Gendered Racial Microaggressions and Black Women’s Sleep Health

Abstract

Gendered racial microaggressions reflect historical and contemporary gendered racism that Black women encounter. Although gendered racial microaggressions are related to psychological outcomes, it is unclear if such experiences are related to sleep health. Moreover, the health effects of gendered racial microaggressions dimensions are rarely investigated. Using a cohort of Black women (N = 400), this study employs an intracategorical intersectional approach to (1) investigate the association between gendered racial microaggressions and sleep health, (2) assess whether gendered racial microaggressions dimensions are related to sleep health, and (3) examine whether the gendered racial microaggressions–sleep health association persists after accounting for depressive symptoms and worry. Gendered racial microaggressions were associated with poor sleep quality overall and four specific domains: subjective sleep quality, latency, disturbance, and daytime sleepiness. Two gendered racial microaggressions dimensions were especially detrimental for sleep: assumptions of beauty/sexual objectification and feeling silenced and marginalized. After accounting for mental health, the effect of gendered racial microaggressions on sleep was reduced by 47 percent. Future research implications are discussed.

The most disrespected person in America is the Black woman. The most unprotected person in America is the Black woman. The most neglected person in America is the Black woman.

—Malcolm X (1962)

Several decades ago, Malcolm X came to this sobering conclusion. Today, Black women experience intersectional invisibility as they navigate their gendered and racialized subjugation (Collins 2000; Crenshaw 1991; Purdie-Vaughns and Eibach 2008). Erasure of Black women has manifested most recently in the lack of attention to the murders of Black women, with media, research, and activism primarily rallying around senseless killings of Black men by law enforcement officers (Crenshaw et al. 2015). Despite the movement originating with three Black women, marginalization of Black women in the Black Lives Matter movement led to the development of the #SayHerName campaign (Crenshaw et al. 2015). Nevertheless, millions of Black women continue to experience societal-level devaluation. This study centers Black women through an investigation of how their intersectional experiences with microaggressions, rooted in gendered racism, are linked to sleep health.

The current study enhances social psychological research in three ways. First, we examine the health implications of microaggressions. Sue, Bucceri, et al. (2007:273) define racial microaggressions as “brief and commonplace daily verbal, behavioral, and environmental indignities, whether intentional or unintentional, that communicate hostile, derogatory, or negative racial slights and insults to the target person or group.” As opposed to explicit or blatant forms of discrimination captured by the commonly used everyday discrimination scale (Williams et al. 1997; e.g., being followed around in stores; called names, threatened, or harassed), microaggressions manifest as subtle and often ambiguous assaults on one’s personhood and may even be characterized as compliments or praise (e.g., when someone comments, with surprise, to a Black person that they “are so articulate”). Compared to the robust literature on discrimination, research on microaggressions remains in its nascent stage. Sociological social psychology has yet to embrace microaggressions as central to empirical research on race, racism, and discrimination. Microaggressions are crucial for investigations of racial inequality because they are micro-level manifestations of structural racism (Sue, Bucceri, et al. 2007a; Sue, Capodilupo, et al. 2007; Sue, Capodilupo, and Holder 2008). Moreover, microaggressions have health implications (e.g., Davenport et al. 2021), thereby making them a critical measure of discrimination.

Second, we utilize an intersectional perspective to assess the frequency with which gendered racial microaggressions are reported among Black women. Empirical research adopting intersectionality as a theoretical framework has proliferated in recent years (Collins 2019), yet the intersectional mechanisms underlying Black women’s health are worthy of further empirical investigation (Aguayo-Romero 2021). In the previous Race, Racism, and Discrimination special issue, Bobo and Fox (2003) proposed that studies could be enhanced through directly engaging a minority group perspective, illuminating how race intersects with class and gender, and building new types of theoretical and methodological bridges. The present study examines the intersectional experience of microaggressions among Black women specifically and connects these experiences to sleep health.

Third, this study explores mental health as a potential mechanism underlying the gendered racial microaggressions–sleep association. The pathways linking discrimination to sleep health are relatively underexplored (Lewis and McKinnon 2019), yet prior research confirms that gendered racial microaggressions have negative implications for mental health, including depressive symptomatology (Lewis and Neville 2015; Williams and Lewis 2019). Also, there is a larger extant literature linking microagressions more generally to mental health (e.g., Nadal et al. 2014). Thus, it stands to reason that mental health may be one pathway linking gendered racial microaggressions to poor sleep. We explore this conjecture by assessing the extent to which depressive symptoms and worry attenuate the association between gendered racial microaggressions and sleep health. Overall, we enhance the discrimination literature by ascertaining how intersectional microaggressions are linked to sleep health and explore one potential mechanism through which racialized stress exposure influences Black women’s health.

BACKGROUND

Intersectionality and Black Women’s Sleep Health

Intersectionality provides an orientating framework to better understand how microaggressions influence Black women’s health. Intersectionality, as a concept, was formally articulated by legal scholar Kimberlé Crenshaw (1991) and further elaborated by sociologist Patricia Hill Collins (2000, 2019). Even prior to the development of the nomenclature, intersectionality is rooted in a long tradition of Black feminist scholarship (Hull, Bell-Scott, and Smith 1982). A core tenet of intersectional theorizing is that multiple systems of oppression (e.g., racism, sexism, and classism) interact in multiplicative ways to impinge on the life chances of individuals at the intersections of their multiple social identities. Black women thus experience oppression not only due to their racial status as Black but also because of their gendered status as women. As such, Black women are subjected to gendered racism that operates to disadvantage them across multiple health dimensions (Collins 2019; Perry, Pullen, and Oser 2012).

Black women’s structural disadvantage translates into poor health. Sleep health is critical to study among Black women because they experience worse sleep outcomes relative to other race/gender groups. Compared to White women, Black women exhibit shorter sleep duration, poorer sleep efficiency, and longer sleep-onset latency (Gaston et al. 2019; Hall et al. 2009; Lauderdale et al. 2009). Moreover, Black women report worse sleep quality than their Black male peers (Johnson et al. 2016). Poor sleep is associated with mortality risk (Cappuccio et al. 2010), cardiovascular disease (Cappuccio et al. 2011), obesity (Piccolo et al. 2013), and diabetes (Zizi et al. 2012). Therefore, sleep health challenges may be the previously unexamined linchpin for explaining racial and gender disparities in physical health.

A recent review of research on racial patterns of sleep called for greater integration of intersectional theorizing (Johnson et al. 2019). Harari and Lee (2021) found only two sleep studies that drew on intersectionality theory (Assari et al. 2017; Trinh et al. 2017). Both studies utilized an intercategorical approach to intersectionality (i.e., systematic comparisons between social groups; McCall 2005), which can obscure within-group heterogeneity and nuances in health profiles (Harari and Lee 2021). In contrast, the present study employs an intracategorical complexity approach to assess the unique experiences of Black women, recognizing the dynamic nature of their intersectional identities. While scholars have convincingly demonstrated that sleep is an indicator of gender (e.g., Burgard 2011) and racial (e.g., Sheehan, Walsemann, and Ailshire 2020) inequality, we know significantly less regarding sleep as an important site reflecting the simultaneity of gender and racial inequality. This study addresses the gap in this literature by examining how macro-level forces of sexism and racism operate through gendered racial microaggressions to relate to the sleep of Black women.

Gendered Racial Microaggressions and Implications for Health

Assessing Black women’s gendered racial microaggression experiences aligns with intersectional research emphasizing their gendered and racialized subordination. Nevertheless, the majority of past research on microaggressions utilizes measures of racial microaggressions. Instruments such as the Racial and Ethnic Microaggressions Scale (Nadal 2011) and the Ethnic Microaggressions Scale (Huynh 2012) capture subtle statements and behaviors that communicate negative messages to ethnoracial minorities. Yet these measures are not adapted to the specific experiences of racism that different populations encounter. Certain microaggressions are unique to a particular racial group, such as ascriptions of intelligence and treatment as perpetual foreigners for Asians (Sue, Capodilupo, et al. 2007) and assumptions of criminality and intellectual inferiority for Blacks and Latinxs (Sue et al. 2008). General racial microaggression scales can therefore miss some experiences when not tailored to a specific racialized group. In addition, extant measures do not evaluate intersectional microaggressions, such as intersecting experiences based on race and gender (for a similar argument regarding racial discrimination scales, see Harnois 2022). Although measures of gendered microaggressions have also emerged (e.g., Capodilupo et al. 2010), these scales, like the racial microaggression scales, are disconnected from Black women’s intersectional realities.

Lewis and Neville’s (2015) Gendered Racial Microaggressions Scale appraises the specific experiences of Black women due to their social location at the intersection of race and gender. Although the gendered racial microaggressions scale is moderately correlated to both the Racial and Ethnic Microaggressions Scale (Nadal 2011) and the Schedule of Sexist Events (Klonoff and Landrine 1995), it also encompasses unique experiences of Black women that cannot be captured by measures of racial or gendered microaggressions alone. The gendered racial microaggressions scale captures four dimensions of gendered racism: (1) assumptions of beauty and sexual objectification, (2) silenced and marginalized, (3) strong Black woman stereotype, and (4) angry Black woman stereotype.

First, the assumptions of beauty and sexual objectification dimension is aligned with Black women’s frequent encounters with stereotypes about their physical appearance as well as sexual objectification rooted in the racialization of their bodies (Lewis and Neville 2015). This form of gendered racism is reflected in the broader symbolic violence Black women endure as their physical features are critiqued and dissected in public discourse (Chaney 2017). For instance, a former governor of West Virginia referred to Michelle Obama, the former First Lady of the United States, as an “ape in heels” (Bever and Lexi 2016). In an article praising Michelle Obama’s fashion sense, words like “hideous,” “drag queen,” and “whore” were trending in the comments section (Chaney 2017). Prominent tennis player Serena Williams has been referred to as a “gorilla,” “manly,” and “ugly” (Desmond-Harris 2016). Even at the height of her professional success, sexist comments about her “formidable” derriere, “lugging” breasts, and sexual undesirability accompanied every professional milestone (Litchfield et al. 2018). The hyper-sexualization of Black women and simultaneous reference to them as unattractive and undesirable come with psychological costs (Lewis and Neville 2015; Uzogara and Jackson 2016). Accordingly, Black women’s sleep may be negatively affected.

Second, Black women frequently report feeling silenced and marginalized at work, school, and other professional settings (Leyva 2021; Melaku 2019). Elements of Black women’s silencing and marginalization were operationalized in the gendered racial microaggressions scale with items such as being “disrespected in the workplace” and “excluded from networking opportunities” (Lewis and Neville 2015). Many of these complaints pertain to the workplace, a “high stakes” environment in which Black women experience hyper-visibility and high levels of scrutiny (Wingfield 2007, 2010).

Third, Black women are exoticized through projected notions of being strong, independent, and assertive. The strong Black woman stereotype includes items such as “I am assumed to be a strong Black woman” and “I have been told I am too independent” (Lewis and Neville 2015). A recent study of undergraduate women attending a Historically Black College/University identified a bivariate association between endorsing strong Black woman ideals and poor sleep quality (McLaurin-Jones et al. 2021). It is interesting that adopting a “strong Black woman” self-orientation could have negative implications for sleep health given that many Black women view this characterization as desirable and necessary (Beauboeuf-Lafontant 2009; Woods-Giscombé 2010). Expectations for Black women to fulfill the needs of others, to appear unbreakable in the face of adversity, and to resist asking for help may place disproportionate burdens on Black women that pose barriers to maintaining healthy sleep practices. More research is needed to better understand how Black women’s notions of strength could be problematic for sleep health.

Last, Black women encounter the stereotype of being angry even when acting or speaking calmly (Lewis and Neville 2015). Black women’s characterization as “angry” was personified in fictional TV character, Sapphire, from a 1950s U.S. television show Amos ’n’ Andy (Judd 2019). According to Thomas, Witherspoon, and Speight (2004:437), “Sapphire is seen as argumentative and harsh. Women who internalize this stereotype may fear being perceived as overly aggressive and may have difficulty expressing their anger.” This controlling image of Black womanhood is ever present in professional and educational settings (Harlow 2003; Wingfield 2010). Even when attempting to defy the stereotype, “Sapphire is so totalizing that it does not take any real anger on the part of a Black woman to be accused of being angry” (Judd 2019:185). Given the negative connotation of Black women’s characterization as “angry,” it is anticipated that such a label being attached to their personhood will be associated with poor sleep health.

Past research confirms that greater gendered racial microaggressions frequency is associated with depressive (Williams and Lewis 2019), anxiety (Wright and Lewis 2020), and traumatic (Dale and Safren 2019) symptoms. Although discrimination and microaggressions more broadly have been associated with sleep across studies (Bethea et al. 2020; Davenport et al. 2021; Lewis and McKinnon 2019; Ong et al. 2017), it is unclear how gendered racial microaggressions are associated with sleep health among Black women. Moreover, the health effects of specific gendered racial microaggressions dimensions are rarely assessed (for exceptions, see Dale and Safren 2019; Erving et al. 2022). We hypothesize that gendered racial microaggressions will be associated with poor sleep health. The analysis of the association between specific dimensions of gendered racial microaggressions and sleep is exploratory in nature. Although any experienced microaggression is problematic and worthy of eradication, identifying which specific gendered racial microaggressions components are detrimental to Black women’s sleep health may help identify intervention strategies in reducing intersectional stress and improving health in this population.

Mental Health as an Underlying Mechanism

In addition to exploring the relationship between gendered racial microaggressions and sleep health, this study explores a potential mechanism underlying this association: mental health. As noted previously, gendered racial microaggressions is associated with several mental health outcomes, including depression, anxiety, and traumatic symptoms (e.g., Dale and Safren 2019; Erving et al. 2022; Moody and Lewis 2019; Williams and Lewis 2019; Wright and Lewis 2020). Thus, the mental health effects of gendered racial microaggressions may be also linked to poor sleep health among Black women. An example might be illustrative. Upon being on the receiving end of chronic silencing and marginalization in a workplace or educational setting, Black women may feel demoralized and worry about how these exclusionary practices may affect them professionally, which, in turn, could have negative implications for sleep. Hence, depressive symptoms and worry may underlie any identified association between gendered racial microaggressions and sleep.

Prior research indicates that race-related stress exposure is related to poor sleep outcomes (Goosby, Straley, and Cheadle 2017); moreover, worry, a form of repetitive negative thought patterns associated with anxiety symptoms, is one pathway linking stress to poor sleep (Beatty et al. 2011; Tousignant et al. 2019). The body of research on anticipatory stress suggests that the mere perceived threat of experiencing stress can have negative implications for mental health, particularly among Black Americans (Grace 2020; Utsey et al. 2012). Recent research on stress and poor sleep among Black Americans also implicates mental health as an underlying mechanism (Hart et al. 2021; Hoggard and Hill 2018). This study extends this work by applying an intersectional lens to research on discrimination stress to explore whether mental health operates as a pathway through which gendered racial microaggressions is linked to poor sleep.

In sum, this study engages three research questions.

• Research Question 1: Among Black women, what is the association between gendered racial microaggressions and sleep health?

• Research Question 2: Among Black women, what specific gendered racial microaggressions dimensions (e.g., silenced and marginalized, angry Black woman stereotype) are associated with sleep health?

• Research Question 3: Among Black women, to what extent does mental health (i.e., depressive symptoms and worry) attenuate the association between gendered racial microaggressions and sleep health?

DATA AND METHODS

This study leveraged data from the Mechanisms Underlying the Impact of Stress and Emotions (MUSE) in African-American Women’s Health Study, a cohort of 422 Black or African American-identified women (Spikes et al. 2022). The overarching goal of the MUSE data collection effort was to ascertain the extent to which psychosocial stressors (e.g., expectations of stress) influence cardiovascular disease risk. Baseline data, collected from December 2016 to March 2019, were used for the present study. National Opinion Research Center (NORC) services were acquired to identify participants representing a wide range of socioeconomic backgrounds and census tracts in the Atlanta, Georgia, metropolitan area. NORC utilized consumer residential and voter registration lists to identify potential Black women in the target age range (30–45 years of age) across census tracts. These individuals were then sent a flyer introducing the study, followed by a phone call. A total of 1,989 persons who expressed interest in the study were then prescreened via telephone to determine eligibility. Inclusion criteria were self-identifying as a Black woman, being ages 30 to 45 at screening, and being premenopausal with at least one ovary. Exclusion criteria included a history of clinical cardiovascular disease, being pregnant or lactating, any chronic illness known to influence atherosclerosis (e.g., HIV/AIDS, autoimmune or chronic inflammatory diseases, such as lupus/rheumatoid arthritis, renal disease, liver disease), current treatment for psychiatric disorders, current illicit drug use (i.e., marijuana, cocaine), or alcohol abuse. Women who reported working overnight shifts were excluded because of the impact of shift work on diurnal rhythms and ambulatory blood pressure (Spikes et al. 2022).

Based on these inclusion and exclusion criteria, 1,158 individuals were ineligible, and 831 were eligible to participate in the study. Eligible participants were contacted by study staff and scheduled for an in-person visit. Of those who were eligible, a total of 422 respondents completed the in-person three-hour interview. Respondents represent approximately 200 unique census tracts in the Atlanta metropolitan area. Interviews were conducted in English by interviewers who identified as Black women. All procedures were approved by the Institutional Review Board, and all participants provided informed consent. After dropping missing cases on the study measures, 400 respondents are included in the analysis for this study. Data were missing at random, and the number of respondents missing data was highest for household income (N = 10), sleep health (N = 6), depressive symptoms (N = 4), and employment status (N = 3).

Measures

Dependent measures.

Sleep health was assessed using the Pittsburgh Sleep Quality Index (PSQI), a self-rated questionnaire that evaluates sleep quality and disturbances during the past month (Buysse et al. 1988). PSQI includes 19 items covering seven sleep health domains: sleep quality, sleep latency, sleep duration, habitual sleep efficiency, sleep disturbances, use of sleeping medication, and daytime dysfunction. Composite scores on the PSQI have a possible range of 0 to 21, with higher scores indicating lower quality sleep and more sleep complaints (what we refer to as composite PSQI score). PSQI has been widely used across study populations with good validity and high test-retest reliability (Backhaus et al., 2002). The PSQI scale is psychometrically valid among African Americans (Bidulescu et al., 2010). PSQI reliability is also stable over time among Black early middle-aged adults (Knutson et al. 2006). In addition to the PSQI composite score, in supplemental analyses, we assessed six sleep domains as separate dependent measures.¹

Independent measures.

Gendered racial microaggressions were assessed using the Gendered Racial Microaggressions Scale, a 26-item measure that assessed Black women’s experiences of everyday and subtle gendered racism (Lewis and Neville 2015). Study participants reported the frequency of experiences in the past year on a 6-point Likert-type scale ranging from 0 = never to 5 = once a week or more. Scores were averaged to calculate a total mean frequency score and had high reliability (α = .94). Convergent validity has been supported by significant positive correlations with measures of racial microaggressions and perceived sexist events (Lewis and Neville 2015). Internal consistency reliability estimates for gendered racial microaggressions frequency have been good in community-based samples of Black women (Lewis and Neville 2015).

In addition to the global gendered racial microaggressions scale, we ascertained whether specific components were associated with sleep. The four gendered racial microaggressions dimensions included: assumptions of beauty and sexual objectification (11 items; e.g., “Someone made negative comments about my hair when natural”; α = .88), silenced and marginalized (7 items; e.g., “Someone has tried to ‘put me in my place”’; α = .91), strong Black woman stereotype (5 items; e.g., “I have been told that I am too assertive”; α = .81), and angry Black woman stereotype (3 items; e.g., “Someone accused me of being angry when speaking calm”; α = .81).

Mental health.

First, depressive symptoms were assessed by the 21-item Beck Depression Inventory (BDI; Beck et al. 1961; Beck, Steer, and Garbin 1988). The BDI is a reliable and validated measure of depressive symptoms for use among Black women (Gary et al. 2018). Items assessed emotional states such as sadness, feeling like a failure, crying, loss of interest, and fatigue (Beck et al. 1988). Each item reflected a depressive symptom, and respondents were asked to indicate the severity of the symptom during the past few days using a scale from 0 to 3. Higher composite scores indicated greater severity of the symptoms, and BDI had high reliability in this sample (Cronbach’s α = .90). Consistent with prior research, we removed the “sleep” item from the BDI (Van Dyke et al. 2016). Second, the Penn State Worry Questionnaire (PSWQ) is a 16-item measure assessing the intensity and excessiveness of worry (Meyer et al. 1990). PSWQ is a reliable, validated measure of worry for use among Black Americans (Rucker, West, and Roemer 2010) and has been linked to anxiety disorders such as generalized anxiety disorder (Meyer et al. 1990). Some items included the following: “As soon as I finish one task, I start to worry about everything else I have to do”; “If I don’t have enough time to do everything, I don’t worry about it” (reverse-scored); and “I am always worrying about something.” Response options included a 5-point scale ranging from not at all typical of me to very typical of me. Higher composite scores reflected higher worry levels. Worry had high reliability in this sample (Cronbach’s α = .93).

Controls.

All analyses were adjusted for several factors associated with sleep health. Marital status distinguished between those currently married (reference), “living as married” (i.e., cohabiting), divorced/separated/widowed, and never married. Parents referred to individuals who reported being a parent (= 1). Respondents were considered employed if they reported working for pay (= 1). Two socioeconomic status indicators were included in the analysis. First, annual household income included the following categories: less than $35,000 (reference), between $35,000 and $49,999, between $50,000 and $74,999, and $75,000 or more. Second, educational attainment included three categories: high school diploma/equivalent or less (reference), some college or occupational training, and bachelor’s degree or higher. Age ranged from 30 to 46 years. Number of household members referred to the number of individuals living in the respondent’s household; the measure ranged from 1 to 12. Last, we adjusted for health and health behavior including body mass index (Van Dyke et al. 2016) and current smoking status (Sheehan et al. 2020).

Analytic strategy.

The analysis began with descriptive statistics for key study measures and controls (Table 1). Next, Pearson’s r product-moment correlation coefficients were calculated for the key study measures in Table 2. Ordinary least squares (OLS) regression analysis was utilized to assess the association between gendered racial microaggressions overall and sleep health (Table 3) as well as gendered racial microaggressions subscales and sleep health (Table 4). Model 1 was unadjusted. Model 2 adjusted for sociodemographic and health factors, including marital status, parental status, employment status, household income, educational attainment, age, number of household members, body mass index, and current smoking status. Next, analyses examined the roles of depressive symptoms and worry because they are psychological risk factors for poor sleep and potential mediators of the association between gendered racial microaggressions and sleep health. To understand the role of mental health, we evaluated changes in the gendered racial microaggressions coefficients after controlling for depressive symptoms and worry. Therefore, the next set of models adjusted for depressive symptoms (Model 3), worry (Model 4), and both simultaneously (Model 5), and the percent attenuation of the gendered racial microaggressions coefficients was calculated for Model 2 compared to Model 5. Substantial reductions in the magnitude and statistical significance of the gendered racial microaggressions–sleep association when accounting for mental health in Models 3 through 5 would suggest that mental health at least partially explained the relationship. All analyses were conducted using STATA 16 (StataCorp 2019).

Table 1.

Descriptive Statistics (N = 400)

	Mean or percentage	(N)	SD	Minimum	Maximum
Composite Pittsburgh Sleep Quality Index score	6.71		3.55	.00	19.00
Gendered Racial Microaggressions Scale	1.41		.95	.00	4.50
Gendered racial microaggressions subscales
Assumptions of beauty and sexual objectification	1.15		.95	.00	4.91
Silenced and marginalized	1.21		1.12	.00	5.00
Strong black woman stereotype	2.02		1.28	.00	5.00
Angry black woman stereotype	1.80		1.30	.00	5.00
Mental health measures
Depressive symptoms	5.29		6.40	.00	38.00
Worry	34.12		15.17	1.00	64.00
Control measures
Marital status
Married (reference)	32%	(127)
Living as married	5%	(20)
Divorced/separated/widowed	13%	(53)
Never married	50%	(200)
Parent	74%	(297)
Employed	87%	(349)
Household income
Less than $35,000 (reference)	25%	(99)
Between $35,000 and $49,999	21%	(85)
Between $50,000 and $74,999	23%	(91)
$75,000 or more	31%	(125)
Educational attainment
High school or less (reference)	31%	(124)
Some college/occupational training	21%	(84)
College or higher	48%	(192)
Age (years)	37.48		4.29	30.00	46.00
Number of household members	3.58		1.79	1.00	12.00
Body mass index	32.91		8.40	17.16	62.13
Current smoker	10%	(38)

Source: Mechanisms Underlying the Impact of Stress and Emotions in African-American Women’s Health Study, 2016–2019.

Note: Depressive symptoms were assessed using the Beck Depression Inventory, and worry was assessed using the Penn State Worry Questionnaire.

Table 2.

Bivariate Pearson’s Correlations for Key Dependent and Independent Measures (N = 400)

	1	2	3	4	5	6	7
1. Composite Pittsburgh Sleep Quality Index score	1
2. Gendered racial microaggressions	.22*	1
3. Beauty and objectification	.21*	.92*	1
4. Silenced and marginalized	.21*	.87*	.74*	1
5. Strong Black woman	.17*	.82*	.64*	.56*	1
6. Angry Black woman	.16*	.81*	.64*	.60*	.73*	1
7. Depressive symptoms	.37*	.28*	.28*	.26*	.17*	.20*	1
8. Worry	.31*	.34*	.33*	.33*	.22*	.24*	.49*

Source: Mechanisms Underlying the Impact of Stress and Emotions in African-American Women’s Health Study, 2016–2019.

^*p < .05.

Table 3.

Beta Coefficients from Ordinary Least Squares Analysis of the Association between Gendered Racial Microaggressions and Sleep Health–Composite Pittsburgh Sleep Quality Index Score (N = 400)

	Model 1	Model 2	Model 3	Model 4	Model 5
Gendered Racial Microaggressions Scale	.83***	.88***	.57**	.60**	.47*
	(.18)	(.18)	(.18)	(.19)	(.19)
Mental health
Depressive symptoms			.16***		.14***
			(.03)		(.03)
Worry				.05***	.03*
				(.01)	(.01)
Controls
Marital status (married = reference)
Living as married		−.54	−.37	−.47	−.36
		(.82)	(.79)	(.80)	(.79)
Divorced, separated, widowed		−.59	−.47	−.47	−.42
		(.62)	(.59)	(.60)	(.59)
Never married		−.07	−.10	.07	−.02
		(.50)	(.48)	(.49)	(.47)
Parent		.42	.50	.23	.39
		(.47)	(.45)	(.46)	(.45)
Employed		.06	.43	.21	.45
		(.53)	(.51)	(.52)	(.51)
Household income (less than $35,000 = reference)
Between $35,000 and $49,999		−.25	−.03	−.06	.04
		(.53)	(.51)	(.52)	(.51)
Between $50,000 and $74,999		.32	.53	.54	.62
		(.54)	(.51)	(.53)	(.51)
$75,000 or more		−.23	.01	−.09	.04
		(.58)	(.56)	(.57)	(.55)
Education (high school diploma or less = reference)
Some college		.72	.61	.71	.62
		(.51)	(.49)	(.50)	(.48)
College or higher		−.88	−.73	−.73	−.67
		(.47)	(.45)	(.46)	(.45)
Age		.01	.02	.02	.02
		(.04)	(.04)	(.04)	(.04)
Number of household members		.06	.04	.09	.06
		(.12)	(.12)	(.12)	(.12)
Body mass index		.06**	.05*	.05**	.05*
		(.02)	(.02)	(.02)	(.02)
Current smoker		.58	.27	.28	.16
		(.61)	(.59)	(.60)	(.59)
Constant	5.55***	2.86	2.19	1.12	1.36
	(.31)	(1.92)	(1.84)	(1.92)	(1.87)
Adjusted R2	.05	.10	.17	.14	.18

Source: Mechanisms Underlying the Impact of Stress and Emotions in African-American Women’s Health Study, 2016–2019.

Note: Standard errors are in parentheses.

^*p < .05.

^**p < .01.

^***p < .001.

Table 4.

Beta Coefficients from Ordinary Least Squares Analysis of the Association between Gendered Racial Microaggressions Subscales and Sleep Heath– Composite Pittsburgh Sleep Quality Index Score (N = 400)

	Assumptions of beauty and sexual objectification		Silenced and marginalized		Strong Black woman stereotype		Angry Black woman stereotype
Model	ß	(SE)	ß	(SE)	ß	(SE)	ß	(SE)
Model 1: unadjusted	.78***	(.18)	66***	(.16)	.48 **	(.14)	.43**	(.14)
Model 2: adjusted for sociodemographic factorsa	.84***	(.18)	.79***	(.15)	.45**	(.14)	.40**	(.14)
Model 3: Model 2 + depressive Symptomsb	.52**	(.18)	.51**	(.16)	.31*	(.13)	.23	(.13)
Model 4: Model 2 + worryc	.56 **	(.19)	.54**	(.16)	.31*	(.14)	.25	(.13)
Model 5: Model 2 + depressive symptoms + worry	.42*	(.19)	.42**	(.16)	.26	(.13)	.18	(.13)

Source: Mechanisms Underlying the Impact of Stress and Emotions in African-American Women’s Health Study, 2016–2019.

^aMarital status, parental status, employment, household income, education, age, household size, body mass index, and current smoker.

^bBeck Depression Inventory.

^cPenn State Worry Questionnaire.

^*p < .05.

^**p < .01.

^***p < .001.

RESULTS

Descriptive Statistics

Table 1 includes means, proportions, and, where appropriate, standard deviations for study measures. The mean composite PSQI score was 6.71 (SD = 3.55), higher than the standard cutoff of 5 for poor sleep quality (Buysse et al. 1988). The frequency of gendered racial microaggressions overall was moderate, falling between the “less than once a year” to “a few times a year” categories (M = 1.41, SD = .95). Compared to other studies that assessed gendered racial microaggressions frequency, the means and standard deviations for this cohort were lower (Moody and Lewis 2019; Wright and Lewis 2020). Regarding the gendered racial microaggressions subscales, the strong Black woman stereotype was the most frequently reported (M = 2.02, SD = 1.28), followed by the angry Black woman stereotype (M = 1.80, SD = 1.30). Assumptions of beauty and sexual objectification (M = 1.15, SD = .95) and silenced and marginalized (M = 1.21, SD = 1.12) were less commonly reported microaggressions.

With regards to mental health, worry levels were high, with a mean of 34.12 (SD = 15.17), whereas depressive symptoms were moderate (M = 5.29, SD = 6.40). In reference to the study control measures, 32 percent were married; 5 percent were living as married; 13 percent were divorced, separated, or widowed; and 50 percent were never married. Nearly three-quarters of respondents were parents, and 87 percent were employed. With regards to household income, 25 percent fell in the category of less than $35,000, 21 percent reported $35,000 to $49,999, 23 percent reported $50,000 to $74,999, and 31 percent were in the highest household income category of $75,000 or more. In terms of education, 31 percent had a high school diploma or lower educational attainment. Twenty-one percent had some college education, and nearly half (48 percent) of respondents had at least a bachelor’s degree. The mean age of the sample was 37.48 years (SD = 4.29). The mean number of household members was 3.58 (SD = 1.79). In terms of health indicators, body mass index was above clinical obesity levels (M = 32.91, SD = 8.40), and 10 percent were current smokers.

Bivariate correlations among dependent and independent measures are reported in Table 2. The PSQI composite score was positively correlated with gendered racial microaggressions overall (r = .22, p < .001). That is, gendered racial microaggressions were correlated with worse sleep health. PSQI was also positively correlated with each gendered racial microaggressions dimension; the largest correlation coefficients were identified for assumptions of beauty and sexual objectification (r = .21, p < .001) and silenced and marginalized (r = .21, p < .001), followed by strong Black woman stereotype (r =.17, p < .001) and angry Black woman stereotype (r = .16, p < .01). PSQI composite scores were positively correlated with depressive symptoms (r = .37, p < .001) and worry (r = .31, p < .001). Gendered racial microaggressions overall were correlated with depressive symptoms (r = .28, p < .001) and worry (r = .34, p < .001). Depressive symptoms and worry were also significantly and positively correlated with each gendered racial microaggressions subscale.

Regression Analysis

Table 3 reports beta coefficients (ß) from OLS regression models of the association between gendered racial microaggressions and sleep health (composite PSQI score). In Model 1, the unadjusted model, each unit increase in gendered racial microaggressions overall was associated with a .83 higher PSQI composite score (p < .001). Stated differently, higher gendered racial microaggressions frequency was associated with worse sleep health. Model 2 adjusted for study controls, and the association between gendered racial microaggressions and PSQI remained significant (ß = .88, p < .001). In addition, body mass index was associated with worse sleep quality (ß = .06, p <.01).

Subsequent models assessed whether the association between gendered racial microaggressions and PSQI was attenuated after accounting for depressive symptoms (Model 3), worry (Model 4), and both simultaneously (Model 5). In Model 3, after adjusting for depressive symptoms (ß = .16, p < .001), the coefficient for gendered racial microaggressions was reduced by 35 percent (ß = .57, p < .01). In Model 4, after adjusting for worry (ß = .05, p < .001), the coefficient for gendered racial microaggressions was reduced by 32 percent (ß = .60, p < .01). When adjusting for depressive symptoms (ß = .14, p < .001) and worry (ß = .03, p < .05), the coefficient for gendered racial microaggressions was reduced by 47 percent (ß = .47, p < .05). In other words, the association between gendered racial microaggressions and PSQI was reduced by nearly half after accounting for mental health.

Table 4 includes analyses assessing the association between each gendered racial microaggressions subscale and sleep health before and after adjusting for sociodemographic factors, depressive symptoms, and worry. The analysis follows the same model progression as Table 4, but only coefficients for gendered racial microaggressions are reported. Two gendered racial microaggressions subscales were associated with higher PSQI scores before and after adjusting for mental health: assumptions of beauty and sexual objectification and silenced/marginalized. However, in comparing the coefficients in Models 2 (adjusted for sociodemographic factors) and 5 (adjusted for sociodemographic factors and mental health), the coefficient for assumptions of beauty and sexual objectification goes from ß = .84 (p < .001) to .42 (p < .05), a 50 percent reduction in the coefficient size. The coefficient for silenced and marginalized goes from ß = .79 (p < .001) to .42 (p < .01), a 47 percent reduction in the coefficient magnitude.

A different pattern of findings emerged for strong Black woman and angry Black woman stereotype dimensions of gendered racial microaggressions. These subscales were both associated with higher PSQI scores in the unadjusted model (Model 1) and the model adjusting for sociodemographic factors (Model 2). However, once depressive symptoms and worry were accounted for, both dimensions fell to nonsignificance. This suggests that mental health (specifically depressive symptoms and worry) may be a pathway linking gendered racial microaggressions to poor sleep health, especially for strong Black woman and angry Black woman stereotype gendered racial microaggressions dimensions.

DISCUSSION

Gendered racial microaggressions reflect historical and contemporary gendered racism Black women experience at the hands of American society (Lewis and Neville 2015). The goals of this study were to ascertain the extent to which Black women’s self-reported gendered racial microaggression experiences were related to sleep health, investigate which dimensions of gendered racial microaggressions were associated with sleep health, and assess whether mental health attenuated the linkage between gendered racial microaggressions and sleep health. In addition to assessing sleep health using a measure of global sleep quality, we also assessed whether specific domains of sleep (e.g., duration, disturbance) were sensitive to having experienced gendered racial microaggressions among Black women.

Study results revealed that gendered racial microaggressions were associated with worse sleep health overall and four specific sleep domains: subjective sleep quality, sleep latency, sleep disturbance, and daytime sleepiness.² Although gendered racial microaggressions were associated with four sleep domains, they were not associated with sleep duration or sleep efficiency. Two-thirds of Black women in this sample reported short sleep (i.e., less than seven hours). Given the high prevalence of short sleep, other dynamics like caregiving responsibilities for children and older family members, which disproportionately befall Black women (Thorne 2020), might play a larger role in understanding why most of the sample reported sleep duration of less than seven hours each night. Sleep efficiency assessed the proportion of time asleep over the time spent in bed. Perhaps this measure is less sensitive to gendered racial microaggressions because time spent in bed could be related to other social arrangements such as nursing a young child or spending time on the phone while in bed. In sum, although two sleep domains were unrelated to gendered racial microaggressions, we identified four other sleep domains that were related to Black women’s experiences of gendered racial microaggressions.

In addition to gendered racial microaggressions overall, we explored whether specific gendered racial microaggressions dimensions predicted poor sleep health. In particular, the associations were strongest for two dimensions: assumptions of beauty and sexual objectification and silenced and marginalized. Assumptions of beauty and sexual objectification included items capturing the presumed hyper-sexualization of Black women (e.g., “Someone assumed I was sexually promiscuous”) and the devaluation of Black physical features (e.g., negative comments about skin tone; Lewis and Neville 2015). These caricatures of Black womanhood are rooted in historically embedded controlling images of them as sexual objects (Collins 2000) and the historical and contemporary pressures for Black women to subscribe to Eurocentric standards of beauty, including fairer complexion, straight hair, and narrow facial features (Uzogara and Jackson 2016). Low levels of self-worth may accompany these types of gendered racial microaggressions, which, in turn, negatively impact sleep.

Second, reports of feeling silenced and marginalized were associated with worse sleep health. These two features of gendered racism provide evidence of Black women’s intersectional invisibility (Purdie-Vaughns and Eibach 2008). Although silencing has been conceptualized as an experience shared by all women because of gendered expectations of them to be self-sacrificing and quiet (Crowley Jack 1991), Black women’s silencing is also accompanied by marginalization due to their disadvantaged racial status. Aspects of the silenced and marginalized subscale consisted of Black women feeling as if others assumed they had nothing to contribute to a conversation and feeling as if someone tried to “put [them] in [their] place” (Lewis and Neville 2015). These experiences, unfortunately, are common in the workplace (Melaku 2019; Wingfield 2007), educational settings (Haynes et al. 2020), and even in the Black community more broadly (Jones and Shorter-Gooden 2003). Given that the current study only adjusted for employment status (and 87 percent of the sample was employed), future research should examine how other aspects of the work environment may interact with gendered racial microaggressions to influence Black women’s sleep health.

Of note, the strong Black woman and angry Black woman stereotypes were associated with sleep health, but the effects were relatively smaller compared to the other microaggression subscales. On one hand, although these aspects of gendered racism are relevant for Black women, being perceived as “strong” is also perceived as a desirable trait for Black women in some contexts (Beauboeuf-Lafontant 2009). Moreover, being typecast as “angry” (or even potentially violent) is also ascribed to Black men (Wingfield 2007). On the other hand, to be deemed physically unattractive yet sexually available is rooted in the historical exploitation of Black women’s bodies in the antebellum period to the present (Collins 2000) and is unique to their gendered and racialized oppression. In addition, silencing and marginalization are also related to Black women’s unique intersectional invisibility because they do not fall within the prototypical categories associated with race and gender (Crenshaw 1991; Purdie-Vaughns and Eibach 2008). For these reasons, perhaps assumptions of beauty and sexual objectification and silenced and marginalized had the strongest negative association with Black women’s sleep health.

The third aim of our study was to ascertain whether the association between gendered racial microaggressions and sleep health persisted after adjusting for mental health. Nearly half of the association between gendered racial microaggressions and sleep was attenuated after accounting for depressive symptoms and worry. Our results are consistent with prior research reporting mental health as a mechanistic factor linking stress exposure to sleep health among Black Americans (Beatty et al. 2011; Hart et al. 2021; Hoggard and Hill 2018). This finding also extends the small but growing literature on gendered racial microaggressions by showing that psychological health acts as a potential underlying mechanism affecting sleep.

With regard to specific gendered racial microaggressions dimensions, assumptions of beauty and sexual objectification and silenced and marginalized had the strongest association with sleep even after adjusting for negative affect (i.e., depressive symptoms), which suggests that other mechanisms underlie this relationship. However, mental health was a potential pathway linking strong Black woman and angry Black woman stereotypes to sleep. Being perceived as strong and angry is linked to poor mental health among Black women (Thomas et al. 2004; Woods-Giscombé 2010). Although one stereotype is ostensibly “positive” (i.e., being strong) while the other is “negative” (i.e., being angry), both appear to be demoralizing and denigrating for Black women, which, in turn, influences sleep via depressive symptoms and worry.

We did not observe significant associations between education, income, and sleep health. Prior research suggests that higher levels of education are generally beneficial to the health of White individuals but not Black individuals, and research has documented this with sleep (Sheehan et al. 2020). However, studies of socioeconomic status and sleep health in cohorts with Black women have produced mixed results, with some finding null associations (Hall et al. 2009), others finding positive associations (Nguyen 2022), and some reporting negative associations (Jackson et al. 2013; Johnson et al. 2016). Although beyond the scope of the current analysis, additional research in this area is needed.

To our knowledge, this study is the first to assess the extent to which gendered racial microaggressions are related to Black women’s sleep health and the potential mediating role of mental health. In doing so, we heed the call of prior research to incorporate intersectional theorizing into empirical research on racial discrimination (Bobo and Fox 2003; Harnois and Ifatunji 2011) and to theoretically enhance the social determinants of sleep literature using an intersectional perspective (Johnson et al. 2019). Future research should build on this study by ascertaining whether other race-related stressors uniquely influence Black Americans’ sleep health. For instance, Hicken and colleagues (2013) discovered that race-related vigilance was associated with poor sleep among Black Americans and that vigilance explained Black-White disparities in sleep quality. Intersectional stress exposures unique to Black women’s experiences should also be explored. For example, sleep health could be related to “shifting,” a phenomenon among Black women that entails changing one’s appearance and speech patterns to appear acceptable or at least more favorable to White people’s sensibilities (Jones and Shorter-Gooden 2003). In sum, this study contributes to analyses of sleep health that account for intersectional measures of stress exposure.

Despite the study strengths, there are some limitations. First, although our sleep health measure was acquired from a validated questionnaire, the domains were self-reported and subject to bias. Self-reported sleep measures are less precise compared to objective measures like actigraphy assessments of sleep duration (Jackson et al. 2020). For example, the questionnaire references the past four weeks, which may not be representative of daily sleep habits. The nuances of weekday versus weekend sleep were also not captured with this measure of global sleep. Nevertheless, longitudinal research demonstrates that discrimination stress influences sleep quality over time (Johnson et al. 2021). Second, the data are based on a select sample of Black women from the Atlanta area, designed to maximize within-group heterogeneity in terms of socioeconomic status and geographic representation (e.g., urban and suburban environments). Although this enhances within-group variability among Black women, it limits generalizability of the results. Atlanta, Georgia, however, is an ideal location in which to study Black women because of the large African American population and its socioeconomic diversity (U.S. Census Bureau 2019). The sample also included a number of health-related exclusion criteria (women with hypertension and diabetes were not excluded, however). Thus, findings may be a conservative estimate of the association between gendered racial microaggressions and sleep health because Black women with severe health problems were not represented in this sample. Given the novelty of the gendered racial microaggressions measure, we encourage the inclusion of gendered racial microaggressions in nationally representative surveys that will enable researchers to investigate how gendered racial microaggressions are linked to health among U.S. Black women more generally. Fourth, given the cross-sectional nature of the data, results regarding the attenuating effects of mental health should be interpreted with caution. In addition, the complexity and challenge of assessing indirect effects for discrete outcome measures make the results presented here suggestive, providing impetus for future research on this topic.

Study results have implications for social psychological research on race, racism, and discrimination as well as research on stress and health. We present strong evidence that microaggressions are indispensable indicators of discrimination, especially given their implications for health. Until recently, social psychological research on discrimination has tended to unidimensionally focus on race and racism. Nevertheless, measures of discrimination that focus on unfair treatment based on racial identity masks discrimination based on other status characteristics (Harnois 2022; Harnois and Ifatunji 2011). Here, we respond to the call of Bobo and Fox (2003) for more theoretical bridges by integrating the intersectionality paradigm with social psychological research on discrimination. In doing so, we focus on a novel measure of microaggressions that operationalizes subtle gendered racism relevant to Black women.

With regards to research on stress and health, this study demonstrates gendered racial microaggressions comprise an important contribution to the small, but growing, literature on intersectional stress exposure. Such measures will allow for a more precise assessment of stress exposure among particular groups occupying a specific constellation of disadvantaged social identities (Ching et al. 2018; Erving, Patterson, and Boone 2021). Future research on stress and health should develop other intersectional stress measures relevant to the specific social identities of other historically disadvantaged groups (e.g., sexual minority Black men). Study results demonstrate the importance of investigating intersectional stress exposures that may operate as mechanistic factors undergirding intersectional disparities in physical and psychological health.

Biographies

BIOS

Christy L. Erving is an associate professor in the Department of Sociology and Population Research Center at The University of Texas at Austin. Using theories, concepts, and perspectives from various disciplines, her research focuses on clarifying and explaining status distinctions in health. Her primary research areas explore how race, ethnicity, gender, and immigrant status intersect to produce health differentials; psychosocial determinants of Black women’s health; and the Black–White mental health paradox. Her research has appeared in Journal of Health and Social Behavior, Society and Mental Health, American Journal of Epidemiology, and Sociology of Race and Ethnicity.

Rachel Zajdel is a postdoctoral fellow in the Minority Health and Health Disparities Population Laboratory, housed within the National Heart, Lung, and Blood Institute at the National Institutes of Health. Her research focuses on the social and legal determinants of racial/ethnic and immigrant health disparities. Her work has appeared in Journal of Racial and Ethnic Health Disparities and Work and Occupations.

Izraelle I. McKinnon is a social epidemiologist whose primary areas of research include identifying and addressing racial and geographic disparities in health, including community-level factors that promote health among Black communities. Dr. McKinnon supported this work as a graduate researcher in the Department of Epidemiology at Emory University Rollins School of Public Health. Her research has appeared in Health and Human Rights Journal, Psychosomatic Medicine, and Social Science & Medicine.

Miriam E. Van Dyke is a social epidemiologist focused on identifying and addressing health inequities at the intersection of race, place, and class. Dr. Van Dyke supported this work as a graduate researcher in the Department of Epidemiology at Emory University Rollins School of Public Health.

Raphiel J. Murden is an assistant professor in the Department of Biostatistics & Bioinformatics at Emory University’s Rollins School of Public Health. Dr. Murden’s research and peer-reviewed publications focus on the development and application of statistical methods for analyzing complex data arising in studies that examine neuroimaging, cardiovascular disease and associated risks, and HIV, with particular attention to medication adherence. His research has appeared in NeuroImage, Health Literacy, Research, & Policy, and Social Science & Medicine.

Dayna A. Johnson is a sleep epidemiologist and assistant professor in Department of Epidemiology at Emory University’s Rollins School of Public Health. Prior to her current appointment, she held a postdoctoral fellowship in sleep and circadian disorders at Harvard Medical School. Dr. Johnson’s research explicates the social contributors to racial/ethnic disparities in sleep by quantifying the contribution of social, household-level, and neighborhood-level factors using data from epidemiologic cohort studies. She is also exploring how stress reduction programs and improvements in the home environment can improve sleep and reduce risk of poor health outcomes.

Reneé H. Moore is a research professor, director of the Biostatistics Scientific Collaboration Center, and director of Diversity, Equity & Inclusion for the Department of Epidemiology and Biostatistics at Drexel University’s Dornsife School of Public Health. Dr. Moore’s research interests are in the design, conduct, and analysis of clinical trials and statistical applications to obesity, sleep apnea, and health disparities. She has a wealth of experience as a biostatistician collaborating with clinicians, public health practitioners, and scientists. Dr. Moore also dedicates time to recruiting, training, and retaining the next generation of collaborative biostatisticians and users of statistics.

Tené T. Lewis is an associate professor in the Department of Epidemiology in the Rollins School of Public Health at Emory University. Her research focuses on understanding how psychological and social factors contribute to the disproportionately high rates of cardiovascular disease morbidity and mortality observed in African American women compared to women of other racial/ethnic groups. She is currently Principal Investigator of two National Institutes of Health R01-funded cohorts examining the impact of discrimination and other race-related psychosocial stressors on various indices of cardiovascular health in healthy African American women and African American women with systemic lupus erythematosus.