Gendered Racism,Coping,Identity Centrality,and African American College Women’s Psychological Distress

Gendered Racism

Some researchers have conceptualized the intersections of racism and sexism experienced by African American women as gendered racism (Essed, 1991; Thomas, Witherspoon, & Speight, 2008). Researchers have tried to empirically explore African American women’s experiences with the intersection of racism and sexism, or gendered racism, in a variety of ways. Drawing on Black feminist scholarship, interdisciplinary researchers have theorized an intersectionality framework to study African American women’s experiences. According to legal scholar Kimberle Crenshaw (1989) who coined the term intersectionality, African American women could experience: (a) similar experiences (single-axis approach), that is, racism similar to African American men and sexism similar to White women; (b) double jeopardy (additive approach) or multiplicative effects (interactional approach), that is, racism and sexism are separate yet accumulative experiences of oppression; and (c) specific oppression (intersectional approach) that is based on the intersection of race and gender. In terms of the latter, scholars have also noted how these intersectional experiences can differ for subgroups of women of color (Buchanan, 2005; Cole, 2009; Thomas et al., 2008).

Much of the extant literature on African American women’s experiences of oppression has been framed around the single-axis or additive approach. For example, two studies (Moradi & Subich, 2003; Szymanski & Stewart, 2010) found that greater sexist and racist discrimination were significantly related to higher levels of psychological distress, at the bivariate level, in a sample of African American women. Both studies found that racist discrimination and the interaction of racist and sexist discrimination did not significantly predict psychological distress, but sexist discrimination did. Although these findings add to our understanding of the additive and interactive effects of racist and sexist experiences, this approach still measures the experiences of racism and sexism separately. Such methods do not meaningfully capture the ways that the intersection of racism and sexism, or gendered racism, uniquely affects psychological distress among African American women.

More recently, psychology researchers have begun to theorize and conduct empirical research on the intersections of race and gender (e.g., Cole, 2009; Thomas et al., 2008). The majority of empirical studies focused on African American women’s experiences with racism and sexism have found that these intersecting forms of oppression are related to poor mental health. In studies that mostly involved community samples of African American women, greater experiences of gendered racism have been related to higher levels of psychological distress (King, 2003; Lewis & Neville, 2015; Thomas et al., 2008), greater depressive symptoms (Carr, Szymanski, Taha, West, & Kaslow, 2014), more post-traumatic stress symptoms (Woods, Buchanan, & Settles, 2009), and lower social self-esteem (King, 2003).

Focus on College Women

We extend the previous research by focusing on college women, because they may be at risk of experiencing gendered racism in their social and collegiate relations, in the academic environment, and in the classroom (Lewis, Mendenhall, Harwood, & Huntt, 2013; Watkins, LaBarrie, & Appio, 2010). In addition, although African American college women typically have greater socioeconomic resources than peers who do not attend college, research suggests that higher socioeconomic status women are not immune to the negative psychological effects of oppression (Jones & Shorter-Gooden, 2003). Given their developmental level, college-age women may also have less experience and practice in navigating discrimination and other life challenges, which may influence their ability to limit or avoid exposure to these negative stimuli and to use more mature coping responses (Charles, 2010). The unique college context may also influence how African American women react to gendered racism by limiting (e.g., difficulty in changing majors or professors) and/or facilitating (e.g., offering students opportunities for collective responses to oppression) certain coping responses.

Coping With Discrimination as Mediators

Given the long history of research on the links between coping styles and mental health outcomes in psychology, it is surprising that very little research has focused on the relation between discrimination and coping variables. Few researchers have examined the potential mediating functions that coping may play in better understanding the link between discrimination and mental health outcomes among individuals from marginalized groups (Pascoe & Smart Richman, 2009). There are several theoretical models and scholarly assertions for why coping might partially mediate the discrimination–mental health outcome links (cf. Clark et al., 1999; Harrell, 2000; Hatzenbuehler, 2009; Szymanski & Obiri, 2011). These theoretical tenets posit that (a) African American women face increased stress due to stigma based on the intersections of race and gender, (b) this stress creates the need to implement coping strategies to deal with these minority-based stressors, and (c) these coping responses partially mediate or explain the link between gendered racism and poorer mental health outcomes. Disengagement coping strategies (e.g., passive, avoidant, and maladaptive coping) are assumed to be related to more psychological distress, and engagement coping strategies (e.g., proactive, problem solving, and adaptive coping) are assumed to be related to less psychological distress.

Supporting these assertions, Utsey, Ponterotto, Reynolds, and Cancelli (2000) found that race-related stress was associated with an increased use of problem solving, avoidant, and support-seeking coping styles among African Americans. Racial discrimination stress predicted the use of Africultural, culturally specific coping (i.e., emotional debriefing, communalistic coping, and spiritually centered coping), after controlling for general coping styles, among low-income African American adolescents (Gaylord-Harden & Cunningham, 2009). In terms of mediation, a study targeting African American adolescents found that avoidant coping strategies explained the relation between racial discrimination and depressive symptoms (Seaton, Upton, Gilbert, & Volpe, 2014). Likewise, Szymanski and Obiri (2011) found that experiences of racial discrimination were related to an increased use of both positive and negative religious coping styles among an African American sample, but only negative religious coping styles mediated the racism and psychological distress link. Thomas et al. (2008) found that culturally specific emotional debriefing (e.g., thinking about other things and not thinking about the stressor) partially mediated the relation between gendered racism and psychological distress among African American women. However, they found no mediational effects for communalistic, ritualistic, or spiritually centered Africultural coping styles.

With a few exceptions (e.g., Lewis et al., 2013; Thomas et al., 2008), most of the literature on coping among African American women has focused on racism and sexism separately; thus, our study advances the coping literature by paying attention to coping strategies used in response to the unique, intersectional experiences of gendered racism. Furthermore, most of the literature on coping with discrimination among African Americans has focused on general coping responses used for any potential stressful event, or culturally specific coping styles. However, three studies found that African American adults cope differently depending on whether the stressor is a general life stressor versus a racial stressor (Brown, Phillips, Abdullah, Vinson, & Robertson, 2011; Hoggard, Byrd, & Sellers, 2012). Yet, very little research has examined coping strategies unique to discriminatory experiences (Miller & Kaiser, 2001). A notable exception is Wei, Alvarez, Ku, Russell, and Bonett’s (2010) study that found that racial and ethnic minority persons frequently use five strategies for coping with discrimination: detachment, internalization, drugs and alcohol, resistance, and education/advocacy. Wei et al. (2010) also found that coping strategies, specific to discrimination, uniquely predicted depressive symptoms, self-esteem, and life satisfaction, after controlling for general coping styles. These findings underscore the need to examine discrimination-specific coping strategies. Thus, we examined how disengagement coping strategies (i.e., detachment, internalization, drugs and alcohol) and engagement coping strategies (i.e., resistance and education/advocacy), employed to deal with discrimination, might partially mediate the link between gendered racism and psychological distress.

Coping with gendered racism via detachment is the process of disengaging from problem solving and distancing oneself from social support (Wei et al., 2010). Lewis et al. (2013) found that African American women reported using detachment coping strategies to deal with subtle forms of gendered racism, especially when the situation was outside their control. Detachment may provide a means of social and cognitive avoidance of these often unpredictable stressors (Miller & Kaiser, 2001; Seaton et al., 2014). In addition, detachment may protect African American women from further assaults, rejection, and misunderstanding from others about the race-related stressor (Miller & Kaiser, 2001). However, detachment may fuel psychological distress due to decreases in emotional support from others, and feeling at a loss for how to solve the problem.

Coping with gendered racism via internalization or self-blame is the process by which a person attributes the cause for the discriminatory events to herself (Wei et al., 2010). Because many gendered racist experiences can be ambiguous, it may leave an African American woman wondering if the action was due to prejudice or other factors; thereby increasing the likelihood that she blames herself for the event (Miller & Kaiser, 2001). It may be that White persons discredit the prejudiced experience, leaving an African American woman to internalize the experience, rather than attribute it to the perpetrator (Kaiser & Miller, 2001). Carr et al. (2014) found that greater experiences of racial discrimination and sexual objectification were related to greater coping with discrimination via internalization, which in turn was related to more depressive symptoms among a low-income clinical sample of African American women. An African American woman may also cope with gendered racism by using drugs and/or alcohol to self-medicate, numb emotions, and decrease feelings of anger and distress (at least in the short-term) related to the discriminatory experience (Szymanski, Moffitt, & Carr, 2011). In a longitudinal study, Gibbons et al. (2010) found that racial discrimination among African American parents and adolescents was linked to increased substance use, and this link was mediated by feelings of anger and hostility.

Resistance is the process of confronting the perpetrators of a discriminatory behavior, while coping via education/advocacy is the process of increasing self and other’s awareness of discrimination and implementing advocacy efforts to fight discrimination at micro- and macro-levels (Wei et al., 2010). The college context is unique in that it often provides diversity-related educational programming (courses and campus events), student groups aimed at increasing awareness of sociocultural identities and experiences of oppression, and opportunities for community engagement and social action. Qualitative studies of community school and college students have found that African American women often use resistance and education/advocacy strategies as a way to proactively cope with gendered racism and avert its potentially negative effects (Everett, Hall, & Hamilton-Mason, 2010; Lewis et al., 2013; Shorter-Gooden, 2004). These strategies may help African American women (a) manage the anger associated with gendered racism, (b) counteract the devaluation associated with gendered racism, (c) contextualize rather than internalize oppression, (d) motivate perpetrators to change their behavior, (e) motivate others to take action, and/or (f) feel empowered that they can do something about gendered racism (Brondolo, ver Halen, Pencille, Beatty, & Contrada, 2009; Szymanski, 2012).

In terms of resistance and education/advocacy, two studies found that increased experiences of racism were related to greater involvement in African American activism (Szymanski, 2012; Szymanski & Lewis, 2015). Other studies have found that African American women who reported “doing something about” discrimination had lower blood pressure levels (Krieger & Sidney, 1996) and were less likely to have a diagnosis of hypertension (Krieger, 1990) than those who did nothing about it. Finally, Hyers (2007) found that women who coped with racism or sexism via confrontation were more likely to feel that they had been efficacious and were less likely to engage in rumination than women who did not confront their perpetrators. Given this small body of research, the need to examine the various methods for coping with discrimination in the gendered racism–distress links is warranted.

Identity Centrality as a Moderator in Predicting Psychological Distress

African American women’s gendered racial identity centrality refers to the degree to which the intersection of one’s race and gender form an important part of one’s self-concept (Leach et al., 2008). Identity centrality is a key aspect of larger theories on racial (Cross, 1991), feminist/womanist (Brown, 1989; Downing & Roush, 1986; Ossana, Helms, & Leonard, 1992), and social (Tajfel & Turner, 1979) identity. It is considered to be stable across situations and contexts (Sellers et al., 1998). A recent meta-analysis of African Americans found that greater levels of racial centrality and other highly related constructs (e.g., Afrocentricity, private regard) were related to lower levels of psychological distress (Lee & Ahn, 2013). A review of gender identity research among adolescents suggests that gender centrality also plays a role in positive psychosocial outcomes (Perry & Pauletti, 2011).

Gendered racial identity centrality can influence the degree to which gendered racism is perceived as threatening to one’s individual self and collective self-concept; theory suggests it is a protective factor when faced with discrimination (Neblett, Rivas-Drake, & Umaña-Taylor, 2012; Sellers & Shelton, 2003; Sellers, Caldwell, Schmeelk-Cone, & Zimmerman, 2003). Drawing attention to one’s race and gender may serve to increase an African American woman’s awareness that she may be targeted for gendered racism and may help her garner the strength she needs to protect herself from its harmful effects (Brondolo et al., 2009). In a similar vein, high identity centrality often encompasses a more serious understanding and commitment to African American women’s experiences and issues, which may lead to blaming the perpetrators of gendered racism, thereby lessening the potential negative influences of discrimination on mental health (Cross & Vandiver, 2001). Furthermore, African American women with high levels of identity centrality might be less likely to internalize the negative messages about being an African American woman that are often conveyed through acts of gendered racism; thereby protecting them from its negative effects (Neblett et al., 2012). When faced with gendered racism, African American women with high levels of identity centrality may be able to feel good about themselves by focusing on the collective strength and positive aspects of their group; thereby buffering themselves from negative mental health consequences (Sellers & Shelton, 2003).

Meta-analytic studies of in-group identification reveal mixed results, with some finding a buffering effect, others finding no effect, and a smaller number finding an exacerbating effect on the link between discrimination and psychosocial outcomes (Pascoe & Smart Richman, 2009). For example, Sellers et al. (2003) found that racial centrality buffered the negative impact of racial discrimination on the psychological distress of African American young adults. Similarly, Seaton (2009) found that high racial identity centrality, high public regard (i.e., beliefs that others think positively about African Americans), and high private regard (i.e., positive feelings about African Americans and about being an African American) buffered the negative effect of racial discrimination on depressive symptoms among a sample of African American adolescents. Feminist identification also has been shown to protect women against negative effects of sexism on various mental health outcomes, including psychological distress and disordered eating (for a review, see Szymanski & Moffitt, 2012). Finally, DeBlaere and Bertsch (2013) found that womanist identification, which is inclusive of racial/ethnic, gender, and sexual orientation identities, played a buffering role in the link between sexism and psychological distress among African American sexual minority women.

Contrary to the buffering effect hypothesis, Burrow and Ong (2010) found that racial identity centrality exacerbated the negative effects of racial discrimination on depression and negative affect among African American doctoral students and graduates of doctoral programs. McCoy and Major (2003) found that lower gender centrality buffered the sexist discrimination and psychological distress links among women. Finally, Pascoe and Smart Richman (2009) in a meta-analysis reported that the majority of studies found no effect for in-group identification; however, these findings are likely to be influenced by the notorious difficulty of detecting interaction effects in nonexperimental studies, due to low sample sizes and a lack of power needed to detect the small effect sizes typically found in social science research (McClelland & Judd, 1993).

The research on identity centrality is limited in understanding individuals with multiple minority identities, by its focus on only one discrimination dimension, and a single identity factor. The conflicting findings related to in-group identification may be due, in part, to the use of measures that do not adequately assess the discrimination experiences and cultural identities of its participants (e.g., African American women). For example, in a qualitative study of young adult African American women, gendered racial identity had greater salience in participants’ lives than did the separate constructs of gender or racial identity (Thomas, Hacker, & Hoxha, 2011). There is no previous research that focuses on the intersections of race and gender in African American women’s identity centrality, as well their potential interactive role with gendered racism, in the links to mental health outcomes. In addition, research has rarely focused on African American female samples. This study advances existing literature by focusing on the intersectional nature of both discrimination experiences and definitions of self-concept that are important to African American women. Given that race-based, gender-based, and social psychological theories on in-group identification articulate a protective function of identity centrality and that more studies in Pascoe and Smart Richman’s (2009) review found support for a buffering, rather than exacerbating, role in the discrimination and mental health links, we hypothesized that gendered racial identity centrality would play a protective role in the link between gendered racism and African American women’s psychological distress.

A Moderated Mediation Model

Gendered racial identity centrality may also influence the implementation of coping strategies when faced with experiences of gendered racism (Brondolo et al., 2009; Sellers et al., 2003; Sellers & Shelton, 2003). For example, African American women with high levels of identity centrality may have a stronger sense that they can respond effectively to discrimination and have more adaptive, effective, and varied coping responses when faced with high levels of gendered racism, because they have more experience dealing with it (Sellers et al., 2003). African American women with high identity centrality may also be more likely to think about race and gender discrimination and thereby may engage in coping efforts aimed at actively dealing with the source of stress (i.e., gendered racism); they may be less likely to disengage from the stressor when faced with gendered racism. Thus, identity centrality should moderate the direct relations between gendered racism and both disengagement and engagement coping strategies. In addition, a pattern of moderated mediation (Hayes, 2013) is likely to be present. That is, identity centrality might also qualify the indirect relation between gendered racism and psychological distress. However, there is a lack of research targeting both in-group identification and coping strategies; the theoretical assertions have not yet been empirically tested.

The Current Study

As illustrated in Figure 1, our hypothesized conceptual model consists of examining potential coping mediators in the link between gendered racism and psychological distress among African American women. In addition, we examined the gendered racial identity centrality as a moderator in the (a) gendered racism and psychological distress link, (b) gendered racism and coping links, and (c) indirect relation between gendered racism and psychological distress. Our specific hypotheses were:

Disengagement and engagement strategies for coping with discrimination will partially mediate the gendered racism–psychological distress link. That is, higher levels of gendered racism will be positively related to the use of more coping via detachment, internalization, and drugs and alcohol, which in turn will be related to more psychological distress. In addition, higher levels of gendered racism will be positively related to more coping via resistance and education/advocacy, which in turn will be related to less psychological distress.

OPEN IN VIEWER

Figure 1.

Hypothesized model predicting mental health outcome. Dashed line indicates conditional indirect effect.

Participants

The initial sample comprised 250 participants who completed an online survey. Eleven men, 1 Asian American, 13 non-students, 7 participants who left at least one measure completely blank, and 6 participants who failed two or more of the 3 validity check items (e.g., “For this item, click the button labeled red”), were eliminated from the data set. The final sample was 212 women and 98% self-identified as African American/Black and 2% as biracial. Participants ranged in age from 18 to 47 years (M = 19.50, SD = 3.87). All participants were currently enrolled as students in a higher education institution, with 61% being first-year undergraduates, 20% sophomores, 10% juniors, 5% seniors, 3% graduate students, and 1% other. Participants self-identified as being a member of the following social class categories: 1% wealthy, 14% upper middle class, 47% middle class, 31% working class, and 7% poor. Participants resided in the South (74%), Northeast (11%), Midwest (8%), and West (7%).

Procedure

Participants were recruited through a department of psychology’s human research pool at a large Southeastern predominantly White public university and through an e-mail announcement of the study that was sent to the listserv owner/contact person of a variety of university/college multicultural or African American/Black student centers, African American student/faculty groups, and African American Studies programs. The e-mail asked recipients to forward our recruitment e-mail to their list serves. The e-mail announcement was sent to individuals on the website listed as either “contact person” or “listserv owner.”

Participants completed an online web-based survey located on a secure firewall protected server accessed via a hypertext link. After participants read the informed consent, they indicated consent to take the survey by clicking a button, which directed them to the next page containing the survey. The survey included a demographic questionnaire and the aforementioned measures, which were randomly ordered. As an incentive, participants were told they were eligible to enter a participant raffle awarding gift certificates to an online merchant. In addition, eligible participants could receive course credit for their undergraduate course. We used a separate raffle/course credit database so that there was no way to connect a person’s online course credit submission with her submitted survey.

Measures

Preliminary Analyses and Descriptive Data

Analysis of missing data patterns for the 212 participants in our final sample indicated that .43%, or less than a half of a percentage of all items for all participants/cases were missing, and 41% of the items were not missing data for any participant/case. Considering individual cases, 82% of participants had no missing data. Finally, no item had 2% or more of missing values. Given the very small amount of missing data, we used available case analyses procedures, wherein mean scale scores are calculated without substitution or imputation of values, which produces similar results to multiple imputation methods (Parent, 2013).

Data met guidelines for univariate normality (i.e., skewness < 3, kurtosis < 10; Weston & Gore, 2006). Preliminary correlational analyses between the demographic variables of age, self-reported socioeconomic status, and geographic region (South vs. Other) and the variables in our study revealed a few significant (p < .05) associations. More specifically, older age was related to greater identity centrality (r = .18) and more coping with discrimination via education/advocacy (r = .17). Higher self-reported socioeconomic status was related to less coping with discrimination via detachment (r = −.16) and lower psychological distress levels (r = −.20). However, we did not include these variables as covariates in the subsequent analyses reported below because the utility of controlling for demographic variables in multiple regression analyses has been questioned by scholars. Because variables such as self-reported socioeconomic status are ordinal variables, they are often not normally distributed and can behave poorly as covariates. In addition, from a theoretical standpoint, problems occur when a demographic variable shares meaningful variance with another variable of theoretical interest (Little, An, Johanns, & Giordani, 2000; Miller & Chapman, 2001; Thompson, 1992). Furthermore, analyses including them as covariates did not change the findings in any meaningful way (i.e., significance levels and the directions of the relations/beta weights were similar for analyses using covariates vs. not using them).

Descriptive statistics and bivariate correlations among all study variables are shown in Table 1. At the bivariate level, gendered racism was significantly and positively related to coping with discrimination via detachment, internalization, resistance, and education/advocacy but not with coping with discrimination via drugs and alcohol. Gendered racism was not related to identity centrality. Gendered racism and disengagement coping (detachment and internalization) yielded significant positive relations with psychological distress, but identity centrality and coping using drugs and alcohol and engagement coping (resistance and education/advocacy) did not. Examination of multicollinearity indexes for all analyses indicated that multicollinearity was not a problem (i.e., absolute value correlations < .90, variance inflation factors < 10; tolerance values > .20 and condition indexes < 30; Field, 2013; Tabachnick & Fidell, 2001).

OPEN IN VIEWER

Table 1.

Means, Standard Deviations, and Correlations for All Study Variables.

Variable	Possible range	M	SD	1	2	3	4	5	6	7
1. Gendered racism	0–4	1.92	1.06	—
2. Coping via detachment	1–6	2.36	0.86	.17*	—
3. Coping via internalization	1–6	2.55	1.14	.17*	.29*	—
4. Coping via drugs and alcohol	1–6	1.65	0.93	−.02	.09	.07	—
5. Coping via resistance	1–6	2.53	0.99	.22*	.02	.09	.39*	—
6. Coping via education/advocacy	1–6	3.24	1.33	.32*	.04	.12	.10	.25*	—
7. Identity centrality	1–7	5.89	1.21	.13	.01	.11	−.08	.11	.34*	—
8. Psychological distress	1–4	1.82	0.52	.31*	.40*	.29*	.03	.08	.12	.11

*p < .05.

Mediation Analyses

We used the PROCESS macro (Hayes, 2013; Model 4) to test the mediation model described in Hypothesis 1. Current recommendations for testing indirect effects, which do not require both the component paths of the indirect effect to be statistically significant, were followed (Mallinckrodt, Abraham, Wei, & Russell, 2006). We used bootstrapping analyses with 1,000 bootstrapping resamples to produce 95% confidence intervals for the indirect effect, because it does not assume normality in the distribution of the mediated effect and can be applied with confidence to smaller samples (cf. Mallinckrodt et al., 2006; Preacher & Hayes, 2008). If the confidence interval does not contain zero, one can conclude that mediation is significant and meaningful (Preacher & Hayes, 2008).

The results of our mediation model are shown in Figure 2. The test of mediation using bootstrapping analyses revealed that both coping via detachment (mean indirect [unstandardized] effect = .03; SE = .01, 95% CI [.005, .050], β = .05) and internalization (mean indirect [unstandardized] effect = .01; SE = .01, 95% CI [.001, .033], β = .02) mediated the gendered racism–psychological distress links. Contrary to our hypothesis, no mediated effects were found for coping using drugs and alcohol (mean indirect [unstandardized] effect = .00; SE = .00, 95% CI [−.006, .0.005], β = .00), resistance (mean indirect [unstandardized] effect = .00; SE = .01, 95% CI [−.016, .020], β = .00), and education/advocacy (mean indirect [unstandardized] effect = .00; SE = .01, 95% CI [−.019, .022], β = .00). Finally, the variables in the model accounted for 25% of the variance in psychological distress scores.OPEN IN VIEWER

Moderator and Moderated Mediation Analyses

To test Hypotheses 2–4, we again used PROCESS (Hayes, 2013; Model 8). Results of these moderated analyses are shown in Table 2. Contrary to Hypothesis 2, results indicated that identity centrality did not moderate the gendered racism–psychological distress link. Consistent with Hypothesis 3, results indicated that identity centrality (β = .16, R² change = .025, significant F change = .02) moderated the relations between gendered racism and detachment coping. Follow-up simple slopes analysis revealed that gendered racism did not predict coping via detachment for women with low (SD = −1) identity centrality, B = −.04, t = −.401, p = .69; whereas gendered racism predicted psychological distress for women with high (SD = +1) identity centrality, B = .26, t = 3.422, p = .001, and at the mean, B = .12, t = 2.09, p = .04 (see Figure 3). Contrary to Hypothesis 3, identity centrality did not moderate the links between gendered racism and coping, using internalization, drugs and alcohol, resistance, and education/advocacy. Because a significant interaction between the predictor (gendered racism) and the moderator (identity centrality) on the mediator outcome variable is needed to establish moderated mediation (Hayes, 2013), there was no support for the associated conditional indirect effects of gendered racism on psychological distress through coping via internalization, drugs and alcohol, resistance, and education/advocacy as proposed in Hypothesis 4.

OPEN IN VIEWER

Table 2.

Test of Identity Centrality as a Moderator of the Predictor–Mediator and Predictor–Criterion Links.

Predictor Variable	Criterion Variable	B	β	t	R 2	F	df
	Criterion: detachment
Gendered racism		.12	.14	2.09*	.05	3.91*	3, 208
Identity centrality		.00	.00	0.02
Gendered Racism × Identity Centrality		.13	.16	2.33*
	Criterion: internalization
Gendered racism		.19	.18	2.58*	.05	3.46*	3, 208
Identity centrality		.08	.08	1.18
Gendered Racism × Identity Centrality		−.11	−.10	−1.46
	Criterion: drugs and alcohol
Gendered racism		−.01	−.11	−0.23	.01	.49	3, 208
Identity centrality		−.06	−.07	−1.06
Gendered Racism × Identity Centrality		.02	.03	0.39
	Criterion: resistance
Gendered racism		.20	.22	3.13*	.06	4.26*	3, 208
Identity centrality		.07	.09	1.27
Gendered Racism × Identity Centrality		−.01	−.01	−.16
	Criterion: education/advocacy
Gendered racism		.34	.27	4.21*	.20	17.65*	3, 208
Identity centrality		.35	.32	5.03*
Gendered Racism × Identity Centrality		.11	.09	1.42
	Criterion: psychological distress
Gendered racism		.11	.23	3.47*	.25	8.46	8, 203
Detachment		.19	.32	4.92*
Internalization		.07	.15	2.24*
Drugs and alcohol		.00	.01	.07
Resistance		.00	.00	.02
Education/advocacy		−.00	−.08	−.12
Identity centrality		.03	.06	.93
Gendered Racism × Identity Centrality		−.02	−.04	−.57

Note. β and t reflects values from the final regression equation. df = degrees of freedom.

*p < .05.

OPEN IN VIEWER

Figure 3.

Interaction of gendered racism and identity centrality on coping with discrimination via detachment.

However, supporting Hypothesis 4, results using 1,000 bootstrap samples for the moderated mediation analyses revealed that the indirect effect of gendered racism on psychological distress through detachment coping was moderated by identity centrality (Index of Moderation Mediation = .025, SE [boot] = .013, 95% CI [.002, .052]). The indirect path was not significant when identity centrality was low (SD = −1; B = −.01; boot estimate = .022; 95% CI [−.054, .037]) but was significant when identity centrality was high (SD = +1; B = .05; boot estimate = .015; 95% CI [.026, .084]) or at the mean (B = .02; boot estimate = .012; 95% CI [.002, .048]).

Limitations and Directions for Future Research

The dearth of intersectional measures for use with African American women is one significant limitation in the extant research literature generally and in our study specifically. Although the Racialized Sexual Harassment Scale (Buchanan, 2005) performed well in our study, this scale was not developed specifically for use with African American women to assess gendered racism. Future studies should use a gendered racism measure, such as the newly developed Gendered Racial Microaggressions Scale (Lewis & Neville, 2015), to more adequately assess the intersections of racism and sexism. In addition, using an African American women’s identity scale, rather than the more general identity centrality scale, might have better captured the intersections of gendered racial identity. Last, although the use of a coping-specific measure, such as the CDS (Wei et al., 2010), was useful in this study, this scale was not designed to capture the intersecting experiences of racism and sexism that African American women experience. Thus, there might be some coping strategies specific to African American women that were not captured in the scale we used. For example, previous qualitative work with African American women (e.g., Everett et al., 2010; Lewis et al., 2013; Shorter-Gooden, 2004) has found that relying on social support is a very important form of coping for African American women, but this was not measured in the CDS. In addition, there might be more nuanced coping strategies in response to gendered racism that have not been uncovered in the extant literature.

Another limitation of our study was the cross-sectional study design, which precludes inferences about causality. For example, although we found that gendered racism predicted detachment coping, which then predicted psychological distress for African American women with high identity centrality, but not for women with low identity centrality, we do not know the exact nature and direction of these findings. It is possible that greater psychological distress predicts detachment coping, which in turn predicts greater self-reported experiences of gendered racism. In addition, although it could be interpreted that low identity centrality serves a buffering function in the gendered racism–distress links, this conclusion seems to miss the fundamental way that identity centrality works. Namely, lower identity centrality might prevent individuals from detecting and self-reporting gendered racism when they experience it, but it does not mean that these experiences might not negatively impact psychological well-being on an unconscious level. Future research should use longitudinal designs to better understand the direction of the interrelations among these variables.

The focus on African American women college students might also reduce the generalizability of the findings. In addition, our sample had a limited age range of participants since the majority were first-year undergraduate students. Thus, from a developmental perspective, there may be a limited range of women at various stages of gendered racial identity represented in this sample, which could affect the scores on the identity centrality scale. A community sample might provide a greater diversity of African American women in terms of age, socioeconomic status, level of education, and other demographic factors, which would provide greater variability in the potential gendered racial experiences of the sample. Future research should move beyond convenience samples of college students to be able to generalize to the greater range of African American women’s experiences.

Practice Implications

Our study has implications for college clinical and counseling psychologists as well as other mental health providers. It is important for practitioners to increase their awareness about the unique types of gendered racism that African American women experience and how it can manifest both inside the college context (e.g., in the learning environment; in relations with university faculty, administrators, staff, and other students) and outside in the community. In terms of specific interventions, practitioners might lessen the potential impact that gendered racism has on African American women’s psychological distress, by using therapeutic strategies designed to provide validation, support, and empathy for their experiences. Clinicians might increase clients’ awareness of their coping mechanisms and help them select methods of coping that will decrease, rather than increase, symptoms. For example, clinicians might help an African American female client reframe the cause of a gendered racist event to the perpetrator, rather than to herself, in order to decrease her internalization/self-blame; they might explore whether the internalization of the strong Black woman stereotype is contributing to the utilization of detachment coping strategies. Given the complexities of reaching out to support networks to decrease detachment coping, clinicians might engage clients in an exploration and discussion of potential pros and cons associated with reaching out to different groups. Therapists should be mindful that some women might feel less inclined to reach out to support networks, so as not to expose friends or family members to the vicarious trauma that could come from hearing these stories of racism and sexism. In addition, reaching out to support networks that do not share similar experiences of oppression (e.g., White persons) could be met with invalidation, rather than with understanding and affirmation. Practitioners might also explore with clients whether their detachment is a long-term coping response or a temporary one, used to emotionally recoup before engaging in strategic problem-solving efforts.

Another effective culturally specific intervention is the use of “sister circles” or Black women therapy groups. Previous research has found that Black women support groups can be very helpful in providing a safe space for African American women to gain support from other Black women who might also experience gendered racism, particularly in predominantly White environments, such as college and university settings (see Neal-Barnett et al., 2011). Finally, practitioners should try to assess their clients’ racial and gender identities to determine how central and salient their identity is to their lives. This can help clinicians determine how best to help their clients develop coping strategies to combat gendered racism. For example, an African American female client whose race and gender are less salient to them might need help increasing her overall awareness of her gendered racial identity. She might be experiencing gendered racism, but be unaware that these experiences are contributing to her distress; whereas an African American woman with high gendered racial centrality might need her therapist to validate her experiences with gendered racism and support her to ward off detachment that might lead to greater psychological distress. Our findings suggest the importance of applying an intersectionality framework to explore the experiences of gendered racism and gendered racial identity centrality in African American women’s lives. Our results also lead us to recommend future work that helps African American women reduce the use of disengagement strategies to cope with discrimination.