Stereotypes of Black American Women Related to Sexuality and Motherhood

Abstract

Intersectionality theorists and researchers suggest the importance of examining unique stereotypes associated with intersecting group identities. We focus on the unique stereotypes of Black women in the United States related to sexuality and motherhood. In an online experimental study, 435 undergraduates from a Northeastern U.S. university were randomly assigned to one of the four conditions in which they viewed a photograph and read a description of a target young woman. The target’s race (Black vs. White) and pregnancy status (pregnant vs. no pregnancy information) were varied. A Black female target (pregnant or not) was perceived more negatively on items related to historically rooted societal stereotypes about sexual activity, sexual risk, motherhood status, and socioeconomic status than was a White female target, but there were no differences on items unrelated to societal stereotypes. A Black target described as pregnant was also perceived as more likely to be a single mother and to need public assistance than was a White target described as pregnant. Current findings, along with evidence that societal stereotypes have damaging effects, underscore the importance of diversifying images of Black women and increasing awareness of how stereotypes affect perceptions of Black women. Findings also highlight the value of research employing intersectionality to understand stereotypes.

From the mammies, jezebels, and breeder women of slavery to the smiling Aunt Jemimas on pancake mix boxes, ubiquitous Black prostitutes, and ever-present welfare mothers of contemporary popular culture, negative stereotypes applied to African-American women have been fundamental to Black women’s oppression.

(Collins, 2000, p. 5)

Societal stereotypes about a wide range of social groups persist in the United States, are promoted or reinforced through the mass media, and can have numerous detrimental consequences (e.g., see Dovidio & Gaertner, 2010; Rosenthal & Overstreet, 2016). The majority of research studies on stereotypes have focused on single identities, such as that of a marginalized racial or ethnic group or a gender group. However, as intersectionality theory suggests (e.g., Cole, 2009; Collins, 2000; Crenshaw, 1993; Warner, 2008), many societal stereotypes uniquely pertain to the intersection of multiple identities, including race or ethnicity in combination with gender (e.g., Ghavami & Peplau, 2013; Thomas, Witherspoon, & Speight, 2004).

Our broader interest is in whether the unique stereotypes of Black American women, due to their intersecting racial identity and gender identity, may contribute to persistent, disproportionately high rates of adverse sexual and reproductive health outcomes among Black women in the United States (Rosenthal & Lobel, 2011). Because of this focus, the research referenced throughout this article and the study participants are from the United States. Black Americans experience an infant mortality rate that is more than twice as high as that experienced by White Americans and that disparity is mostly explained by group disparities in low birth weight and preterm delivery (Martin, Hamilton, Ventura, Osterman, & Mathews, 2013; Murphy, Xu, & Kochanek, 2013). Also, the rate of new HIV infections in the United States among Black women was more than 20 times the rate among White women in 2010 (Centers for Disease Control and Prevention [CDC], 2012). The U.S. federal government has increased its interest and investments in identifying factors that contribute to and can reduce existing health disparities, such as those in women’s sexual and reproductive health (U.S. Department of Health and Human Services, 2011). There is increasing evidence that stereotypes and discrimination play an important role in these sexual and reproductive health outcomes (e.g., Duvall et al., 2013; Earnshaw et al., 2013; Giscombé & Lobel, 2005; Rosenthal et al., 2014). In previous work, we proposed that it is critical to examine the unique stereotypes of Black women that are related to sexuality and motherhood, in order to understand racial disparities in women’s sexual and reproductive health (Rosenthal & Lobel, 2011). Examining these unique stereotypes is also important more broadly, because the assumptions people make about Black women have a wide range of implications for the ways that Black women are treated in our society (e.g., Rosenthal & Lobel, 2011; Rosenthal & Overstreet, 2016).

In the current investigation, we conducted an experiment with a large sample of undergraduates attending a university in the Northeastern United States. The aim was to explore whether individuals’ perceptions of a Black female target compared to a White female target, described as pregnant or with no information about pregnancy status, are consistent with stereotypes about the sexuality and motherhood of Black women in the United States. This investigation draws on the work of Black feminist theorists and intersectionality theory (e.g., Cole, 2009; Collins, 2000; Crenshaw, 1993). We sought to examine the stereotypes people have about Black women related to their sexuality and motherhood and how those stereotypes affect perceptions of Black women who are pregnant or whose pregnancy status is not known. We examined how stereotypes shape social categories, such as race and pregnancy, and people’s perceptions of women in the United States.

Stereotypes and Their Effects

We adopted a standard and accepted definition of stereotypes as, “beliefs about the characteristics, attributes, and behaviors of members of certain groups” (Hilton & von Hippel, 1996, p. 240). Stereotypes are sometimes based in some amount of real difference that distinguishes groups, and sometimes they are not (Hilton & von Hippel, 1996). Even if stereotypes reflect some amount of reality, they can lead to inaccurate overgeneralizations to all members of a group and to ignoring other important information about individuals, with a variety of adverse consequences (e.g., Rosenthal & Overstreet, 2016; Shavers et al., 2012; Steele, 1997). We are all exposed to, and learn, societal stereotypes about groups. Although individuals with low levels of prejudice often consciously try to avoid the activation and application of stereotypes, even those individuals demonstrate automatic stereotype activation when cognitive resources are limited (e.g., Devine, 1989). Stereotypes, either consciously or unconsciously, can affect people’s judgments of and behaviors toward others. Stereotypes have a connection to decision-making, such as decisions about public policies (Kunda & Spencer, 2003). For example, endorsement of stereotypes about gay men and lesbian women has been associated with less support for gay and lesbian rights (Wood & Bartkowski, 2004); endorsement of stereotypes about older workers has been associated with more negative views toward working with, and the organizational treatment of, older workers (Chiu, Chan, Snape, & Redman, 2001); endorsement of stereotypes about Black Americans related to responsibility has been associated with greater opposition to affirmative action for Black Americans (Reyna, Henry, Korfmacher, & Tucker, 2005); and holding implicit stereotypes (i.e., stereotypes outside of conscious awareness assessed with an implicit associations test) about individuals who are overweight has been associated with being less likely to invite a job applicant who is overweight for an interview (Agerström & Rooth, 2011). Thus, stereotypes have important consequences in many realms of life.

There is evidence that when health-care professionals, consciously or unconsciously, hold stereotypes, their attitudes can lead to discrimination in provision of care and in health outcomes for patients, contributing to health disparities (Shavers et al., 2012). A recent study found that medical students’ racial stereotypes resulted in them being less willing to prescribe pre-exposure prophylaxis for HIV (use of medication to prevent HIV infection among those at high risk) to Black than to White male hypothetical patients (Calabrese, Earnshaw, Underhill, Hansen, & Dovidio, 2014). Research demonstrating how stereotypes affect medical and other decision-making highlights the importance of identifying unique stereotypes of Black women in the United States.

Whereas there is already much research documenting stereotypes that exist about various social groups and the consequences of those stereotypes, the majority of this research has focused on single identities. We have very limited understanding of the unique stereotypes of groups with multiple marginalized identities, including Black women. In the current study, we used an intersectionality framework to explore the extent to which people’s perceptions of Black women, who are known to be pregnant or whose pregnancy status is unknown, are influenced by unique stereotypes about Black women related to sexuality and motherhood.

Intersectionality and Stereotypes of Black Women in the United States

Intersectionality theory is rooted in the experiences and ideas of Black activists and scholars, particularly Black feminist activist-scholars who sought to bring to light the marginalization of Black women’s experiences and the interconnections among multiple systems of oppression (e.g., Collins, 2000; Combahee River Collective, 1995; Crenshaw, 1993, 1995; Davis, 1981; hooks, 1989). We use intersectionality as a framework because it is particularly relevant to understanding stereotypes about Black women and their sexual and reproductive health outcomes. Intersectionality draws our attention to the importance of multiple intersecting identities (e.g., being Black, being a woman, and being pregnant) to understand stereotypes. Intersectionality also highlights that the roots of stereotypes are in historical and contemporary systemic oppression (e.g., Cole, 2009). It also offers a strong framework of theory and research that has delineated the unique stereotypes and other historical and contemporary experiences of Black women (e.g., Alexander-Floyd, 2012; Collins, 2000; Thomas et al., 2004). Intersectionality theory grows out of an activist tradition that impels us not only to document the existence of stereotypes and their connections to consequences such as adverse health outcomes, but also to use the research we conduct to advocate for social and structural change to reduce those stereotypes and other manifestations of oppression (Alexander-Floyd, 2012; Rosenthal, 2015), which is imperative given the large and persistent sexual and reproductive health disparities faced by Black women (CDC, 2012; Martin et al., 2013; Murphy et al., 2013).

Researchers in history, law, philosophy, psychology, public health, sociology, and women’s studies (e.g., Cole, 2009; Collins, 2000; Crenshaw, 1993; Davis, 1981; Essed, 1991; Thomas et al., 2004; Thomas, Witherspoon, & Speight, 2008; Townsend, 2008) have drawn on intersectionality theory to focus on the intersection of race and gender in creating unique experiences, including those related to stereotypes, that affect Black women in the United States. Certain stereotypes of Black women are similar to, or the same as, stereotypes of Black men or White women, but there are also unique stereotypes of Black women that are not applied to Black men or to White women. Three prevalent, pernicious, images or archetypes of Black women that have been promulgated dating back to the period of American slavery are the “mammy,” “sapphire,” and “jezebel” (e.g., see West, 2008; Woodard & Mastin, 2005). The mammy archetype is the image of an unattractive Black mother who is strong and content in her caregiving role for many children, in the service of White slave owners or White employers. The sapphire (or “matriarch”) archetype is the image of an aggressive, dominating, angry, emasculating Black woman. The jezebel (or “sexual siren”) is the image of an immoral, sexually promiscuous, and sexually available Black woman (West, 2008; Woodard & Mastin, 2005). Another more recent archetype of Black women is that of the “welfare queen,” which is connected to images of Black women as “breeders” dating back to slavery (Collins, 2000). The welfare queen is an image of an uneducated, poor, single Black woman who does not want to work but has many children in order to take advantage of public assistance (Woodard & Mastin, 2005). An understanding of these archetypes requires an intersectional analysis of the multiple stereotypical images of Black women that are distinct from those of Black men and White women, and the connection of these stereotypes to historical and contemporary structural oppression of Black women.

There is evidence that Black women are aware of unique stereotypes about them and have experiences during which others perceive or treat them differently, based on those stereotypes. Black female academics report stories of people mistaking them for waitresses, administrative staff, or prostitutes (Byrd & Solomon, 2005; Thomas et al., 2004; West, 2008). There also is evidence that the archetypes associated with Black women continue to be relevant; research shows that current stereotypes of Black women have close connections with the historical, stereotypical archetypes or images (e.g., Donovan, 2011; Ghavami & Peplau, 2013; Goff, Thomas, & Jackson, 2008). For example, Donovan (2011) found that Black women, compared to White women, were stereotyped as stronger and more domineering, consistent with the sapphire archetype. Goff, Thomas, and Jackson (2008) found that Black women were perceived as less attractive than White women; Black women were perceived as more masculine, because of the strong association that people hold between “Blackness” and “maleness,” which is also consistent with the sapphire archetype. More specifically related to sexuality, Black American women continue to be stereotyped as promiscuous, hypersexual, sexually available, and as having “animalistic” sexuality, all of which have a long history connected to the sexualized exploitation of Black women during slavery and are consistent with the jezebel archetype (Collins, 2000; Davis, 1981; hooks, 1990; Thomas et al., 2004). Black American women also continue to be stereotyped as poor, uneducated, young, single mothers who sleep with and use men for money (“gold diggers”); they are stereotyped as purposefully having children to take advantage of public assistance programs, and as being unable to have the financial resources to adequately care for their children (Collins, 2000; Roberts, 2002). All these perceptions are consistent with the welfare queen archetype. A recent study conducted with a sample of racially/ethnically diverse undergraduates found that the societal stereotypes associated with particular racial or ethnic × gender groups (e.g., Black women and Latino men) were distinct from, and not simply the sum of, stereotypes associated separately with particular gender groups (e.g., women and men) or racial or ethnic groups (e.g., Black and Latino; Ghavami & Peplau, 2013). Participants spontaneously reported unique stereotypes about Black women being overweight, confident, assertive, promiscuous, not feminine, and aggressive, as well as having “big butts,” “wearing hair weaves,” and “liking to eat fried chicken” (Ghavami & Peplau, 2013). Although the study was not specifically focused on stereotypes about sexuality and motherhood, the results indicated that there are indeed unique stereotypes about Black women in these domains that emerged spontaneously and that are consistent with the archetypes identified above.

Although there is increasing interest in psychology and other fields in incorporating intersectionality theory (Cole, 2009; Warner, 2008) and researchers have begun to explore the intersections of gender and race in societal stereotypes (see also Galinsky, Hall, & Cuddy, 2013), most empirical research on stereotypes has still focused on single identities. Little stereotype research has drawn on interdisciplinary studies to pinpoint historically relevant stereotypes of groups, particularly multiply marginalized groups such as Black women (Ghavami & Peplau, 2013; Purdie-Vaughns & Eibach, 2008). To the best of our knowledge, no research to date has examined pregnancy as another identity or status that may by itself, or in combination with other identities or statuses such as race, contribute to stereotypes about women.

Many women report that pregnancy is a pivotal period in their lives, as it involves significant changes in physical appearance, roles and responsibilities, self-perceptions, and social relationships, and it heralds further changes to identity and status that are associated with motherhood (Lobel, 1998; Lobel, Hamilton, & Cannella, 2008). These changes are prominent topics of concern to pregnant women (e.g., Alderdice, Lynn, & Lobel, 2012; Lobel et al., 2008). Pregnant women are often treated distinctively by others, sometimes with particular kindness but also sometimes with disapproval and denigration (especially if a pregnant woman is young or single), infantilizing (e.g., uninvited touching), and even by maintaining greater physical distance from them (Lobel, 1998). These observations of others’ reactions to pregnant women and women’s own perceptions substantiate the view that pregnancy is a contributor to identity and status. Pregnancy, motherhood, and caregiving are also particularly important themes connected to the historical stereotyping of Black American women, as demonstrated in the sapphire and welfare queen archetypes; the theme of sexual promiscuity, demonstrated in the jezebel archetype, is also connected to the potential for pregnancy. Thus, the intersecting role of pregnancy status with race is critical to fully understand unique stereotypes of Black women; we examine these stereotypes in the current investigation.

The Current Investigation

In the current investigation, we manipulated a female target’s race (image of Black woman’s face vs. White woman’s face) and pregnancy status (information indicating target is pregnant vs. no information about pregnancy status). We examined whether manipulation of the female target’s race and pregnancy status would affect participants’ perceptions of the female target related to marital/relationship status, sexual activity, protection/risk during sex, motherhood status, socioeconomic status, other health behaviors, compliance with health care, single motherhood, needing public assistance for one’s child, and health behaviors during pregnancy. We selected items related to these themes because of their relevance to sexual and reproductive health as well as their connections to the historical and contemporary stereotypes of Black women described above. Based on the stereotypes of Black American women about sexuality and motherhood that are connected to the jezebel and welfare queen archetypes, we hypothesized that a Black female target would be more likely to be perceived by participants as sexually promiscuous (engaging in more sexual activity), engaging in more unprotected sexual activity, currently having children or having been pregnant in the past, and having lower socioeconomic status, than a White female target.

To investigate whether perceptions of Black women are distinctly related to these historically rooted stereotypes about sexuality and motherhood, as opposed to whether all characteristics associated with Black women are viewed negatively, we also assessed perceptions of the target’s involvement in other stigmatized health behaviors (alcohol and drug use) and compliance with general and pregnancy-specific health-care recommendations. We did not expect to find differences in these perceptions because they are not connected to historically rooted stereotypical images of Black women and have not been found in more recent research on stereotypes of Black women (e.g., Ghavami & Peplau, 2013). In addition, based on the stereotypes of Black American women that are specific to motherhood, and thus most relevant to pregnancy, we hypothesized that a pregnant Black female target would be more likely to be perceived by participants as being a single mother (not having the father involved) and needing public assistance than a pregnant White female target, but that there would not be any differences in perceptions about her health behaviors during pregnancy.

We also examined main effects of pregnancy status as well as interactions between the race and pregnancy status of the target. Although we did not have specific hypotheses about interactions, we explored whether pregnancy status is another identity or a status that interacts with race in contributing to stereotypical perceptions of women. Our experimental design, in which everything about the target woman is the same, except for random assignment to variations on her race and pregnancy status, also allowed us to tease apart stereotypes that are specific to these identities. We avoided the co-occurrence of race and pregnancy status with other characteristics that may be apparent when women are encountered in daily living and may also affect others’ views of them, such as their socioeconomic status, age, or occupation. Although our focus is specifically on stereotypes of Black women, we chose to compare perceptions of a Black female target to a White female target to verify that these stereotypes are indeed unique to Black women and not applied to women more broadly.

The current investigation was designed to make several contributions to the existing literature. This study draws on intersectionality theory to focus on unique stereotypes about Black women due to the intersection of gender and race that are related to sexuality and motherhood. It moves beyond documenting whether people are aware of existing stereotypes of Black women and investigates whether people apply stereotypes in making judgments about Black women, and whether the stereotypes are applied broadly to all judgments or selectively applied based on the specificity of historically rooted societal stereotypes. Finally, the current study explicitly examines—to our knowledge, for the first time—the role of pregnancy status in interaction with a woman’s race in affecting people’s perceptions of Black and White women.

Method

Participants

A total of 474 undergraduates at a mid-sized public university in the Northeastern United States were recruited through the psychology department’s subject pool to complete an online survey. However, 39 participants failed a manipulation test (21 for the White target and 18 for the Black target), which was a question that asked participants to identify the race/ ethnicity of the target, Jasmine. Thus, the final analytic sample included 435 participants (291 or 66.9% female, 139 or 32% male, 5 or 1.1% not reported; mean age = 20.48, SD = 3.39). Participants reflected the composition of the campus population (186 or 42.8% White/European American, 146 or 33.6% Asian American, 36 or 8.3% Latino American, 28 or 6.4% Black/African American, 39 or 9% Multiracial/ethnic or Other; 22 or 5.1% reported being “lower class,” 105 or 24.1% reported being “lower middle class,” 228 or 52.4% reported being “middle class,” 74 or 17% reported being “upper middle class,” 6 or 1.4% reported being “wealthy”). Participants received psychology subject pool credit.

Procedure

The investigation involved a 2 × 2 between-subjects experimental design with race (image of Black woman’s face vs. White woman’s face) and pregnancy status (information indicating target is pregnant vs. no information about pregnancy status) of the target woman randomly varied. Participants logged onto the secure study website, and first completed an online consent form, in which they were told that the purpose of the study was to understand people’s perceptions and beliefs about other people. Then they were randomly assigned to one of the four conditions. In all conditions, participants viewed an image of a target woman named Jasmine and read a brief description of her. Race of the target woman was manipulated by displaying an image at the top of each survey page in which the target was Black or White. The images were created using FaceGen (©Singular Inversions Inc.) software; faces were identical except that skin color and other racially relevant facial features, such as nose and lips, were varied to cue whether the image was of a Black versus White woman (see Appendix). Pregnancy status of the target woman was manipulated by a brief description. For a pregnant target, participants read the description, “Jasmine is a 25 year-old woman who was born and has lived her whole life in New York City. Jasmine just came from a doctor’s appointment where she found out that she is pregnant.” For a target for whom no information about pregnancy status was given, participants read the description, “Jasmine is a 25 year-old woman who was born and has lived her whole life in New York City. Jasmine just came from a doctor’s appointment.” Thus, the four conditions involved participants being exposed to a pregnant Black female target (n = 110), a Black female target for whom no information about pregnancy status was given (n = 96), a pregnant White female target (n = 110), and a White female target for whom no information about pregnancy status was given (n = 119).

After exposure to the image and brief description, participants read instructions saying, “Please take a few moments to think about Jasmine and what type of person she probably is. Then, please answer the following questions, based on your thoughts about Jasmine.” All participants completed the same questions asking about their perceptions of the target; those who were told that the target was pregnant answered additional questions pertinent to pregnancy. Participants then answered several sociodemographic questions. Finally, participants were debriefed, told that the true purpose of the study was to see if people’s perceptions of one of the four possible women vary based on other information given about her, and participants were thanked for their participation. The institutional review board at the university where the study was conducted approved all procedures.

Measures

Participants completed the following questions after viewing the image and description of the target woman, in the order described below. Items evaluating perceptions of the female target were developed for this study to assess perceptions related to several categories: marital/relationship status, sexual activity, protection/risk during sex, motherhood status, socioeconomic status, other health behaviors, compliance with health care, single motherhood, needing public assistance for one’s child, and health behaviors during pregnancy. After completing those items, participants reported their age, gender, race/ethnicity, and socioeconomic status.

Perceptions of target (all conditions)

All participants reported on a scale from 1 (not likely at all) to 4 (very likely) how likely they thought several things were to be true about the target woman: she “is married,” she “is in a serious relationship,” she “uses some form of birth control or contraception regularly,” she “uses condoms regularly during sexual activity,” she “has children,” she “has been pregnant some time in the past,” “it is a good idea for her to be a mother right now,” she “is currently employed full-time,” she “receives some sort of public assistance,” she “smokes cigarettes regularly,” she “drinks alcohol regularly,” she “will follow the instructions that her doctor told her during the appointment she just came from,” and she “will show up to the follow-up appointment her doctor scheduled for her.” The 4-point response scale for these items was used to avoid participants choosing a “neutral” response in the middle for all items; they indicated their perception of the target with little information given about her. In addition, all participants were asked to “guess” the following about the target: at what age she became sexually active, how many people she has had sex with in her life and in the past month, her highest level of education (ranging from 1 = some high school but not a high school diploma to 6 = some graduate school or professional degree), and the salary she earns per year.

Perceptions of target (pregnant conditions only)

Only participants who rated a pregnant target reported on a scale of 1 (not likely at all) to 4 (very likely) how likely they thought several things were true about the target: “the father of her baby will play a role in raising the child,” she “would make a good mother,” she “will need some sort of public assistance if she has her baby,” she “will smoke cigarettes while she is pregnant,” and she “will drink alcohol while she is pregnant.”

Manipulation check

All participants were asked to indicate what they perceived Jasmine’s race/ethnicity to be based on the photograph, and responses confirmed that the manipulation was successful, with 92% of participants correctly identifying Jasmine as Black versus White in the appropriate conditions. Those who did not correctly identify Jasmine’s race/ethnicity were not included in analyses reported. However, removing them from the analyses did not change the overall pattern of findings.

Results

All analyses were conducted using SPSS Version 22.0 (IBM Corp, 2013). Mean responses to all questions by condition are in Table 1.

Table 1.

Means (and Standard Deviations) by Condition.

	Pregnant Black Target	No Pregnancy Information Black Target	Pregnant White Target	No Pregnancy Information White Target
Marital/relationship status
She is marrieda	2.32 (0.80)	2.26 (0.67)	2.35 (0.74)	2.08 (0.65)
She is in a serious relationshipa	2.70 (0.81)	2.63 (0.62)	2.88 (0.69)	2.66 (0.68)
Sexual activity
At what age she became sexually active	17.02 (2.46)	17.60 (2.44)	17.26 (1.69)	17.36 (1.59)
How many people she has had sex with in her lifea	6.14 (5.09)	5.24 (3.38)	6.51 (10.69)	4.34 (2.35)
How many people she has had sex with in the past monthb	1.31 (0.66)	1.35 (1.19)	1.14 (0.39)	1.09 (0.49)
Protection/risk during sex
She uses some form of birth control or contraception regularlya,b	2.25 (0.90)	2.70 (0.77)	2.31 (0.88)	2.97 (0.83)
She uses condoms regularly during sexual activitya	2.24 (0.86)	2.67 (0.82)	2.38 (0.84)	2.82 (0.72)
Motherhood status
She has childrenb	2.05 (0.82)	2.10 (0.84)	1.60 (0.65)	1.69 (0.71)
She has been pregnant some time in the pastb,c	2.05 (0.78)	1.95 (0.79)	1.66 (0.71)	1.87 (0.78)
Perception of motherhood
It is a good idea for her to be a mother right nowa,c	2.28 (0.78)	2.25 (0.81)	2.48 (0.83)	2.03 (0.86)
Socioeconomic status
She is currently employed full-timea	2.76 (0.81)	2.99 (0.72)	2.90 (0.70)	2.98 (0.69)
She receives some sort of public assistanceb	2.02 (0.84)	1.86 (0.66)	1.65 (0.69)	1.54 (0.64)
Her highest level of educationa,b	3.75 (1.26)	4.31 (1.04)	4.40 (1.03)	4.66 (0.80)
The salary she earns a yearb	36,118 (20,222)	38,885 (14,064)	41,718 16,503)	42,782 (16,760)
Other health behaviors
She smokes cigarettes regularly	1.75 (0.73)	1.61 (0.61)	1.76 (0.72)	1.85 (0.74)
She drinks alcohol regularlya,b	1.94 (0.76)	2.01 (0.75)	2.03 (0.76)	2.28 (0.82)
Compliance with health care
She will follow the instructions that her doctor told her during the appointment she just came fromc	3.01 (0.88)	3.22 (0.74)	3.30 (0.77)	3.09 (0.77)
She will show up to the follow-up appointment her doctor scheduled for her	3.15 (0.83)	3.27 (0.77)	3.37 (0.75)	3.25 (0.77)
Single motherhood
The father of her baby will play a role in raising the childb	2.40 (0.85)	—	2.77 (0.74)	—
Perception of motherhood current pregnancy
She would make a good mother	2.80 (0.81)	—	2.89 (0.65)	—
Public assistance for child
She will need some sort of public assistance if she has her babyb	2.23 (0.87)	—	1.86 (0.74)	—
Health behaviors during pregnancy
She will smoke cigarettes while she is pregnant	1.33 (0.61)	—	1.32 (0.57)	—
She will drink alcohol while she is pregnant	1.36 (0.66)	—	1.31 (0.60)	—

Note. All items were rated on a 1 (not likely at all) to 4 (very likely) scale, except the target’s highest level of education was rated on a 1 (some high school but not a high school diploma) to 6 (some graduate school or professional degree) scale, and at what age she became sexually active, how many people she has had sex with in her life, how many people she has had sex with in the past month, and the salary she earns a year used open-ended, continuous responses. N = 435.

^aIndicates main effect of pregnancy status significant.

^bIndicates main effect of race significant.

^cIndicates interaction between race and pregnancy status significant.

Perceptions by Race and Pregnancy Status of Target for All Conditions

To test our hypotheses about unique stereotypes about Black women, we used multivariate analysis of variance (MANOVA) to test for main effects of race and pregnancy status of the target, and interactions between race and pregnancy status of the target for all of the questions that were asked of participants in all four conditions. Wilks’s λ statistics were examined for omnibus tests of effects. The omnibus main effects of race and pregnancy status were significant, Wilks’s λ = 0.82, F(18, 414) = 5.11, p < .001, ${\mathrm{\eta }}_{\mathrm{p}}^2=.18$ and Wilks’s λ = 0.76, F(18, 414) = 7.13, p < .001, ${\mathrm{\eta }}_{\mathrm{p}}^2=.24$, respectively. The omnibus interaction between race and pregnancy status did not reach significance, Wilks’s λ = 0.94, F(18, 414) = 1.58, p = .063, ${\mathrm{\eta }}_{\mathrm{p}}^2=.06$.

Perceptions by race of target

Compared to a White female target, participants rated a Black female target as having more sexual partners in the last month, F(1, 431) = 9.71, p = .002, ${\mathrm{\eta }}_{\mathrm{p}}^2=.02$; less likely to use some form of birth control regularly, F(1, 431) = 4.15, p = .042, ${\mathrm{\eta }}_{\mathrm{p}}^2=.01$; more likely to have children, F(1, 431) = 35.73, p < .001, ${\mathrm{\eta }}_{\mathrm{p}}^2=.08$; more likely to have been pregnant before, F(1, 431) = 9.63, p = .002, ${\mathrm{\eta }}_{\mathrm{p}}^2=.02$; more likely to be receiving public assistance, F(1, 431) = 26.39, p < .001, ${\mathrm{\eta }}_{\mathrm{p}}^2=.06$; having a lower level of education, F(1, 431) = 24.83, p < .001, ${\mathrm{\eta }}_{\mathrm{p}}^2=.05$; having a lower yearly salary, F(1, 431) = 8.32, p = .004, ${\mathrm{\eta }}_{\mathrm{p}}^2=.02$; and less likely to drink alcohol regularly, F(1, 431) = 5.77, p = .017, ${\mathrm{\eta }}_{\mathrm{p}}^2=.01$.

Perceptions by pregnancy status of target

Compared to a target for whom no information about pregnancy status was given, participants rated a pregnant target as more likely to be married, F(1, 431) = 5.69, p = .018, ${\mathrm{\eta }}_{\mathrm{p}}^2=.01$; more likely to be in a serious relationship, F(1, 431) = 4.95, p = .027, ${\mathrm{\eta }}_{\mathrm{p}}^2=.01$; having more sexual partners in her life, F(1, 431) = 6.45, p = .011, ${\mathrm{\eta }}_{\mathrm{p}}^2=.02$; less likely to use some form of birth control regularly, F(1, 431) = 46.28, p < .001, ${\mathrm{\eta }}_{\mathrm{p}}^2=.10$; less likely to use condoms regularly during sexual activity, F(1, 431) = 31.58, p < .001, ${\mathrm{\eta }}_{\mathrm{p}}^2=.07$; it being a better idea for her to be a mother right now, F(1, 431) = 9.56, p = .002, ${\mathrm{\eta }}_{\mathrm{p}}^2=.02$; less likely to be employed full-time, F(1, 431) = 4.83, p = .029, ${\mathrm{\eta }}_{\mathrm{p}}^2=.01$; having a lower level of education, F(1, 431) = 16.88, p < .001, ${\mathrm{\eta }}_{\mathrm{p}}^2=.04$; and less likely to drink alcohol regularly, F(1, 431) = 4.74, p = .030, ${\mathrm{\eta }}_{\mathrm{p}}^2=.01$.

Interactions between race and pregnancy status of target

There were significant interactions between race and pregnancy status of the target for 3 items: how likely it was that she had been pregnant before, F(1, 431) = 4.40, p = .037, ${\mathrm{\eta }}_{\mathrm{p}}^2=.01$; that it was good idea for her to be a mother right now, F(1, 431) = 7.23, p = .007, ${\mathrm{\eta }}_{\mathrm{p}}^2=.02$; and that she will follow the instructions her doctor told her, F(1, 431) = 7.44, p = .007, ${\mathrm{\eta }}_{\mathrm{p}}^2=.02$. To understand these interactions, we ran two follow-up MANOVAs examining the main effects of race of target separately for a pregnant target versus a target for whom no information about pregnancy status was given. The omnibus tests of the effects of race remained significant for both the pregnant and the no pregnancy information target, Wilks’s λ = 0.78, F(18, 201) = 3.12, p < .001, ${\mathrm{\eta }}_{\mathrm{p}}^2=.22$ and Wilks’s λ = 0.72, F(18, 196) = 4.24, p < .001, ${\mathrm{\eta }}_{\mathrm{p}}^2=.28$, respectively. Focusing specifically on the 3 items for which there were significant interactions between race and pregnancy status of the target (whether she has been pregnant before, whether it is a good idea for her to be a mother right now, and whether she will follow the instructions her doctor told her), we looked at the effects of race separately for the pregnant and the no pregnancy information target. The main effect of race for whether she has been pregnant before was significant when the target was pregnant, F(1, 218) = 14.41, p < .001, ${\mathrm{\eta }}_{\mathrm{p}}^2=.06$, but not when no information about pregnancy status of the target was given, F(1, 213) = 0.48, p = .491, ${\mathrm{\eta }}_{\mathrm{p}}^2=.002$. When the target was pregnant, participants rated the Black target as more likely to have been pregnant before than the White target. The main effect of race for whether it is a good idea for her to be a mother right now was nonsignificant for both the pregnant and the no pregnancy information target, F(1, 218) = 3.38, p = .067, ${\mathrm{\eta }}_{\mathrm{p}}^2=.02$ and F(1, 213) = 3.84, p = .051, ${\mathrm{\eta }}_{\mathrm{p}}^2=.02$, respectively. The main effect of race for likelihood she will follow the instructions her doctor told her was significant for the pregnant target, F(1, 218) = 6.76, p = .010, ${\mathrm{\eta }}_{\mathrm{p}}^2=.03$, but nonsignificant for the no pregnancy information target, F(1, 213) =1.48, p =.226, ${\mathrm{\eta }}_{\mathrm{p}}^2=.01$. When participants were told that the target was pregnant, participants rated the Black target as less likely to follow the instructions her doctor told her than the White target. However, these interactions should be interpreted with caution, as the omnibus MANOVA for the interaction between race and pregnancy status did not reach significance.

Perceptions by Race of Target for Pregnant Conditions

To further test our hypotheses about the unique stereotypes about Black pregnant women specifically, we ran another MANOVA examining the main effects of race of target for questions that only participants who rated a pregnant target were asked. The omnibus test of the effect of race was significant, Wilks’s λ = 0.91, F(5, 214) = 4.20, p = .001, ${\mathrm{\eta }}_{\mathrm{p}}^2=.09$. The individual significant effects of race indicated that participants who viewed a Black pregnant target rated it less likely that the father of the child would be involved in raising the child, F(1, 218) = 12.10, p = .001, ${\mathrm{\eta }}_{\mathrm{p}}^2=.05$, and more likely that she would need public assistance if she has her baby, F(1, 218) = 11.14, p = .001, ${\mathrm{\eta }}_{\mathrm{p}}^2=.05$, compared to those who viewed a White pregnant target.

Participants’ Gender, Race/Ethnicity, and Socioeconomic Status

We then tested whether participants’ own gender, race/ ethnicity, and socioeconomic status might play a role in their perceptions of the target. First, we examined whether controlling for participants’ gender, race/ethnicity (coded dichotomously as White participants vs. participants of color), or socioeconomic status (using the same categories as described above) as covariates in the analyses substantively changed the pattern of results, and it did not. The only differences when including all of these controls in analyses were that the race of target effect for likelihood of using some form of birth control regularly was no longer significant (p = .058), while significant effects for race of target on likelihood of using condoms regularly, F(1, 423) = 4.47, p = .035, ${\mathrm{\eta }}_{\mathrm{p}}^2=.01$, and for pregnancy status of target on likelihood of receiving public assistance, F(1, 423) = 2.13, p = .040, ${\mathrm{\eta }}_{\mathrm{p}}^2=.01$, emerged that were not previously significant. The Black target was rated as less likely than a White target to use condoms regularly, and a pregnant target was rated as more likely to be receiving some form of public assistance than a target for whom no information about pregnancy status was given.

Next, we tested whether participants’ gender, race/ ethnicity, or socioeconomic status moderated any of the effects found. There were no significant omnibus interaction effects with participants’ gender or their socioeconomic status. The omnibus test for the interaction of race of participant with pregnancy status of target was the only significant omnibus interaction with race of participant, Wilks’s λ = 0.91, F(18, 410) = 2.17, p = .004, ${\mathrm{\eta }}_{\mathrm{p}}^2=.09$. Examining the interaction effects for each of the items, the interaction of race of participant with pregnancy status of target was significant for how likely it is that the target has been pregnant before, F(1, 427) = 4.53, p = .034, ${\mathrm{\eta }}_{\mathrm{p}}^2=.01$; uses some form of birth control regularly, F(1, 427) = 15.38, p < .001, ${\mathrm{\eta }}_{\mathrm{p}}^2=.04$; and uses condoms regularly, F(1, 427) = 20.68, p < .001, ${\mathrm{\eta }}_{\mathrm{p}}^2=.05$. Follow-up analyses revealed that participants of color perceived a pregnant target as significantly less likely than a no pregnancy information target to have been pregnant before, F(1, 245) = 4.02, p = .046, ${\mathrm{\eta }}_{\mathrm{p}}^2=.02$, but this was not the case for White participants, F(1, 182) = 1.16, p = .282, ${\mathrm{\eta }}_{\mathrm{p}}^2=.01$. White participants also perceived a pregnant target as less likely to use condoms regularly than a no pregnancy information target, F(1, 182) = 54.72, p < .001, ${\mathrm{\eta }}_{\mathrm{p}}^2=.23$, but this was not the case for participants of color, F(1, 245) = 1.69, p = .195, ${\mathrm{\eta }}_{\mathrm{p}}^2=.01$. Finally, both White participants and participants of color perceived a pregnant target as less likely to use some form of birth control regularly than a no pregnancy information target, but this effect was stronger among White participants, F(1, 182) = 59.29, p < .001, ${\mathrm{\eta }}_{\mathrm{p}}^2=.25$, than among participants of color, F(1, 245) = 6.99, p = .009, ${\mathrm{\eta }}_{\mathrm{p}}^2=.03$.

Discussion

Consistent with our hypotheses, we found that among a sample of undergraduates, a Black female target, regardless of pregnancy status, was perceived as having had sex with more people in the past month, less likely to use birth control regularly during sex, more likely to have children and to have been pregnant some time in the past, more likely to receive some form of public assistance, to have lower education, and to earn less income per year than a White female target. These effects support our hypotheses that there are negative stereotypes about Black women related to sexuality, motherhood, and socioeconomic status that are consistent with the historical images of the jezebel and welfare queen archetypes. However, as expected, a Black target was not perceived more negatively than a White target on other stigmatized items, such as those relating to marital/relationship status, other health behaviors (smoking cigarettes and drinking alcohol regularly), or compliance with health care. The Black target was not perceived more negatively in all ways but was negatively judged in ways that reflect the unique, historically-rooted stereotypes that exist of Black women in the United States. When participants were told that the target was pregnant, the Black pregnant target was perceived as less likely to have the father of the child involved in raising the child and more likely to need public assistance to help with their child than the White pregnant target. This result is consistent with our hypotheses regarding the unique, historically-rooted stereotypes about Black American women related to single motherhood and public assistance and related to the welfare queen archetype. There were no differences by race of target in perceptions of whether she would make a good mother or of the target’s health behaviors during pregnancy (smoking cigarettes and drinking alcohol). These results extend prior research, build upon intersectionality theory, and provide evidence that people not only hold unique stereotypes about Black women—both pregnant and in general—that are related to sexuality and motherhood, but that these stereotypes are applied when making judgments about a Black female target.

A pregnant target, regardless of race, was perceived as being more likely to be married and in a serious relationship, to have had sex with more people in her life, not to use birth control or condoms regularly during sex, and for it to be a good idea for her to be a mother. A pregnant target overall was also perceived as less likely to be employed full-time and having less education as well as less likely to drink alcohol regularly. These findings suggest that young pregnant women are also evaluated based on distinct perceptions, such as being more likely to be married or in a relationship, to engage in unprotected sexual behaviors, and to have lower socioeconomic status. This offers support for our suggestion that being pregnant is itself a type of identity or status that elicits distinct perceptions by others.

Furthermore, although the overall test of the interaction between race and pregnancy status of target was not significant, there were some interactions between race and pregnancy status of target for specific items: whether she had been pregnant before, whether it was a good idea for her to be a mother now, and whether she would follow the doctor’s instructions. When no information about pregnancy status was given, there were no differences based on race of target for perceptions of whether she had been pregnant before or whether she would follow her doctor’s instructions. However, when participants were told that the target was pregnant, the Black target was perceived as more likely to have been pregnant before and less likely to follow the doctor’s instructions than the White target. Although we should interpret these findings with caution due to non-significance of the omnibus test, they support the idea that pregnancy is an identity or status that interacts with race to elicit unique stereotypes. A Black target tended to be viewed more negatively on these particular items only when she was pregnant, suggesting that pregnancy interacts with race to create a unique vulnerability to negative stereotypes for Black women. The main effects of target race on items about a pregnant target also reinforce the idea that pregnancy makes a key contribution to stereotypes about Black women, which is consistent with the historically rooted mammy, jezebel, and welfare queen archetypes.

Strengths, Limitations, and Future Directions

The current investigation provides evidence that unique stereotypes about Black women related to sexuality and motherhood influence people’s judgments of Black women in comparison to White women. Further, pregnancy is an important part of these stereotypes. Our findings corroborate and extend past research and support claims about the importance of intersectionality as a framework for understanding societal stereotypes and other experiences of stigmatization of Black women, in addition to other groups. Some scholars have recently been concerned about the “erasure” of Black women and the “flattening” or “depoliticizing” of intersectionality in some work by disconnecting it from its roots in Black women’s experiences of marginalization, activism, and struggles against oppression (e.g., see Alexander-Floyd, 2012; Bilge, 2013). To avoid these pitfalls, we drew on the interdisciplinary work of Black feminist scholar-activists who have focused on the historical and contemporary oppression of Black women to inform the stereotypes we studied. Moreover, we suggest that current study findings not only be considered a documentation of existing stereotypes of Black women related to sexuality and motherhood, but along with other past and current interdisciplinary research, be considered a call to action. Our findings suggest that stereotypical images of Black women that are historically rooted in the systemic oppression of Black women in the United States continue to affect people’s perceptions of and judgments about Black women; research demonstrates such judgments can have a range of adverse consequences (see Rosenthal & Overstreet, 2016, for a review). This work supports the continuing need to work toward eliminating stereotypical media images of Black women and to replace them with more diverse, positive, complex, and dynamic images that reflect the reality of Black women’s lives (e.g., see Mastro, 2015). Our study also supports the continuing need to increase awareness of the influence that stereotypes have on perceptions of Black women and to help people learn how to work toward reducing that influence and respond to Black women in more socially just ways.

Our manipulation of pregnancy status was merely a one-sentence description. Future work might examine whether stronger effects are found with pregnancy status manipulated by visual cues, such as having images of women with and without a visibly pregnant belly. This could potentially reveal stronger effects because visual cues of pregnancy may elicit stereotypes in people’s real-life interactions with pregnant women. Indeed, some of our past work has found that day-to-day experiences with discrimination that young pregnant women of color face, change over the course of pregnancy and postpartum, at times when visual cues of pregnancy (e.g., growing belly) and motherhood (e.g., infant in stroller) may be more or less prominent (Rosenthal et al., 2015).

Future work also might compare perceptions of Black and White women who are either pregnant or not to perceptions of Black and White men who are either expecting to become a parent or not, to study the role that the intersections of race and gender play in stereotyping. Given that we (Rosenthal & Lobel, 2011) have noted that some experiences of discrimination and stereotypes of Latina women are similar, intertwined, or parallel with those of Black women (e.g., Davis, 1981; Ghavami & Peplau, 2013), it would be valuable to expand this investigation to a Latina target that is pregnant or not as well as to targets of other racial/ethnic backgrounds. Also, given the importance of motherhood in stereotypes about Black women, future work may want to examine motherhood as an identity or status that intersects with other identities, such as race, to affect perceptions of women.

Effect sizes were generally small in this study. However, even these small differences in perceptions may be meaningful for women’s well-being through several mechanisms, including stereotype internalization, stereotype threat, and stereotyping leading to discrimination (Rosenthal & Overstreet, 2016). It is not known how perceptions based on race and pregnancy status affect the actual treatment of women. This remains an important issue that should be explored in future research. In both quantitative and qualitative studies, researchers might explore Black women’s awareness of stereotypes related to sexuality and motherhood, and how they feel those stereotypes affect how they are treated in various contexts. Some (e.g., Bowleg, 2008) have suggested that intersectional issues are best studied with qualitative methods, which can give Black women (and other groups of people) the ability to express how stereotypes affect them in ways that may not be captured with quantitative measures. The current findings offer a foundation for examining the connections of stereotypes about Black American women related to sexuality and motherhood to their sexual and reproductive health outcomes through multiple mechanisms; future research is needed to explore those connections. Experimental studies might examine whether the same stereotypes are found among health-care providers and other professionals who interact with women in institutional settings, and whether stereotypes have an influence on services Black women receive. It would be particularly valuable to examine whether perceptions of women affect decision-making and treatment by health-care providers.

The study sample was comprised of undergraduates from a public university in the Northeastern United States, who were on average about 20 years old, close in age but younger than the target Jasmine’s age (25). Given the characteristics of the study sample, it is possible that the results do not accurately estimate the extent of stereotyping that exists in other individuals, and the generalizability of findings should be tested in other samples. Furthermore, the largest racial/ethnic groups represented in the sample were White/European Americans and Asian Americans, with other groups each comprising less than 10% of the sample. Although we found that controlling for race of participant—using a dichotomous variable that compared all White participants to all participants of color—did not change the pattern of results nor moderate any of the effects, our ability to discern the impact of participant race/ethnicity was constrained. Also, controlling for socioeconomic status did not alter nor moderate results, but our self-report measure of socioeconomic status used categories whose interpretation by participants cannot be verified, and some categories were chosen only by a few participants, with the majority identifying as middle class. Thus, our ability to discern the impact of participants’ socioeconomic status was also constrained. Furthermore, because less than 7% of participants identified themselves as Black/African American, the current study was not able to address potential stereotype internalization and its implications, which is an important area for future research. Future research should attempt to replicate study findings in more diverse samples and using more sophisticated assessments of socioeconomic status.

We created and tested single items to assess participants’ stereotypes of Black women. Future work may benefit from using composite measures to assess these stereotypes. Although a strength of the current study is that we studied the application of stereotypes to perceptions or judgments about Black women, exactly how this translates to real-life perceptions, judgments, and treatment of Black women remains unclear. In this tightly controlled experimental context in which participants were asked to make assumptions about a target woman, participants may have been drawing on their understanding of social and economic realities (e.g., Black women being paid less than White women). However, there is evidence that stereotypes can affect judgments and interactions with others both consciously and unconsciously, regardless of the source of those stereotypes (Kunda & Spencer, 2003), with adverse consequences for members of stereotyped groups (e.g., Rosenthal & Overstreet, 2016; Shavers et al., 2012; Steele, 1997).

Possible Implications for Black Women’s Sexual and Reproductive Health

As mentioned, the current investigation provides evidence that unique stereotypes about Black women related to sexuality and motherhood influence people’s judgments of Black women in comparison to White women. More work is needed to explore the potential consequences of stereotypes for Black women’s sexual and reproductive health. It is important to explore the effects of stereotypes on health outcomes. There is evidence that stereotypes can be harmful to the well-being of members of stereotyped groups through at least three mechanisms: (1) stereotype internalization, (2) stereotype threat, and (3) stereotyping leading to discrimination (Rosenthal & Overstreet, 2016). Stereotype internalization is when someone endorses stereotypes about their own group; it has been associated with various indicators of poorer well-being for numerous groups (e.g., Davison, Schmalz, Young, & Birch, 2008; Gupta, Szymanski, & Leong, 2011). For example, a recent study found that in a sample of adult Black women, greater internalization of sexual stereotypes of Black Americans, even when based on items that were framed “positively” (e.g., “Black people have a unique quality of sexuality that most White people don’t have”), was associated with more risky sexual attitudes and behavior (Duvall et al., 2013).

Stereotype threat is when someone is worried or anxious about the possibility of confirming or being judged according to stereotypes about their group; it has been found to have a variety of adverse consequences, affecting well-being, performance in the stereotyped domain (e.g., Steele, 1997; Steele & Aronson, 1995), and health behaviors (e.g., Seacat & Mickelson, 2009). For example, a recent study found that stereotype threat, based on health-related stereotypes about Black Americans, was associated with delays in seeking medical care among a sample of Black American adults (Jones et al., 2013). We have also proposed that stereotype threat related to unique stereotypes about Black women’s sexuality and motherhood could lead Black women to experience greater distress throughout their lives and specifically during pregnancy, when some of these stereotypes may be especially salient (Rosenthal & Lobel, 2011). Ample research supports that distress increases risk for adverse birth outcomes, including preterm birth and low birth weight (see Lobel & Dunkel-Schetter, 2016, for a review).

Stereotyping of a group can also lead to conscious or unconscious discrimination toward members of that group (see, e.g., Dovidio & Gaertner, 2010), and experiences of discrimination are associated with many adverse mental and physical health outcomes for a wide range of groups (see, e.g., Pascoe & Smart Richman, 2009; Williams & Mohammed, 2009, for reviews). Evidence has been growing that women’s experiences with discrimination, both throughout their lifetime and during pregnancy, predict greater risk of adverse birth outcomes (e.g., Earnshaw et al., 2013; Giscombé & Lobel, 2005; Parker Dominguez, Dunkel-Schetter, Glynn, Hobel, & Sandman, 2008). Also, research has begun to accumulate which finds that experiences with discrimination are related to sexual risk, including among Black and Latina women (Rosenthal et al., 2014). Further, as discussed earlier, if health-care providers hold stereotypes consciously or unconsciously, this can affect their provision of care (e.g., Calabrese et al., 2014; Shavers et al., 2012). The finding that stereotypes of Black women affected participants’ perceptions of Black women illustrates that it is important to explore the role of these stereotypes in relation to existing disparities that Black women experience in sexual and reproductive (as well as other) health outcomes.

Practice Implications

Study findings support the continuing need to eliminate stereotypical images of Black women and replace them with more diverse, positive, complex, and dynamic images that reflect the reality of Black women’s lives (see, e.g., Mastro, 2015). In this way, these findings are relevant to activists, artists, educators, and those who generate and promote public media. It is also important that as consumers of media, we voice our concerns about stereotypical images that are promoted of Black women as well as other groups and hold the producers of these images accountable for them. It is important to critically question why these stereotypical images have persisted for centuries and what societal structures are connected to their persistence. Historically, these stereotypical images resulted from and justified the oppression of Black women in slavery, including the sexualized abuse and exploitation of Black women to the benefit of White male slave owners (e.g., Sublette & Sublette, 2015). To understand how we might change these images, we must question what sustains them today, and what ends they serve in our society.

These findings also have implications for practitioners as well as individuals training to be practitioners in any field who work and interact with Black women, particularly pregnant Black women, Black mothers, and Black women seeking sexual and reproductive care. Programs for trainees and established practitioners could use the current findings to increase these individuals’ awareness of how unique stereotypes about Black women, due to the intersection of race and gender (as well as pregnancy), may explicitly or implicitly affect their perceptions, judgments, and treatment of the Black women with whom they interact. By increasing this awareness and helping these individuals learn how to reduce the influence of stereotypes and respond to Black women in more socially just ways, the impact of these stereotypes may be reduced. Further, given current interest in ways to foster positive intergroup relations at diverse institutions, including universities and colleges across the United States (e.g., Sidanius, Levin, Van Laar, & Sears, 2010) where controversies are erupting over institutional oversight of diversity and discrimination (e.g., Hartocollis & Bidgood, 2015), the current findings could be used to foster constructive dialogue about stereotypes and the ways they affect intergroup relations in our increasingly diverse society.

Conclusions

Researchers across disciplines are more frequently adopting an intersectionality framework, confirming that identity arises from multiple intersecting factors, including one’s race and gender, and that the experiences that come along with those intersecting identities are connected to interlocking systems of oppression. The current investigation supports the notion that intersectionality influences the stereotypes that people possess about others, and it offers evidence that pregnancy may be an additional intersecting identity that influences stereotypes about women, particularly in conjunction with women’s race. In particular, our findings provide evidence that people hold unique stereotypes about Black women related to sexuality, motherhood, and socioeconomic status, which are relevant to pregnancy. We also uncovered some preliminary evidence for distinct stereotypes about pregnant Black women. Intersectionality may help us to better understand stereotypes and their implications in many realms of society. Current findings, along with past evidence of damaging effects of societal stereotypes, underscore the importance of diversifying images of Black women in media (e.g., Mastro, 2015), increasing awareness of how stereotypes affect perceptions of Black women, and working toward reducing those effects. Furthermore, it will be valuable for future work to explore the influence of stereotypes related to sexuality and motherhood on the persistent disparities in sexual and reproductive health outcomes that Black women face.

Appendix

These were the two images of the target Jasmine that varied her race: