmidt

L. C.

Lyons

Treistman

(2005). Social cognitive predictors of academic interests and goals in engineering: Utility for women and students at historically Black universities. Journal of Counseling Psychology, 52, 84–92. https://doi.org/10.1037/0022-0167.52.1.84

37.

Lent

R. W.

Sheu

H.-B.

Miller

M. J.

Cusick

M. E.

Penn

L. T.

Truong

N. N.

(2018). Predictors of science, technology, engineering, and mathematics choice options: A meta-analytic path analysis of the social-cognitive choice model by gender and race/ethnicity. Journal of Counseling Psychology, 65, 17–35. https://doi.org/10.1037/cou0000243

38.

Lewis

K. L.

Stout

J. G.

Finkelstein

N. D.

Pollock

S. J.

Miyake

Cohen

G. L.

Ito

T. A.

(2017). Fitting in to move forward: Belonging, gender, and persistence in the physical sciences, technology, engineering, and mathematics (pSTEM). Psychology of Women Quarterly, 41, 420–436. https://doi.org/10.1177/0361684317720186

39.

Lewis

K. L.

Stout

J. G.

Pollock

S. J.

Finkelstein

N. D.

Ito

T. A.

(2016). Fitting in or opting out: A review of key social-psychological factors influencing a sense of belonging for women in physics. Physical Review Physics Education Research, 12, Article 020110. https://doi.org/10.1103/PhysRevPhysEducRes.12.020110

40.

Lindberg

S. M.

Hyde

Petersen

J. L.

Linn

M. C.

(2010). New trends in gender and mathematics performance: A meta-analysis. Psychological Bulletin, 136, 1123–1135. https://doi.org/10.1037/a0021276

41.

Major

O’Brien

L. T.

(2005). The social psychology of stigma. Annual Review of Psychology, 56, 393–421. https://doi.org/10.1146/annurev.psych.56.091103.070137

42.

Martiny

S. E.

Nikitin

(2019). Social identity threat in interpersonal relationships: Activating negative stereotypes decreases social approach motivation. Journal of Experimental Psychology: Applied, 25, 117–128. https://doi.org/10.1037/xap0000198

43.

Milless

K. L.

Wout

D. A.

Murphy

M. C.

(2021). Diversity or representation? Sufficient factors for Black Americans’ identity safety during interracial interactions. Cultural Diversity and Ethnic Minority Psychology. Advance online publication. https://doi.org/10.1037/cdp0000492

44.

Morgenroth

Ryan

M. K.

Peters

(2015). The motivational theory of role modeling: How role models influence role aspirants’ goals. Review of General Psychology, 19, 465–483. https://doi.org/10.1037/gpr0000059

45.

Moser

C. E.

Branscombe

N. R.

(2021). Male allies at work: Gender-equality supportive men reduce negative underrepresentation effects among women. Social Psychological and Personality Science. Advance online publication. https://doi.org/10.1177/19485506211033748

46.

Moss-Racusin

C. A.

Dovidio

J. F.

Brescoll

V. L.

Graham

M. J.

Handelsman

(2012). Science faculty’s subtle gender biases favor male students. Proceedings of the National Academy of Sciences of the USA, 109, 16474–16479. https://doi.org/10.1073/pnas.1211286109

47.

Murphy

M. C.

Steele

C. M.

Gross

J. J.

(2007). Signaling threat: How situational cues affect women in math, science, and engineering settings. Psychological Science, 18, 879–885. https://doi.org/10.1111/j.1467-9280.2007.01995.x

48.

Murphy

M. C.

Taylor

V. J.

(2012). The role of situational cues in signaling and maintaining stereotype threat. In Inzlicht

Schmader

(Eds.), Stereotype threat: Theory, process, and application (pp. 17–33). Oxford University Press.

49.

National Science Foundation. (2019). Women, minorities, and persons with disabilities in science and engineering (Special Report NSF 19-304). https://www.nsf.gov/statistics/wmpd

50.

O’Brien

L. T.

Garcia

D. M.

Blodorn

Adams

Hammer

Gravelin

(2020). An educational intervention to improve women’s academic STEM outcomes: Divergent effects on well-represented vs. underrepresented minority women. Cultural Diversity and Ethnic Minority Psychology, 26, 163–168. https://doi.org/10.1037/cdp0000289

51.

O’Brien

L. T.

Hitti

Shaffer

Camp

A. R. V.

Henry

Gilbert

P. N.

(2017). Improving girls’ sense of fit in science: Increasing the impact of role models. Social Psychological and Personality Science, 8, 301–309. https://doi.org/10.1177/1948550616671997

52.

Ostrove

J. M.

Brown

K. T.

(2018). Are allies who we think they are?: A comparative analysis. Journal of Applied Social Psychology, 48, 195–204. https://doi.org/10.1111/jasp.12502

53.

Park

J. W.

Vani

Saint-Hilaire

Kraus

M. W.

(2020). Beneficiary activists’ attitudes toward allies in social movements. PsyArXiv. https://doi.org/10.31234/osf.io/umzk2

54.

Pietri

E. S.

Drawbaugh

M. L.

Johnson

I. R.

Colvin

V. E.

(2020). Maybe he is relatable too: Encouraging women to identify with male scientists by highlighting bias against fathers. Psychology of Women Quarterly. Advance online publication. https://doi.org/10.1177/0361684320972118

55.

Pietri

E. S.

Johnson

I. R.

Majid

Chu

(2020). Seeing what’s possible: Videos are more effective than written portrayals for enhancing the relatability of scientists and promoting Black female students’ interest in STEM. Sex Roles. Advance online publication. https://doi.org/10.1007/s11199-020-01153-x

56.

Pietri

E. S.

Johnson

I. R.

Ozgumus

(2018). 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. Journal of Experimental Social Psychology, 74, 291–306. https://doi.org/10.1016/j.jesp.2017.06.021

57.

Pietri

E. S.

Johnson

I. R.

Ozgumus

Young

A. I.

(2018). Maybe she is relatable: Increasing women’s awareness of gender bias encourages their identification with women scientists. Psychology of Women Quarterly, 42, 192–219. https://doi.org/10.1177/0361684317752643

58.

Purdie-Vaughns

Steele

C. M.

Davies

P. G.

Diltmann

Crosby

J. R.

(2008). Social identity contingencies: How diversity cues signal threat or safety for African Americans in mainstream institutions. Journal of Personality and Social Psychology, 94, 615–630. https://doi.org/10.1037/0022-3514.94.4.615

59.

Radke

H. R. M.

Hornsey

M. J.

Barlow

F. K.

(2018). Changing versus protecting the status quo: Why men and women engage in different types of action on behalf of women. Sex Roles, 79, 505–518. https://doi.org/10.1007/s11199-017-0884-2

60.

Radke

H. R.

Kutlaca

Siem

Wright

S. C.

Becker

J. C.

(2020). Beyond allyship: Motivations for advantaged group members to engage in action for disadvantaged groups. Personality and Social Psychology Review, 24, 291–315. https://doi.org/10.1177/1088868320918698

61.

Roberts

S. O.

Bareket-Shavit

Dollins

F. A.

Goldie

P. D.

Mortenson

(2020). Racial inequality in psychological research: Trends of the past and recommendations for the future. Perspectives on Psychological Science, 15, 1295–1309. https://doi.org/10.1177/1745691620927709

62.

Rossignac-Milon

Bolger

Zee

K. S.

Boothby

E. J.

Higgins

E. T.

(2021). Merged minds: Generalized shared reality in dyadic relationships. Journal of Personality and Social Psychology, 120, 882–911. https://doi.org/10.1037/pspi0000266

63.

Steele

C. M.

Spencer

S. J.

Aronson

(2002). Contending with group image: The psychology of stereotype and social identity threat. In Zanna

M. P.

(Ed.), Advances in experimental social psychology (Vol. 34, pp. 379–440). Academic Press.

64.

Stout

J. G.

Dasgupta

Hunsinger

McManus

M. A.

(2011). STEMing the tide: Using ingroup experts to inoculate women’s self-concept in science, technology, engineering, and mathematics (STEM). Journal of Personality and Social Psychology, 100, 255–270. https://doi.org/10.1037/a0021385

65.

Tam

M.-Y. S.

Bassett

G. W.

Jr. (2006). The gender gap in information technology. In Bystydzienski

J. M.

Bird

S. R.

(Eds.), Removing barriers: Women in academic science, technology, engineering, and mathematics (pp. 119–133). Indiana University Press.

66.

Thorne

K. M.

Jones

M. K.

Davis

T. M.

Settles

I. H.

(2021). The significance of race in cross-racial mentoring of faculty of color. Translational Issues in Psychological Science. Advance online publication. https://doi.org/10.1037/tps0000286

67.

U.S. Department of Education. (2019). Status and trends in the education of racial and ethnic groups (Indicator 26: STEM degrees). National Center for Education Statistics. https://nces.ed.gov/programs/raceindicators/indicator_reg.asp

68.

U.S. Department of Labor. (2019). Women in the labor force: A databook (Report 1084). Bureau of Labor Statistics. https://www.bls.gov/opub/reports/womens-databook/2019/home.htm

69.

Walton

G. M.

Cohen

G. L.

(2007). A question of belonging: Race, social fit, and achievement. Journal of Personality and Social Psychology, 92, 82–96. https://doi.org/10.1037/0022-3514.92.1.82

70.

Walton

G. M.

Logel

Peach

J. M.

Spencer

S. J.

Zanna

M. P.

(2015). Two brief interventions to mitigate a “chilly climate” transform women’s experience, relationships, and achievement in engineering. Journal of Educational Psychology, 107, 468–485. https://doi.org/10.1037/a0037461

71.

Wiley

Dunne

(2019). Comrades in the struggle? Feminist women prefer male allies who offer autonomy- not dependency-oriented help. Sex Roles, 80, 656–666. https://doi.org/10.1007/s11199-018-0970-0

72.

Wilton

L. S.

Bell

A. N.

Vahradyan

Kaiser

C. R.

(2020). Show don’t tell: Diversity dishonesty harms racial/ethnic minorities at work. Personality and Social Psychology Bulletin, 46, 1171–1185. https://doi.org/10.1177/0146167219897149

73.

Windscheid

Bowes-Sperry

Kidder

D. L.

Cheung

H. K.

Morner

Lievens

(2016). Actions speak louder than words: Outsiders’ perceptions of diversity mixed messages. Journal of Applied Psychology, 101, 1329–1341. https://doi.org/10.1037/apl0000107

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.16 MB