Examining the Associations of Racism, Sexism, and Stressful Life Events on Psychological Distress among African American Women

Abstract

African American women may be susceptible to stressful events and adverse health outcomes as a result of their distinct social location at the intersection of gender and race. Here, racism and sexism are examined concurrently using survey data from 204 African American women residing in a southeastern U.S. urban city. Associations between racism, sexism, and stressful events across social roles and contexts (i.e., social network loss, motherhood and childbirth, employment and finances, personal illness and injury, and victimization) are investigated. Then, the relationships among these stressors on psychological distress are compared, and a moderation model is explored. Findings suggest that racism and sexism are a significant source of stress in the lives of African American women, and are correlated both with one another and with other stressful events. Implications for future research and clinical considerations are discussed.

Scholars have highlighted the distinct socio-contextual risk factors for adverse mental health conditions among African American women, such as race and gender (Brown, 2003). However, there is a dearth of literature focused on understanding the complexities of racism, sexism and the underlying mechanisms contributing to psychological distress in African American women. The current study uses data from 204 African American women from a southeastern urban U.S. city. We explore relationships between racism, sexism, and stressful life events occurring in distinct social contexts (i.e., social network loss, motherhood and childbirth, employment and finances, personal illness and injury, and victimization). Further, we compare the magnitude of the effects of racism, sexism, and context-specific stressors on the mental health of African American women, and examine interactions between racism and sexism on psychological distress. The current study adds to the literature by promoting an in-depth understanding of domains of risk that may be linked to African American women’s social locations and presentation of psychological distress.

Racism and Sexism: Intersectionality as a Conceptual Framework

Psychological distress experienced by African American women has been explained by several theoretical constructs in the literature. Wingfield (2008) emphasized the importance of an intersectional perspective by the illustration of Don Imus' reference to the women's Rutger's basketball team as "nappy-headed hoes" while on-air. This is a clear example reflecting the current climate of negative stereotypes held by society about African American women providing space for perpetuation of racism and sexism. Recently, Bowleg (2012) called for health disparities researchers to utilize intersectionality as a conceptual framework when examining individual level outcomes among multiple historically oppressed populations. Of significance is recognizing how systems of privilege and oppression such as racism and sexism intersect at the macro-level to maintain adverse individual level outcomes, specifically psychological distress in the current study. The core tenants of intersectionality are: (1) an individual has multiple social identities that are not independent and unidimensional, but multiple and intersecting, (2) the focus of the framework are individuals from multiple historically oppressed and marginalized groups and (3) micro level social identities such as race and gender intersect with macro-level socio-structural factors such as racism and sexism resulting in adverse outcomes (Bowleg, 2012). The term "double jeopardy" has been used to explain the impact of both racism and sexism experienced by African American women (Beal, 1969). However, for the purpose of the current study, we utilize an intersectional framework to explain how the impact of both racism and sexism relate to psychological distress and stressful-life events experienced by African American women. We propose that the socio-contextual factors of racism and sexism intersect in the lives of African American women contributing to their experience of more stressful life events and an increase in psychological distress because African American women’s experiences of being both a women and African American often cannot be easily separated (Essed, 1991).

Psychological literature has long used a deficit approach (Guthrie, 1998) by examining individual level outcomes by comparing Whites and African Americans without considering the social location of African Americans and how these identities intersect within a larger social context. This study contributes to the literature by examining psychological distress among African American women within their own context, rather than examining psychological distress as it deviates from the historical norms of White women (Bowleg, 2012).

Socio-contextual Factors and Stressors Affecting African American Women

African American women are susceptible to experiencing stress across a range of social contexts compared to their White counterparts. The intersection between race and gender has distinct differences in employment opportunities and outcomes (Skaggs, 2012). Economic hardships are contextual factors influencing stress among African American women and may be a risk factor for violence within African American families. African American women encounter many forms of violence, including childhood sexual abuse and adult victimization (West, 2002), and racism and discrimination may increase their susceptibility to such violence and types of harassment. Waltermaurer and colleagues (2006) found that perceived racial discrimination and being the victim of interpersonal violence were significantly correlated among African American women.

With regard to social roles, research suggests that partner status among African American women may not be a significant source of social support nor a protective factor against psychological distress (Brown & Gary, 1985). Furthermore, African American women are less likely to be married compared to their White counterparts (Keith & Brown, 2010). When examining racism and motherhood, Banks-Wallace (2001) found that living in a racist society impacts maternal thinking and practices of African American women. The combination of employment and parenthood roles increases stress, and some African American women experience role overload often with limited coping resources (Keith & Brown, 2010). Furthermore, recurring psychological stress combined with limited resources impacts the physical health and risk for personal illness among African American women. Specifically, a review of literature on health outcomes found that African American women have higher rates of diabetes, hypertension, and lupus which lead to higher mortality rates than White women (Keith & Brown, 2010).

Racism and Sexism as Stressors

There is significant evidence to suggest that experiences of racism and sexism are largely intertwined. In a sample of African American female college students, King (2003) found that racism and the interaction of racism and sexism predicted increased stress. However, Szymanski and Stewart (2010) found that when examining internalized racism and sexism concomitantly in a sample of adult African American women, only sexism emerged as a significant predictor of distress. In a mixed sample of college students and adult women in the community, Moradi and Subich (2003) hypothesized that racism and sexism might have a multiplicative effect on psychological distress, such that sexism is more harmful in the presence of racism. They found that recent sexist events explain variance in psychological distress after controlling for socio-demographics and a number of other potential confounding variables. Although the constructs were substantially correlated, racism and sexism did not interact to predict psychological distress (Moradi & Subich, 2003). Furthermore, in a study of socioeconomically disadvantaged African American women, Woods-Giscombé and Lobel (2008) found that generic stress, race-related stress, and gender-related stress equally contributed to overall distress, as opposed to either of latter possessing unique associations with distress. Prior studies examining racism and sexism among college age and adult African American women have resulted in mixed findings.

Buchanan and colleagues (2009) examined sexual harassment and racial harassment experiences in a sample of male and female college students (12% of the sample were African American women). The purpose of the study was to examine the unique, additive, and interactive effects on psychological outcomes (Buchanan, Bergman, Bruce, Woods, & Lichty, 2009). African American women reported higher levels of both racial and sexual harassment compared to African American men. Furthermore, unique contributions were found for both racial and sexual harassment on posttraumatic stress, depression, general clinical symptoms, and health satisfaction. Specifically, the relationship between sexual harassment and posttraumatic stress was strongest for the African American women in the study (Buchanan, Bergman, Bruce, Woods, & Lichty, 2009). The authors concluded that although the study only assesses rates of harassment over the past year, findings specific to the African American students were likely a result of a cumulative effect of sex and/or race-based traumatic experiences. Although the African American women in the study had weaker relationships between harassment and the clinical outcomes, they reported higher levels of psychological distress than other groups at lower levels of harassment. Based on the findings from previous studies, it is possible that women who have experienced excessive racism may be more sensitive to or perceptive of various forms of discrimination (Collins, 1986).

Present Study

African Americans are 20% more likely to report serious psychological distress compared to non-Hispanic Whites (U.S. Department of Health and Human Services, 2012). However, it is likely the disparities in mental health are even larger due to the underreporting and underutilization of mental health services by African Americans (Snowden, 2001). Though researchers posit that the intersection of racism and sexism places African American women at increased risk for psychological distress (Greer, Laseter, & Asiamah, 2009; Woods-Giscombé & Lobel, 2008), findings from empirical studies have been mixed. To add to the literature, we examine the following: 1) How prevalent are racism, sexism, and other more commonly measured stressful life events among African American women?; 2) Are there correlations between racism, sexism, and stressors occurring in social contexts (e.g. social network loss, motherhood, employment and finances, personal injury and illness, victimization)?; and 3) What are the magnitudes of the correlation between racism, sexism, and context-specific stressors on the mental health of African American women, and is there a moderating relationships between these variables? Our hypotheses are as follows: There will be significant positive correlations between racism, sexism, and stressful events; and the effects of racism and sexism together will be significantly associated with psychological distress as opposed to examining each construct separately. The current study contributes to understanding underlying mechanisms in the manifestation of psychological distress among African American women. In addition, it adds to the dearth of literature examining the concurrent impact of racism and sexism among African American women.

METHODS

Sample

Data were drawn from the B-WISE (Black Women in a Study of Epidemics) project, which was part of a larger epidemiological study examining health problems among a non-random, stratified sample of African American women. Data was collected from samples of self-identified drug-using and non-drug using women in the community with no history of criminal justice involvement, women in prison, and women on criminal justice probation. The current study only utilized data from the community-based sample. As part of the stratified sampling design, approximately half the sample reported having used an illicit drug in the past year. Participants were recruited using newspaper ads and fliers posted in various parts of a southeastern urban area with a large African American population (based on census data). Eligibility criteria included: (1) self-identifying as an African American woman; (2) being at least 18 years old; and (3) not currently being involved in the criminal justice system. All data were collected by African American female interviewers. Interviews were face-to-face and lasted approximately two hours using computer assisted interviewing (CAPI). Interviews were conducted in private rooms at a research office on a university campus or at public libraries across the city. Institutional Review Board approval was received and participants were compensated $20 for the interview.

After deletion of missing data (two cases), the analysis sample contained 204 African American women. The median household income in the sample ($17,500) and the percent college educated (15%) are significantly different from national statistics for African American women ($29,423; Z = −6.44; p < .001 and 24%; Z = 2.57; p < .01; 2000 Census). Also, the percent currently married in the B-WISE sample (13%) is significantly different from the national percentage (26%; Z = 3.71; p < .001; 2000 Census). Thus, the African American women in the current study are not representative of African American women nationally as expected due to the sampling strategy. For this reason, socio-demographic variables are used as controls in all models.

Measures

Psychological distress

The dependent variable is a count of common symptoms of psychological distress. It was computed using a modified version of the Addiction Severity Index Lite – CF (ASIL-CF) (McLellan, Cacciola, Carise, & Coyne, 1999). The Addiction Severity Index has been determined to be reliable for use in research across ethnicities (Makela, 2004) due to its design by assessing a wide range of domains and ability to compare between groups. Respondents were asked whether or not they had ever experienced any significant period of psychological distress or psychiatric problems. Symptoms include serious depression, serious anxiety, hallucinations, trouble understanding, concentrating or remembering, trouble controlling violent behavior, serious thoughts of suicide, attempted suicide, and prescription of drugs for psychological problems. These items are coded 1 if yes and 0 if no. The count variable was calculated by adding the responses.

Socio-demographic variables

Age and education were coded in years. Annual household income was coded in thousands of dollars. Finally, a dichotomous variable for marital status was coded 1 if the respondent is currently married and 0 if unmarried.

Stressful life events

Stressful life events were largely measured using The Traumatic Life Events Questionnaire (TLEQ) (Kubany et al., 2000). The TLEQ is intended capture an individual’s “trauma history,” and measures exposure to a broad range of events. For each item, respondents were asked how often they had experienced the event in their lifetime. Response categories are “never”, “once”, “twice”, “three times”, “four times”, “five times”, or “more than five times.” The version of the TLEQ that appears in the study survey contains modifications designed to capture more common stressors and to make the instrument more applicable to the experiences of African American women. Specifically, questions about relationships, employment, financial problems, and motherhood were added to the questionnaire (seven items; See Table 1), and questions pertaining to natural disasters and combat experiences were omitted. The TLEQ, both with and without the additional items, was significantly correlated with psychological distress, and the alpha is identical (0.78). However, the scale with the additional items provided a better model fit when examining the multivariate analyses, as indicated by the LRX² and BIC, suggesting that it may more comprehensively capture the experiences of African American women than the original scale.

Table 1.

Summary of items comprising context-specific sub-scales for stressful life events

Domain	Items (abbreviated)	Source
Employment and finances	unemployed or seeking work unsuccessfully more than one month* sacked (laid off, fired) from a job* major financial crisis*	B-WISE (Oser 2007)
Personal illness and injury	motor vehicle accident resulting in injury or death another type of accident resulting in injury or death life-threatening personal illness	TLEQ (Kubany et al. 2000)
Social network loss	close friend or loved one died unexpectedly close friend or loved one had a life-threatening or disabling illness separation due to marital difficulties* divorce or break up of steady relationship*	TLEQ (Kubany et al. 2000) B-WISE (Oser 2007)
Childbirth and motherhood	miscarriage abortion baby born prematurely or with health problems* open case with child protective services*	TLEQ (Kubany et al. 2000) B-WISE (Oser 2007)
Lifetime Victimization	victim of a mugging or robbery involving a weapon severely physical assault by an acquaintance or stranger witness to severe assault of an acquaintance or stranger threat of death or serious bodily harm intimate partner violence non-consensual sexual contact as an adult stalking victim	TLEQ (Kubany et al. 2000)
Childhood Victimization	physical abuse in childhood witness to domestic violence non-consensual sexual contact prior to age 18	TLEQ (Kubany et al. 2000)
Sexism	unfair treatment by employers, bosses, and supervisors unfair treatment by teachers, school administrators, and coaches unfair treatment by coworkers or fellow students unfair treatment by people in service jobs unfair treatment by people in helping jobs made fun of, picked on, pushed, shoved, hit, or threatened with harm gender discrimination in the workplace inappropriate or unwanted sexual advances extreme anger about being a victim of sexism called a sexist name extreme anger about sexist or sexual jokes unfair treatment by boyfriend, husband, or other important men unfair treatment by family	SSE-LM (Bowleg, Neilands, & Choi 2008)
Racism	Items 1–6 above, plus: (7)unfair treatment by strangers (8)unfair treatment by neighbors (9)unfair treatment by institutions (10)unfair treatment by supposed friends (11)called a racist name (12)false accusation or suspicion of wrongdoing (13)misunderstanding about intentions and motives (14)inaction in the face of desire to respond to racism (15)extreme anger about being a victim of racism (16)took legal or other drastic action in response to racism (17)argued or fought in response to racism	SRE (Landrine and Klonoff 1996)

^*Items were developed by PI for B-WISE

A novel contribution of this study is the conceptualization of stressful life events as endemic to distinct social roles or contexts. This strategy permits an exploration of the relationships between racism, sexism, and risk across multiple domains. Accordingly, the stressful life events are separated into six sub-scales for the purposes of this study (see Table 1 for a list of items): employment and finances, personal illness and injury, social network loss, childbirth and motherhood, and victimization. Factor analysis was not performed because these indices represent counts of conceptually-similar experiences and do not necessarily co-occur.

Scores were calculated for the total number of stressful life events experienced. The total number of stressful life events was calculated by adding the score (ranging from 0 when the event was never experienced to 6 when experienced six or more times) for each item in a sub-scale. For instance, the “Childbirth and motherhood” sub-scale contains four items, so the potential range of number of types of events is 0–4. However, respondents could experience each event up to six times, resulting in a potential range of 0–24 (or 6*4) for total number of stressful life events. A higher score for each sub-scale indicates more adverse experiences.

Sexism

Experiences of sexism were measured using the Schedule of Sexist Events (SSE; Bowleg, Neilands & Choi, 2008). This is a 13-item scale that asks whether respondents experienced a series of events in their lifetime “because you are a woman” (See Table 1 for a list of items). The total number of sexist events was calculated in an identical manner to the stressful life events sub-scale (α = 0.87).

Racism

The Schedule of Racist Events (SRE) measures lifetime experiences of racism (Landrine & Klonoff, 1996). This instrument contains 17 items, 6 of which are identical to those on the Schedule of Sexist Events. Respondents were asked whether they experienced a series of events “because you are Black” (See Table 1 for a list of items). The total number of racist events experienced was calculated in an identical manner to the stressful life events sub-scale (α = 0.92).

Analysis

Analyses explored relationships between racism, sexism, stressful events, and psychological distress. Correlations were computed to measure these associations. Because there is some overlap in the biographical circumstances captured by the stressful life events variables and the racism and sexism variables, response patterns across similar items are examined. In other words, correlation between stressful life events, racism, and sexism could be due to reporting of the same event in multiple scales rather than true association (e.g., experiencing racism is associated with an increase in vulnerability to negative life events). For this reason, items in different scales that appear similar are evaluated using cross-tabulations to determine whether response patterns are unique. Likewise, findings from correlations and regression models where items with potential overlap are omitted from the racism and sexism scales are compared to the original analyses to check robustness.

To determine the extent to which racism, sexism and other stressful events are associated with psychological distress, negative binomial regression models were computed. Incidence rate ratios (IRR) are presented in tables and express the factor change in the rate ratio of events associated with a one-unit increase in the independent variable. A series of regressions model the effects of control variables and one stressful life event on psychological distress. Models are X-standardized to compare the magnitude of coefficients across models, and Wald tests are employed to assess the equality of coefficients. Additionally, the likelihood ratio X² statistic, Nagelkerke/Cragg-Uhler pseudo-R², and BIC are used to compare fit across models. The combined relationship of both racism and sexism are explored using moderation analysis. A model that introduces a multiplicative term (racism x sexism) is computed to identify whether there is a significant interaction between racism and sexism on psychological distress. Simple slopes and dummy variable methods for assessing interaction were employed, but are not presented here because of non-significance. The stratified sampling of drug users and non-drug users could bias results. To address this concern a dummy variable for illicit drug use (1 = yes; 0 = no) was included in all models and it did not change the results or substantive findings. Also, interaction terms multiplying drug use by all the independent variables were not significant, suggesting that drug use does not moderate the effects of the other variables in the models. Racism and sexism remain significant in the model. Results from these analyses reduce concerns about the proportion of drug users in the sample (full models are available upon request).

RESULTS

Table 2 presents descriptive statistics on variables in regression analyses. Women in the BWISE sample report experiencing an average of 1.80 psychological symptoms, with a substantial degree of variation across individuals. With respect to the more commonly occurring life-stressors, African American women in the sample report experiencing an average of 4.65 events involving employment and finances, 6.04 involving lifetime victimization, 10.59 involving social network loss, 13.02 involving sexism, and 15.04 involving racism.

Table 2.

Descriptive sample characteristics (BWISE Community Sample, N=204)

	M	SD	Range
Socio-demographics
Education (years)	12.75	2.26	3.00–20.00
Household income (thousands)	20.85	21.24	2.50–87.50
Age (years)	36.39	14.19	18.00–68.00
Currently married (1=yes; 0=no)	0.13
Number of total stressful life events
Employment and finances	4.65	4.18	0.00–18.00
Personal illness and injury	0.78	1.17	0.00–7.00
Social network loss	10.59	5.20	0.00–27.00
Childbirth and motherhood	1.32	1.55	0.00–7.00
Lifetime victimization	6.04	7.56	0.00–36.00
Childhood victimization	3.42	4.60	0.00–18.00
Sexism (SSE)	13.02	7.23	0.00–33.00
Racism (SRE)	15.04	12.30	0.00–67.00
Psychological distress scale	1.80	1.87	0.00–8.00

Results from a correlation matrix examining the relationship between each of the eight stressful event sub-scales are presented in Table 3. The strongest correlation is between racism and sexism, such that experiencing more instances of sexism is associated with higher levels of racism. In addition, sexism is significantly and positively related to every other type of stressful event measured here. More specifically, there is a moderate correlation with lifetime victimization and employment/financial problems, as well a modest relationship with events pertaining to childbirth and motherhood, childhood victimization, social network loss, and illness and injury. Moreover, racism is significantly and positively correlated with lifetime victimization, employment/financial problems, illness and injury, and social network loss.

Table 3.

Correlation matrix of context-specific sub-scales for number of total stressful life events

	1	2	3	4	5	6	7
1) Employment and finances
2) Illness and injury	−0.02
3) Social network loss	0.17**	0.13
4) Childbirth and motherhood	0.17*	0.15*	0.02
5) Lifetime victimization	0.40***	0.27***	0.25***	0.33***
6) Childhood victimization	0.19**	0.22**	0.19**	0.19**	0.49***
7) Sexism (SSE)	0.34***	0.22**	0.20**	0.26***	0.44***	0.24***
8) Racism (SRE)	0.25***	0.21**	0.20**	0.07	0.30***	0.13	0.61***

^*= p < .05;

^**= p < .01;

^***= p < .001 (two-tailed tests)

Because there is modest overlap in the biographical circumstances captured by the life events variables and the racism and sexism variables (i.e., the same event could be reported in multiple scales), response patterns across similar items are examined. In all, findings show that while responses to many similar items across scales are significantly related (as determined by chi-square tests), response patterns are unique. For instance, 39% of respondents who reported being unemployed and 38% who reported being fired or laid off did not report being treated unfairly by employers and supervisors because of their gender. Similarly, 49% who were unemployed and 59% who were fired or laid off did not report unfair treatment in the workplace because of their race. This variation exists in the opposite direction, as well, with substantial percentages of respondents who reported experiencing racism or sexism in the workplace not reporting being unemployed or terminated. In addition, the Pearson’s r coefficients change very little (full results available upon request), suggesting that relationships between racism, sexism, and negative biographical circumstances are not a methodological artifact. Thus, the full original versions as described in the methods section are used in the following analyses.

Table 4 displays results from negative binomial regression models examining the relationship between the total numbers of lifetime stressors on psychological distress. Results in Models 1–6 indicate that experiencing additional stressors related to employment and finances, personal illness and injury, social network loss, childbirth and motherhood, lifetime victimization, and childhood victimization are all associated with higher levels of psychological distress. Also, Models 7 and 8 demonstrate that experiencing higher levels of sexism and racism are significantly and strongly related to psychological distress.

Table 4.

Standardized negative binomial regression of psychological distress (BWISE Community Sample, N=204)

	Model 1	Model 2	Model 3	Model 4
Socio-demographics
Education (years)	0.94 (0.82–1.09)	0.93 (0.81–1.08)	0.93 (0.80–1.07)	0.96 (0.84–1.11)
HH income (thousands)	0.94 (0.79–1.12)	0.87 (0.73–1.04)	0.88 (0.74–1.05)	0.87 (0.74–1.04)
Age (years)	1.04 (0.89–1.22)	1.09 (0.93–1.28)	1.16 (0.99–1.35)	1.10 (0.94–1.28)
Currently married1	1.04 (0.89–1.21)	1.02 (0.87–1.19)	1.11 (0.94–1.32)	1.01 (0.86–1.18)
Total stressful life events
Employment and finances	1.29 (1.10–1.50)***
Personal illness and injury		1.17 (1.01–1.35)*
Social network loss			1.22 (1.04–1.44)**
Childbirth and motherhood				1.23 (1.06–1.43)**
Pseudo-R2	0.08	0.05	0.06	0.07
LRX2	16.79**	11.06*	12.67*	13.75*
BIC	761.87	767.60	765.99	764.91
	Model 5	Model 6	Model 7	Model 8
Socio-demographics
Education (years)	0.95 (0.83–1.10)	0.97 (0.84–1.12)	0.92 (0.80–1.05)	0.92 (0.81–1.07)
HH income (thousands)	0.91 (0.77–1.07)	0.90 (0.76–1.07)	0.91 (0.77–1.07)	0.90 (0.76–1.06)
Age (years)	1.06 (0.91–1.23)	1.15 (0.99–1.34)	0.99 (0.86–1.16)	1.04 (0.89–1.21)
Currently married1	1.05 (0.90–1.22)	1.03 (0.88–1.19)	1.05 (0.91–1.22)	1.04 (0.89–1.21)
Total stressful life events
Lifetime victimization	1.34 (1.16–1.54)***
Childhood victimization		1.32 (1.15–1.51)***
Sexism (SSE)			1.49 (1.28–1.73)***
Racism (SRE)				1.36 (1.18–1.56)***
Pseudo-R2	0.11	0.10	0.15	0.12
LRX2	23.04***	21.41***	32.44***	25.14***
BIC	755.61	757.25	746.22	753.52

^*= p < .05;

^**= p < .01;

^***= p < .001 (two-tailed tests)

¹Reference category is never married, separated, widowed, or divorced

Note: X-standardized incidence rate ratios are presented, confidence intervals in parentheses

Comparing across models, racism and sexism have the largest impact on psychological distress. Moreover, regressions containing racism and sexism experiences provide a better overall model fit relative to those that include traditional stressful events, as indicated by the pseudo-R², likelihood ratio X² test, and Bayesian Information Criterion. Also, when either racism or sexism is added to Models 1–6 in Table 4, these remain significant at the .001-level across models, suggesting that they have robust, direct relationships (full models available upon request). Additionally, LRX² tests of the equality of coefficients performed following these latter models indicate that the influence of sexism experiences on psychological distress is significantly greater in magnitude than personal illness and injury (X² = 6.54, p < .01), social network loss (X² = 4.54, p < .05), and stressors pertaining to childbirth and motherhood (X² = 4.36, p < .05). Although the association between racism is larger than other stressors across these models, differences in effect size do not achieve statistical significance. Finally, no significant interaction between racism and sexism was identified, and this finding holds across different types of interaction models (i.e., simple slopes and recoding to dummy variables). This result suggests that racism and sexism have additive rather than multiplicative relationship to psychological distress in this sample.

In a full model, only childhood victimization remains significant at the .05-level (IRR = 1.18, p < .05). Racism (IRR = 1.18, p = .06) and sexism (IRR = 1.20, p = .06) are marginally significant in this full model, and become significant at the .001-level when entered separately into the model with all stressful life events. This is consistent with the high level of correlation between these two types of discrimination, as well as the reduced statistical power associated with introducing six additional covariates to the model.

DISCUSSION

Broadly, the findings of the current study provide empirical support for conceptual models of intersectionality that have been vital in explaining the relationship between racism, sexism and psychological distress in African American women (Bowleg, 2012; Collins, 1998). We found experiences of racism and sexism are strongly correlated, and are also modestly associated with more traditional stressful life events. Specifically, African American women who face high levels of sexism are also increasingly susceptible to every other type of stressor measured here, and racism was positively correlated with all stressors except victimization in childhood and problems in childbirth and motherhood.

The results of the current study indicated that racism and sexism have substantial influence on psychological distress, exceeding the magnitude of the relationships between all other types of lifetime traumatic events measured in the current study. Results indicated that the interaction term of racism and sexism was not significant, indicating each have additive as opposed to multiplicative relationship on psychological distress in the presence of other stressful life events. In other words, experiencing racism may increase perception of sexist events, which in turn increases risk for psychological distress among African American women. From a methodological standpoint, these findings add to the literature and suggest that the omission of racism and sexism from standard measures of stressful life events is problematic. Furthermore, results are consistent with previous findings demonstrating the significant relationship between racism and sexism on psychological distress among African American women (King, 2003; Moradi & Subich, 2003).

Racism, Sexism and Risk for Psychological Distress

Importantly, socio-contextual risk may also explain the significant, positive correlation of both racism and sexism to every other stressful life event measured in this study across a range of social contexts and life domains. These findings support existing research and theoretical constructs of intersectionality suggesting that African American women’s social location may increase their susceptibility to adverse life events and chronic stressors, often resulting in psychological distress (Greer, 2010). Specifically, our findings suggest that experiences of racism and sexism tend to co-occur with other life stressors. Consistent with previous research (Keith & Brown, 2010; West, 2002), we found that personal illness and injury among African American women are related to psychological distress. Furthermore, consistent with previous literature (Daily & Humphreys, 2010) results indicate that social network loss was related to psychological distress. Our findings are consistent with the literature on intersectionality and the impact of experiencing multiple forms of oppression (Bowleg, 2012). This is particularly true for our study sample in that their income was significantly lower than the national average and their marital rates were also significantly lower than the national average. The current study illustrates how multiple levels of oppression, being African American, female, lower SES, all may increase susceptibility to psychological stress.

Our findings related to employment and financial stressors were consistent with current research that demonstrated the impact of racism and sexism on adverse employment-related outcomes among African American women (Skaggs, 2012). Buchanan and Fitzergald (2008) found that sexual and racial harassment contributed significantly to occupational and psychological outcomes among African American women involved in a sexual harassment class-action lawsuit against their employer. Along these same lines, our findings indicate that racism and sexism are correlated with African American women’s exposure to violence. For example, compared to other racial/ethnic groups, African American women have a higher risk for exposure to trauma such as witnessing violence or being the victim of violence (Roberts, Gilman, Breslau, Breslau, & Koenen, 2011).

Limitations and future directions

The limitations of the study suggest a need for additional research. First, though we have modeled the association between racism and sexism on psychological symptoms, the reverse process may be driving these results. That is, individuals with high levels of distress might be susceptible to racial and gender discrimination, or more likely to perceive it. Though the impact of reverse causation cannot be ruled out with these cross-sectional data, there is some longitudinal research supporting the explanations given here. Two studies using longitudinal data determined that poor health does not predict subsequent reports of discrimination, but that racism, sexism and other forms of discrimination do lead to later health problems (Gee & Walsemann, 2009; Pavalko, Mossakowski, & Hamilton, 2003). However, additional research is needed to be more certain of the direction of these relationships. The current study is unable to draw causal conclusions based on the cross-sectional design.

Secondly, the data used here are not based on a nationally representative sample of African American women, but rather a sample of women that are often considered hard to reach for the purposes of participating in research. Most notably, half the sample self-reported illicit drug use within the past year. In addition, marriage rates, percent college educated, and household income are lower than national averages for African American women based on census reports. Readers should use caution when interpreting results, and be aware that findings may not extend to African American women in higher socioeconomic strata. Although consistent with an intersectional theoretical framework, when we examined the interaction term of racism and sexism in the models we found insignificant results. This finding is consistent with previous quantitative studies that examined multiplicative effects of racism and sexism on psychological distress (Buchanan & Fitzergald, 2008; Moradi & Subich, 2003). Buchanan and Fitzgerald (2008) proposed that it may be difficult to find significant results for psychological outcomes including a multiplicative term of racism and sexism due to the characteristics of the sample. Specifically, prior studies consisted of college student samples (Buchanan, Bergman, Bruce, and Lichty, 2008; Moradi & Subich, 2003) that may have limited experience as it related to employment and economic stressors and may lack the external validity needed to detect the multiplicative effect. Similarly, although the current study sample consisted of adult African American women, half of the sample indicated past-year illicit drug use and their income was significantly below the national average, thus limiting external validity. It may be that future research may yield significant results on psychological outcomes by examining the multiplicative effects of racism and sexism with a more nationally representative sample of African American women.

Third, our measure of psychological distress is not diagnostic and is not inclusive of all forms of psychological distress. An area for future research would be to focus on the relationship between diagnosed mental disorders such as depressive and anxiety disorders, including post-traumatic stress disorder, as they relate to experiences of racism and sexism among African American women. In addition, it is important to note that while our conceptual framework and literature review highlight the intersection of race and gender, our study measured these constructs separately. Future research is needed to assess the impact of a single construct of gendered racism (Essed, 1991).

Fourth, our study was limited in that we did not assess the magnitude of perceived distress resulting from negative life events. The adverse impact of these events is likely due to the distress itself rather than the indirect effects of the event. Future research should measure distress and take these cultural issues into account.

Clinical implications

In conclusion, the culturally responsive psychologist will include socio-contextual factors in the conceptualization of psychological distress among African American women. The current study results indicated a significant correlation between racism, sexism and other stressful life-events. Due to the intersection of these constructs, it may be difficult for a clinician to tease apart the impact of both racism and sexism among African American women. The ethically responsible psychologist will assess for experiences of racism and sexism in addition to other traumatic and stressful life events with African American female clients.