An Explanation for the Gender Gap in Perceptions of Discrimination among African Americans

Abstract

Studies indicate that African American men report more personal experiences with discrimination than do African American women. According to the subordinate male target hypothesis, this gender difference reflects an underlying reality in which African American men are the primary targets of anti-Black discrimination. From the perspective of intersectionality theory, African American women and men experience racial discrimination differently; and therefore greater reports of discrimination among African American men might be a result of measurement bias that favors the experiences of African American men vis-à-vis African American women. To assess these perspectives, the authors analyze data from the 1995 Detroit Area Study and the 2001–2003 National Survey of American Life. The authors use multiple-group confirmatory factor analytic models with latent means and categorical outcomes to observe the degree to which gender bias in measurement accounts for disparities in perceived discrimination among African American women and men. The results show that gender bias in the measures most often used to assess unfair treatment in social surveys is responsible for the gender gap in certain kinds of perceived discrimination among African Americans. Measures of everyday discrimination are mostly gender balanced, but measurement bias is responsible for a large portion of the gap in perceptions of major life discrimination and the entire gap when major life discrimination is attributed to race. The results highlight the importance of intersectionality theory for assessing discrimination, and the authors argue that revisions in the measurement of perceived discrimination are required to better reflect the experiences of African American women.

Keywords

racism perceived discrimination gender intersectionality measurement

Social science research shows that perceptions of discrimination remain high in the United States, particularly among racial and ethnic minorities (Forman, Williams, and Jackson 1997; Kessler, Mickelson, and Williams 1999; Pérez, Fortuna, and Alegria 2008). However, relative to the study of prejudice, there is a near absence of sociological research on perceived discrimination (Bobo and Fox 2003). Even so, recent studies of perceived discrimination among African Americans present us with an underexplored finding. Although African American women and men report comparable levels of “anti-Black” group discrimination in American society (Kane and Kyyro 2001; Sigelman and Welch 1991), African American men report more personal experience with discrimination than African American women (Broman, Mavaddat, and Hsu 2000; Forman et al. 1997; Sellers and Shelton 2003; Sigelman and Welch 1991). This gender gap¹ exists when respondents are asked about their experiences with racial discrimination (e.g., Sellers and Shelton 2003) and when they are asked about discrimination more generally (e.g., Forman et al. 1997). However, because African American women are both Black and female, and therefore hold two stigmatized social statuses, many would expect that African American women would experience and report more discrimination than African American men.

This gender gap is often interpreted using a version of the subordinate male target hypothesis, which attributes the gap to an underlying reality in which African American men are more often the targets of discrimination than are African American women (e.g., Anderson 2008; Ferguson 2001; Gary 1995; Navarrete et al. 2010; Pieterse and Carter 2007; Sidanius and Veniegas 2000). Intersectionality theory offers an alternative explanation, which centralizes the ways in which systems of inequality reinforce and mutually shape one another at all levels of society (Cho, Crenshaw, and McCall 2013; Collins 2000; Crenshaw 1989; King 1988; May 2015; McCall 2005; Weber 2010; Zinn and Dill 1996). Those who favor an intersectional approach are therefore inclined to consider how gender might shape the way African American women and men experience and perceive mistreatment (Cho et al. 2013; Collins 2000; Crenshaw 1989; King 1988) and the way these experiences and perceptions are measured in social surveys (Harnois 2013; Harnois and Ifatunji 2011). As a result, gender bias in the measurement of perceived discrimination in social surveys may be responsible for the observed gender differences in perceived discrimination among African Americans.

In the present study, we investigate the relative role of these two perspectives in explaining the gender gap in perceived discrimination among African Americans. After reviewing the conceptualization and measurement of perceived discrimination in social surveys, we review an array of studies, all of which show that African American men report more experiences with discrimination than African American women. We then consider two competing explanations for this finding. The subordinate male target hypothesis argues that the gender gap in perceived discrimination reflects an underlying reality in which African American men are the primary targets of racial discrimination. If this view is correct, accounting for gender-related measurement bias will leave the gender gap largely intact. Intersectionality theory suggests that the measures social scientists use to study perceived discrimination in social surveys may be gendered in such a way as to inflate or wholly create the gender gap by more accurately assessing the experiences of African American men. To the extent that this view is correct, the gender gap in perceived discrimination will be substantially reduced after accounting for gender-related measurement bias. We draw on data from the 1995 Detroit Area Study (DAS) and the 2001–2003 National Survey of American Life (NSAL) and use multiple-group confirmatory factor analysis with latent means (Bollen 1989) and categorical outcomes (Sass, Schmitt, and Marsh 2014) to assess gendered measurement bias (or differences in model fit before and after items are allowed to load differently for women and men) in two widely used measures of perceived discrimination (i.e., “everyday” and “major life” discrimination; Harnois and Ifatunji 2011; Williams, Yu, and Jackson 1997). We then examine the degree to which this measurement bias accounts for the gender gap in perceived discrimination among African Americans, by observing reductions in the gender gap that correspond with improvements in model fit associated with allowing items to load differently for women and men.

Background on Perceived Discrimination

Scholars have offered working definitions of racial discrimination (e.g., Jones 2000; Krieger 1999; Williams et al. 1997). According to the National Research Council of the National Academies, racial discrimination occurs when there is “differential treatment on the basis of race that disadvantages a racial group” or “treatment on the basis of inadequately justified factors other than race that disadvantages a racial group” (Blank, Dabady, and Citro 2004:39). That is, racial discrimination is present when there is either disparate treatment or disparate impact along particular racial lines. Given this definition of racial discrimination, we define two types of perceived discrimination. First, we define the general perception of discrimination as the perception or view that one has been treated unfairly personally, or the observation of unjust disparate impact for any particular group in a given context. Second, we define the perception of racial discrimination as the perception of unfair treatment in one’s personal experiences with the attribution of such treatment to one’s (ascribed) racial (or ethnic) identity, or the observation of disparate impact along particular racial or ethnic lines. The distinction between our terms and the definitions cited is that our terms focus on perception. Our first definition concerns discrimination in general (without status attribution), and the second focuses on discrimination that individuals attribute to racial or ethnic factors.

Although there is a wide array of measures for assessing discrimination, including a growing literature on microaggressions (Pierce, Pierce-Gonzalez, and Willis 1978; Sue et al. 2007), in this study we assess two widely used measures of perceived discrimination, each using multiple indicators: everyday and major life discrimination. Everyday discrimination is generally thought to assess “the integration of racism into everyday situations through practices that activate underlying power relations” (Essed 1991:50). Different from such “daily hassles,” major life discrimination represents an encounter with discrimination at a time and place when it might restrict social mobility (e.g., Kessler et al. 1999). Although some conceptualize these major life experiences as isolated and event specific traumatic stressors (Williams et al. 1997), given the structural nature of race in the United States (Bonilla-Silva 1997), we view them as cumulative (Reskin 2012). That is, we source anti-Black discrimination in the social structure of society, which then systematically stigmatizes Black people (albeit in a gendered manner) across time- and location-separated events. These events then share a common cause such that one event is correlated with the likelihood of exposure to another event. Therefore, experiencing unfair treatment by teachers in school and being treated unfairly by the police have a shared genesis in white supremacy.

The Gender Gap in Perceived Discrimination

African American women and men are equally likely to believe that racial inequality in the United States is due to anti-Black discrimination (Forman and Ifatunji 2006; Kane and Kyyro 2001; Sigelman and Welch 1991; Weitzer and Tuch 1999). According to a study that used data from the General Social Survey, 65 percent of African American men and 64 percent of African American women reported that racial inequality is the result of discrimination against African Americans (Kane and Kyyro 2001). Moreover, African American women and men do not differ on questions concerning discriminatory treatment of “Black people” by the criminal justice system and unequal neighborhood protection in African American neighborhoods (Weitzer and Tuch 1999). Findings such as these, on anti-Black group discrimination, persist even after controlling for other social and demographic background factors (Forman and Ifatunji 2006; Sigelman and Welch 1991).

Even though African American women and men are equally likely to believe that discrimination remains a problem for African Americans, African American men report more personal experiences with discrimination than African American women (Barnes et al. 2004; Broman et al. 2000; Brown 2001; Din-Dzietham et al. 2004; Forman et al. 1997; Herring et al. 1998; Ifatunji and Forman 2006; Sigelman and Welch 1991; Weitzer and Tuch 1999; Welch et al. 2001; Williams and Chung 1996). This pattern holds when African American respondents are asked about their personal experiences with racial discrimination in particular (i.e., without attribution; Broman et al. 2000; Brown 2001; Din-Dzietham et al. 2004; Hausmann et al. 2008; Weitzer and Tuch 1999; Welch et al. 2001; Williams and Chung 1996) and when asked about personal discrimination more generally (Barnes et al. 2004; Forman et al. 1997; Herring et al. 1998; Ifatunji and Forman 2006; Sigelman and Welch 1991). Adjusting these estimates for relevant social background factors does not account for this pattern (Forman et al. 1997; Herring et al. 1998; Sigelman and Welch 1991).

With respect to the measures of discrimination considered in this study, African American men report more perceived discrimination than do African American women on a range of indicators associated with major life discrimination. For instance, African American men are six times as likely as African American women to report racial discrimination by the police, even after controlling for social background characteristics (Broman et al. 2000; Weitzer and Tuch 1999). After considering the effects of education, age, and family income, African American men are more than twice as likely to report job discrimination (Broman 1996). In addition to their encounters with police and in seeking employment, African American men report more racial discrimination than African American women in attaining fair wages, getting a quality education (Welch et al. 2001), and engaging with the health care system (Hausmann et al. 2008). These findings also persist after adjusting for the effects of racial composition in several social contexts (Welch et al. 2001).

It appears that gender differences in everyday discrimination are not as stark as they are for major life discrimination. This is likely associated with the fact that these measures were inspired by the book Everyday Racism, which was a qualitative inquiry into the ways in which Black women in Canada experienced racial discrimination in daily life (Essed 1991). Nevertheless, a study focusing on Detroit found scores that are about 9 percent greater for African American men than for African American women (Forman et al. 1997). Another study fielded in Miami put the gap at about 16 percent (Taylor and Turner 2002). However, results from factor analytic methods show that everyday discrimination is composed of two dimensions: “unfair treatment” and “personal rejection” (Barnes et al. 2004; Stucky et al. 2011). When the larger scale is subdivided in this way, African American women and men report similar levels of unfair treatment, but African American men report more personal rejection (Barnes et al. 2004).

Explanations for the Gender Gap

Subordinate Male Target Hypothesis

Much of the research that documents the gender gap in perceived discrimination interprets this gap as reflecting the underlying experiences of African Americans; that is, men are the primary targets of anti-Black discrimination (e.g., Ferguson 2001; Gary 1995; Navarrete et al. 2010; Pieterse and Carter 2007; Veenstra 2013). Social psychologists argue that prejudice and discrimination are driven largely by the perception of group threat, as opposed to intergroup animosity or negative affect (Blumer 1958; Bobo and Tuan 2006). By extension, members of the dominant group are more likely to target men in the subordinate group, because men are viewed as more threatening than women (Sidanius and Pratto 1999). From this perspective, the gender gap in perceived discrimination stems from the fact that whites perceive African American men as posing a greater threat to their group position than African American women (Navarrete et al. 2010; Sidanius and Veniegas 2000). The subordinate male target hypothesis therefore expects the disproportionate targeting of African American men vis-à-vis women and is unsurprised by the greater rates of perceived discrimination found among African American men in social surveys. According to this perspective, because African American men are targeted at greater rates than African American women, no amount of measurement bias in the scales used to assess perceived discrimination can account for the underlying reality that African American men experience more discrimination than African American women. For our analysis of the gender gap in perceived discrimination, this perspective yields our first hypothesis:

Hypothesis 1: Group mean differences between African American women and men remain intact after accounting for gender-related measurement bias in scales that assess perceived discrimination.

The Intersection of Race and Gender

Intersectionality theory is a perspective that centralizes the intersections of multiple inequalities, as they occur within and across multiple levels of society (Cole 2009; Collins 2000; Crenshaw 1989; King 1988; May 2015; Weber 2010). Early articulations of intersectionality argued that because of their experiences with both racism and sexism, African American women face “double disadvantage” or “double jeopardy” because they are both Black and female (e.g., Beale 1970). In the decades since, intersectional scholars have revised these “additive models,” arguing that it is not possible to disentangle and then add up the separate contributions of race and gender, as they are mutually constitutive; that is, notions of race are gendered and notions of gender are shaped by race (Baca Zinn and Thornton Dill 1996; Bowleg 2008; Crenshaw 1989; Hancock 2007; King 1988). Therefore, although African American women and men may share experiences and perspectives as a result of being in the same racial group, intersectionality asserts that the ways in which they experience and perceive unfair and/or mistreatment are likely to be different (Bowleg 2008; Collins 2000; Crenshaw 1989; Harnois and Ifatunji 2011; Wingfield 2007).

According to this perspective, the “racial experience” of African Americans might be quantitatively similar (in terms of prevalence) and yet qualitatively different (in terms of the nature of any given experience) for women and men. In fact, several studies show that African American women and men are stereotyped in qualitatively different ways (Collins 2004; Dill 1988; Shih 2002; Slatton and Spates 2014; Timberlake and Estes 2007) and are often subjected to different types of mistreatment. For instance, Patricia Hill Collins (2000, 2004) has argued that the dominant cultural stereotypes of African American women and men are based not only on race but also on gender, class, and sexuality. These “controlling images” result in portrayals of African American men in the media that range from the lower class, “authentically” Black, hypersexual “thug” or “pimp” to the desexualized, “white-washed” (i.e., inauthentic), upper-middle-class “Black buddy” or “sidekick.” In short, many equate male “Blackness” with poverty, criminality, violence, and hypersexuality. Racial stereotypes for African American women are somewhat different and include the desexualized, lower class mammy, the lower class “welfare queen,” the hypersexualized (and frequently animalistic) “jezebel,” and the “white-washed,” upper-middle-class “Black lady” (e.g., the television character Claire Huxtable).

Given the gendered nature of racial stereotypes, it comes as little surprise that the measures social scientists use to study perceived discrimination are also gendered (Harnois 2013; Harnois and Ifatunji 2011). However, these measures are not evenly gendered. At least one study showed that the measures often used to assess the perception of discrimination in social surveys are gendered in ways that favor the experiences of African American men (Harnois and Ifatunji 2011). The present study builds upon this by analyzing the extent to which this measurement bias explains the gender gap in perceived discrimination. Given the perspective offered by intersectionality theory, we develop our second hypothesis:

Hypothesis 2: Accounting for gender-related measurement bias in scales that assess perceived discrimination results in meaningful reductions in mean differences between African American women and men.

To test these hypotheses, we analyze data from two social surveys that include measures of everyday and major life discrimination. Our analysis begins by assessing gendered measurement bias in these measures. We then examine the degree to which this measurement bias accounts for the gender gap in perceived discrimination among African Americans.

Data and Methods

The data for this study come from two social surveys: the DAS (Jackson and Williams 1995) and the NSAL (Jackson et al. 2004). The DAS is representative of the Detroit metropolitan area,² and the NSAL contains a nationally representative sample of African Americans.³ The DAS was the first study to include measures of major life and everyday discrimination and therefore provides an important “baseline” sample (Williams et al. 1997). Each assessed perceived discrimination with multiple questions, and both used a similar set of questions. The analytic samples for the DAS are 180 men and 393 women for both everyday and major life discrimination and 152 men and 347 women for the assessment of major life discrimination attributed to race. The analytic sample for the NSAL includes 1,120 men and 2,073 women (for all three operationalizations of perceived discrimination).⁴ All analytic samples only include those respondents who self-identified as Black or African American and who answered all of the questions regarding perceived discrimination.

In this study, we use retrospective accounts of respondents’ perceptions to assess their experiences with both everyday and major life discrimination (Forman et al. 1997; Harnois and Ifatunji 2011; Ifatunji and Forman 2006; Kessler et al. 1999; Taylor and Turner 2002; Williams et al. 1997). Although retrospective accounts are less than ideal (Hardt and Rutter 2004; Maughan and Rutter 1997), as it is often difficult to assess the intent or impetus of discrimination (Blank et al. 2004; Phinney, Madden, and Santos 1998), there is an established body of research that connects retrospective reports of the perception of discrimination to various life outcomes, including mental and physical health (Amaro, Russo, and Johnson 1987; Brown 2001; Kessler et al. 1999; Krieger and Sidney 1996; Krieger et al. 2005; Lewis et al. 2009; Lewis, Cogburn, and Williams 2015; Pager and Shepherd 2008; Paradies 2006; Pascoe and Richman 2009; Schulz et al. 2006; Williams et al. 1997). Therefore, although retrospective reports of perceived discrimination are not ideal, they remain an important tool for estimating trends in the national population.

The three measures of discrimination examined in this study are increasingly becoming the standard for measuring perceived discrimination in social survey research (Forman et al. 1997; Harnois and Ifatunji 2011; Ifatunji and Forman 2006; Kessler et al. 1999; Taylor and Turner 2002; Williams et al. 1997). These measures are everyday, major life, and “major life attributed to race.” Following others, we parse everyday discrimination into two dimensions but retain it as a single kind of discrimination (Barnes et al. 2004; Stucky et al. 2011).

Similar to previous work, we theorize two dimensions of everyday discrimination: implicit and explicit (Barnes et al. 2004). Implicit discrimination reflects the daily hassles people experience in more subtle ways. We create this latent factor using five variables: “. . . treated with less courtesy than other people,” “. . . you are treated with less respect than other people,” “. . . receive poor service compared with other people at restaurants or stores,” “. . . people act as if they think you are not smart,” and “. . . people act as if they’re better than you are.” Explicit discrimination refers to the less than subtle or slightly more overt ways in which people encounter daily hassles. We create this latent factor using four variables: “. . . people act as if they are afraid of you,” “. . . people act as if they think you are dishonest,” “. . . you are called names or insulted,” and “. . . you are threatened or harassed.” We then correlate these two latent factors to compose a single measurement model. There are five response options for all of the above questions: “less than once a year,” “a few times a year,” “a few times a month,” “at least once a week,” and “almost every day.” The more frequent the experience, the greater the level of everyday discrimination. The DAS and NSAL measure this construct in about the same way, except that the NSAL uses one additional question, which we include in the explicit dimension: “You are followed around in stores” (see Table 1 to compare questions in the DAS and NSAL). The single-group confirmatory factor analytic model fits the data well in both the DAS (root mean square error of approximation [RMSEA] = .052, comparative fit index [CFI] = .994) and the NSAL (RMSEA = .062, CFI = .992).⁵

Table 1.

Means and Standard Deviations for Items Measuring Everyday Discrimination, by Gender.

	DAS (1995)^a			NSAL (2001–2003)^b
Question	Male	Female	t	Male	Female	t
All items^c,d	2.170	1.990	2.720**	2.320	2.140	5.750***
	(.776)	(.727)		(.945)	(.851)
Implicit items^e	2.390	2.270	1.630	2.600	2.470	3.370***
	(.887)	(.877)		(1.120)	(1.050)
You are treated with less courtesy than other people.	2.380	2.210	1.810	2.600	2.430	3.720***
	(1.040)	(1.030)		(1.320)	(1.210)
You are treated with less respect than other people.	2.260	2.100	1.670	2.470	2.320	3.290***
	(.992)	(1.020)		(1.270)	(1.190)
You receive poorer service than other people at restaurants or stores.	2.390(1.060)	2.210(1.010)	1.920	2.360(1.190)	2.250(1.140)	2.520*
People act as if they think you are not smart.	2.360	2.270	.760	2.590	2.490	1.820
	(1.190)	(1.280)		(1.500)	(1.430)
People act as if they’re better than you are.	2.590	2.530	.530	2.980	2.840	2.390*
	(1.270)	(1.260)		(1.640)	(1.610)
Explicit items^f	1.890	1.650	3.740***	2.050	1.810	7.660***
	(.789)	(.712)		(.921)	(.821)
People act as if they are afraid of you.	2.300	1.850	4.300***	2.380	1.830	11.670***
	(1.250)	(1.100)		(1.460)	(1.140)
People act as if they think you are dishonest.	1.920	1.620	3.360***	2.220	1.850	8.230***
	(1.090)	(.910)		(1.350)	(1.120)
You are called names or insulted.	1.760	1.630	1.420	1.890	1.810	1.810
	(1.000)	(.960)		(1.170)	(1.150)
You are threatened or harassed.	1.600	1.470	1.660	1.620	1.540	2.500*
	(.850)	(.875)		(.865)	(.852)
You are followed in stores.	—	—	—	2.150	2.000	3.110**
	—	—		(1.330)	(1.220)

Note: Statistical tests for t scores are two tailed (difference > 0). DAS = Detroit Area Study; NSAL = National Survey of American Life.

Results in these three columns are from the DAS (1995) (n = 180 men, n = 393 women).

Results in these three columns are from the NSAL (2001–2003) (n = 1,120 men, n = 2,073 women).

A multiplicative index of all nine items concerning “everyday discrimination.”

The preamble for these questions was “In your day-to-day life how often have any of the following things happened to you? First . . . .”

A multiplicative index of the five items that constitute the implicit dimension of the everyday discrimination index.

A multiplicative index of the four items that constitute the explicit dimension of the everyday discrimination index.

p < .05. **p < .01. ***p < .001.

Major life discrimination refers to encountering discrimination that could potentially result in blocks to social mobility. We assess this form of discrimination with six questions in the DAS and nine questions in the NSAL (see Table 2 to compare questions in the DAS and NSAL). In the DAS, we used the questions, “Do you think you have ever been unfairly fired or denied a promotion?” “For unfair reasons, do you think you have ever not been hired for a job?” “Have you been unfairly stopped . . . or abused by the police?” “Do you think you have ever been unfairly discouraged by a teacher or advisor from continuing your education?” “Do you think you have ever been unfairly prevented from moving into a neighborhood?” and “Do you think that neighbors have made life difficult for you and your family?” Respondents answered each of these questions “yes” or “no.” The more experiences a respondent has had, the greater his or her experiences with major life discrimination. The single-group confirmatory factor analytic model fits the data in both the DAS (RMSEA = .019, CFI = .993) and the NSAL (RMSEA = .024, CFI = .983).⁶

Table 2.

Means and Standard Deviations for Items Measuring Major Life Discrimination (Attributed to Race), by Gender.

	DAS (1995)^a,b			NSAL (2001–2003)^c
Question^d	Male	Female	t	Male	Female	t
No specific attribution (general)^e	.325	.212	5.480***	.196	.121	11.490***
	(.243)	(.222)		(.203)	(.161)
Do you think you have ever been unfairly fired or denied a promotion?	.483(.501)	.318(.466)	3.850***	—	—	—
For unfair reasons, do you think you have ever not been hired for a job?	.361(.482)	.282(.451)	1.900	.263(.441)	.180(.385)	5.520***
Do you think you have ever been unfairly stopped, searched, questioned, physically threatened or abused by the police?	.606(.490)	.198(.399)	10.520***	.442(.497)	.118(.322)	22.280***
Do you think you have ever been unfairly discouraged by a teacher or advisor from continuing your education?	.211(.409)	.142(.350)	2.060*	.103(.304)	.092(.289)	1.010
Do you think you have ever been unfairly prevented from moving into a neighborhood because the landlord or a realtor refused to sell or rent you a house or apartment?	.133(.341)	.160(.367)	−.830	.079(.271)	.075(.264)	.430
Have you ever moved into a neighborhood where neighbors made life difficult for you or your family?	.156(.363)	.170(.377)	−.450	.066(.249)	.077(.267)	−1.150
At any time in your life, have you ever been unfairly fired?	—	—	—	.280(.449)	.194(.395)	5.610***
Have you ever been unfairly denied a promotion?	—	—	—	.247	.152	6.620***
	—	—		(.432)	(.360)
Have you ever been unfairly denied a bank loan?	—	—	—	.122	.094	2.540*
	—	—		(.328)	(.291)
Have you ever received service from someone such as a plumber or car mechanic that was worse than what other people get?	—	—	—	.161	.105	4.593***
	—	—		(.367)	(.306)
Attributed to race or ethnicity^f	.181	.097	5.010***	.121	.056	12.830***
Attributed to race or ethnicity^f	(.210)	(.154)		(.170)	(.114)
Do you think you have ever been unfairly fired or denied a promotion?	.257	.141	3.140**	—	—	—
	(.438)	(.349)		—	—
For unfair reasons, do you think you have ever not been hired for a job?	.191	.161	.800	.163	.096	5.580***
	(.394)	(.368)		(.370)	(.295)
Do you think you have ever been unfairly stopped, searched, questioned, physically threatened or abused by the police?	.408	.095	8.790	.342	.078	20.060
	(.493)	(.294)		(.475)	(.268)
Do you think you have ever been unfairly discouraged by a teacher or advisor from continuing your education?	.086	.061	1.02	.057	.044	1.600
	(.281)	(.239)		(.232)	(.206)
Do you think you have ever been unfairly prevented from moving into a neighborhood because the landlord or a realtor refused to sell or rent you a house or apartment?	.086	.084	.070	.067	.055	1.370
	(.281)	(.277)		(.250)	(.228)
Have you ever moved into a neighborhood where neighbors made life difficult for you or your family?	.059	.037	1.09	.039	.028	1.650
	(.237)	(.190)		(.194)	(.166)
At any time in your life, have you ever been unfairly fired?	—	—	—	.129	.074	5.130***
	—	—		(.336)	(.262)
Have you ever been unfairly denied a promotion?	—	—	—	.163	.072	8.040***
Have you ever been unfairly denied a promotion?	—	—		(.369)	(.259)
Have you ever been unfairly denied a bank loan?	—	—	—	.072	.050	2.560*
Have you ever been unfairly denied a bank loan?	—	—		(.259)	(.218)
Have you ever received service from someone such as a plumber or car mechanic that was worse than what other people get?	—	—	—	.062	.021	5.920***
	—	—		(.241)	(.144)

Note: Statistical tests for t scores are two tailed (difference > 0). DAS = Detroit Area Study; NSAL = National Survey of American Life.

Results in these three columns are from the DAS (1995).

There are 180 men and 393 women in the rows for “no specific attribution (general)” and 152 men and 347 women in the rows for “attributed to race or ethnicity.”

Results in these three columns are from the NSAL (2001–2003) (n = 1,120 men, 2,073 women).

The preamble for these questions was “In the following questions, we are interested in the way other people have treated you or your beliefs about how other people have treated you. Can you tell me if any of the following has ever happened to you . . . .”

A multiplicative index of items concerning major life discrimination. A dash indicates that the question was not asked in the survey.

A multiplicative index of items concerning major life discrimination, attributed to race. A dash indicates that the question was not asked in the survey.

p < .05. **p < .01. ***p < .001.

To capture major life discrimination attributed to race, we developed a second operationalization of major life discrimination from respondents who attributed their treatment to race or ethnicity (following Brown 2001). That is, after answering each of the above questions (for major life discrimination), respondents were then asked a follow-up question: “What was the main reason [for this mistreatment]?” For this second operationalization of major life discrimination (i.e., “major life discrimination attributed to race”), if a respondent reported no experience with a particular instance of discrimination or if he or she did not attribute the cause of that experience to race or ethnicity,⁷ we coded that respondent as not having experienced major life discrimination attributed to race. If the respondent reported experiencing discrimination and attributed it to race or ethnicity, we coded him or her as having experienced major life discrimination attributed to race. The single-group confirmatory factor analytic model fits the data in both the DAS (RMSEA = .000, CFI = 1.000) and the NSAL (RMSEA = .024, CFI = .980).⁸

An implicit assumption in nearly all of the studies that use these scales is that the indicators relate to the latent variables in the same way for African American women and men (Harnois and Ifatunji 2011; Stucky et al. 2011). That is, an increase in the underlying latent variable of perceived discrimination would lead to identical changes in any given item for African American women and men. If this assumption is not true, any comparison of these measures is flawed, because the items reflect the latent factor differently for African American women and men. Therefore, we use multiple-group confirmatory factor analysis with latent means (Bollen 1989) and categorical outcomes (Sass et al. 2014) to test for measurement invariance (Cheung and Rensvold 2002) in perceived discrimination between African American women and men (Harnois and Ifatunji 2011). If African American women and men experience discrimination differently, items that are designed to assess discrimination will indicate the underlying factor differently for women and men (Molenaar and Nesselroade 2012). For example, the factor loading for discrimination by the police might be larger for African American men if this more accurately reflects the way in which African American men experience major life discrimination. We determine this differential factor loading by allowing women and men to have different factor loadings and then assessing whether this allowance results in improved model fit. Our index variables were selected because they loaded most similarly between genders. To account for the dichotomous and ordinal nature of the response options, we estimate our models using threshold cut points instead of estimating intercepts (Bollen 1989; Muthén and Muthén 2012; Sass et al. 2014).⁹

After assessing model fit, we then test for the extent to which gender differences in the underlying factor means (or the means for the unobserved latent variables) are due to variance between African American women and men in the association between the observed items and the unobserved latent factors. For example, let us assume that mistreatment by the police better reflects the experience of major life discrimination for African American men (i.e., allowing different factor loadings results in improved model fit). In a standard regression model, women and men are forced to have the same “factor loading” or β coefficient for this item. If we relax this assumption of sameness and allow women and men to have different factor loadings for this item, we are able to observe any corresponding reductions in the underlying factor mean difference between women and men, because women and men would be allowed to have their own covariance structures (i.e., adjusting for measurement error), thus allowing a more accurate assessment of the underlying means and the associated gender gap (Molenaar and Nesselroade 2012).

This measurement strategy has several advantages relative to the use of other strategies to assess measurement, including Cronbach’s α scores and summed or averaged indexes. First, it provides for a more accurate assessment of construct validity, as Cronbach’s α scores do not account for the dichotomous nature of the response options. This limitation might lead to underestimation of the shared variance of items used to measure a given construct. Second, unlike summed or averaged index scores, it allows a targeted assessment of the relative contribution that each item makes to measurement error between groups. Third, it allows an assessment of whether observed differences in validity are statistically different between groups. Finally, it allows the investigator to observe mean differences between groups (i.e., differences in the unobserved latent factor mean) before and after accounting for specific between-group errors in measurement.

Results

Tables 1 and 2 replicate findings from previous studies, which also showed that African American men report more personal experiences with discrimination than do African American women (Broman et al. 2000; Forman et al. 1997; Ifatunji and Forman 2006; Sellers and Shelton 2003; Sigelman and Welch 1991). In all cases in which gender differences are statistically significant, men report more discrimination than women. This is true for everyday, major life, and major life discrimination attributed to race. According to t scores, the largest differences between women and men for everyday discrimination are within the explicit dimension. The two items with the biggest gender differences are “people act as if they are afraid of you” and “people act as if you are dishonest.” The largest differences for items assessing major life discrimination and major life discrimination attributed to race involved experiences with the police and in the labor market.

Do Measures of Perceived Discrimination Contain Gender Bias?

Tables 3a and 3b present findings from a series of multiple-group confirmatory factor analytic models of everyday discrimination, for women and men, using the DAS and NSAL. Allowing women and men to have different factor loadings does not improve overall model fit in the DAS but substantially improves model fit in the NSAL (Δχ² = 168.18, p = .000). Findings from the DAS show that women and men have different factor loadings for the question assessing “poor service” (Δχ² = 8.44, p = .004). Findings from the NSAL show that women and men have different factor loadings for three types of mistreatment: people acting as if they are afraid of you (Δχ² = 114.87, p =.000), being called names or insulted (Δχ² = 132.55, p = .000), and being threatened or harassed (Δχ² = 97.43, p = .000). Because chi-square tests rely heavily on sample size, the difference in significance levels across samples may stem in part from different sample sizes.

Table 3a.

Multiple-group Confirmatory Factor Analysis for Everyday Discrimination, DAS (1995).^a

Variable	1	2	3	4	5	6	7	8	9
Implicit^b
People act as if they are better than you are? Treated with less courtesy than other people?	—	—	—	—	—	—	—	—	—
	.858***(.103)	.854***(.108)	.850***(.101)	.860***(.102)	.874***(.107)	.859***(.103)	.860***(.103)	.861***(.103)	.876***(.105)
	—	.937***	—	—	—	—	—	—	—
	—	(.107)	—	—	—	—	—	—	—
Treated with less respect than other people?	1.200***	1.210***	1.140***	1.210***	1.240***	1.200***	1.200***	1.210***	1.240***
	(.140)	(.139)	(.141)	(.140)	(.148)	(.139)	(.139)	(.140)	(.145)
	—	—	1.040***	—	—	—	—	—	—
	—	—	(.123)	—	—	—	—	—	—
Poorer service . . . at restaurants or stores?	1.050***	1.050***	1.040***	1.150***	1.070***	1.050***	1.050***	1.050***	1.170***
	(.115)	(.116)	(.114)	(.136)	(.120)	(.115)	(.115)	(.115)	(.139)
	—	—	—	.912***	—	—	—	—	.932***
	—	—	—	(.109)	—	—	—	—	(.111)
People act as if they think you are not smart?	1.370***	1.370***	1.370***	1.360***	1.170***	1.370***	1.360***	1.360***	1.190***
	(.175)	(.176)	(.176)	(.170)	(.151)	(.174)	(.174)	(.173)	(.153)
	—	—	—	—	1.190***	—	—	—	1.170***
	—	—	—	—	(.141)	—	—	—	(.137)
Explicit^b
People act as if they think you are dishonest?	—	—	—	—	—	—	—	—	—
People act as if they are afraid of you?	.639***	.639***	.640***	.639***	.633***	.688***	.640***	.638***	.671***
	(.090)	(.090)	(.090)	(.090)	(.090)	(.103)	(.090)	(.090)	(.106)
	—	—	—	—	—	.576	—	—	.624***
	—	—	—	—	—	(.090)	—	—	(.091)
You are called names or insulted?	.562***	.562***	.563***	.562***	.556***	.568***	.509***	.562***	.509***
	(.099)	(.099)	(.099)	(.099)	(.099)	(.097)	(.101)	(.099)	(.110)
	—	—	—	—	—	—	.512***	—	.547***
	—	—	—	—	—	—	(.079)	—	(.084)
You are threatened or harassed?	.435***	.435***	.436***	.434***	.429***	.439***	.435***	.389***	.386***
	(.070)	(.070)	(.070)	(.070)	(.069)	(.068)	(.070)	(.075)	(.080)
	—	—	—	—	—	—	—	.421***	.436***
	—	—	—	—	—	—	—	(.063)	(.063)
Implicit
Factor mean disparity^c	−.204	−.201	−.177	−.194	−.173	−.203	−.203	−.203	−.157
SD	(.113)	(.111)	(.115)	(.115)	(.113)	(.113)	(.113)	(.113)	(.115)
t score	−1.800	−1.810	−1.550	−1.690	−1.540	−1.800	−1.800	−1.800	−1.370
Explicit
Factor mean disparity^d	−.862	−.862	−.860	−.863	−.869	−.818	−.838	−.721	−.644
SD	(.239)	(.239)	(.239)	(.240)	(.242)	(.238)	(.223)	(.217)	(.213)
t score	−3.600***	−3.600***	−3.600***	−3.600***	−3.590***	−3.440***	−3.760***	−3.320***	−3.020**
χ²	127.340	127.530	128.450	122.430	133.990	127.750	130.810	132.810	139.520
df	76	75	75	75	75	75	75	75	71
RMSEA	.049	.049	.050	.047	.052	.050	.051	.052	.058
CFI	.992	.992	.992	.993	.991	.992	.992	.991	.990

Note: This table reflects the following order on the releasing constraints between men and women: (1) factor means, factor variances, correlations between observed errors, and errors for observed terms are allowed to vary; (2) unconstrain the factor loading for “treated with less courtesy”; (3) unconstrain factor loading for “treated with less respect”; (4) unconstrain factor loading for “poor service”; (5) unconstrain factor loading for “think you are not smart”; (6) unconstrain factor loading for “afraid of you”; (7) unconstrain factor loading for “called names or insulted”; (8) unconstrain factor loading for “threatened or harassed”; and (9) unconstrain factor loadings for “poor service,” “not smart,” “afraid of you,” “called names or insulted,” and “threatened or harassed.” Statistical tests are two tailed (difference > 0). CFI = comparative fit index; DAS = Detroit Area Study; NSAL = National Survey of American Life; RMSEA = root mean square error of approximation.

Results in this table are from the DAS (1995) (n = 180 men, n = 393 women).

For all comparisons, the top row presents the factor loading for men, and the bottom row presents the factor loading for women.

This is the disparity for the latent factor mean, between men and women, for everyday implicit discrimination.

This is the disparity for the latent factor mean, between men and women, for everyday explicit discrimination.

p < .01. ***p < .001.

Table 3b.

Multiple-group Confirmatory Factor Analysis for Everyday Discrimination, NSAL (2001–2003).^a

Variable	1	2	3	4	5	6	7	8	9	10
Implicit^b
People act as if they are better than you are?	—	—	—	—	—	—	—	—	—	—
Treated with less courtesy than other people?	.703***	.750***	.705***	.703***	.704***	.703***	.704***	.704***	.703***	.705***
	(.036)	(.042)	(.036)	(.036)	(.037)	(.036)	(.036)	(.036)	(.036)	(.036)
	—	.702***	—	—	—	—	—	—	—	—
	—	(.036)	—	—	—	—	—	—	—	—
Treated with less respect than other people?	.797***	.801***	.822***	.797***	.799***	.797***	.798***	.797***	.797***	.799***
	(.039)	(.039)	(.044)	(.039)	(.040)	(.039)	(.039)	(.039)	(.039)	(.039)
	—	—	.820***⁷	—	—	—	—	—	—	—
	—	—	(.041)	—	—	—	—	—	—	—
Poorer service . . . at restaurants or stores?	.841***	.846***	.844***	.841***	.843***	.841***	.842***	.841***	.841***	.843***
	(.043)	(.043)	(.043)	(.046)	(.043)	(.043)	(.043)	(.043)	(.043)	(.043)
	—	—	—	.840***	—	—	—	—	—	—
	—	—	—	(.044)	—	—	—	—	—	—
People act as if they think you are not smart?	1.110***	1.100***	1.100***	1.110***	1.090***	1.110***	1.110***	1.110***	1.110***	1.110***
	(.058)	(.057)	(.057)	(.058)	(.060)	(.058)	(.058)	(.058)	(.058)	(.058)
	—	—	—	—	1.160***	—	—	—	—	—
	—	—	—	—	(.060)	—	—	—	—	—
Explicit^b
People act as if they think you are dishonest?	—	—	—	—	—	—	—	—	—
People act as if they are afraid of you?	.729***	.728***	.729***	.729***	.729***	.923***	.726***	.724***	.730***	.839***
	(.038)	(.038)	(.038)	(.038)	(.038)	(.052)	(.039)	(.039)	(.038)	(.050)
	—	—	—	—	—	.690***	—	—	—	.730***⁶
	—	—	—	—	—	(.035)	—	—	—	(.039)
You are called names or insulted?	.550***	.549***	.550***	.550***	.550***	.562***	.475***	.547***	.549***	.474***
	(.032)	(.032)	(.032)	(.032)	(.032)	(.031)	(.032)	(.033)	(.032)	(.032)
	—	—	—	—	—	—	.654***	—	—	.681***
	—	—	—	—	—	—	(.036)	—	—	(.039)
You are threatened or harassed?	.542***	.541***	.542***	.542***	.542***	.554***	.539***	.481***	.541***	.477***
	(.032)	(.032)	(.032)	(.032)	(.032)	(.031)	(.032)	(.032)	(.032)	(.033)
	—	—	—	—	—	—	—	.626***	—	.657***
	—	—	—	—	—	—	—	(.035)	—	(.038)
You are followed?	.617***	.617***	.617***	.618***	.617***	.634***	.616***	.615***	.607***	.624***
	(.033)	(.033)	(.033)	(.033)	(.033)	(.032)	(.033)	(.033)	(.035)	(.033)
	—	—	—	—	—	—	—	—	.549	—
	—	—	—	—	—	—	—	—	(.030)	—
Implicit
Factor mean disparity^c	−.164	−.165	−.168	−.158	−.176	−.164	−.164	−.163	−.164	−.163
SD	(.057)	(.057)	(.057)	(.058)	(.057)	(.057)	(.057)	(.057)	(.057)	.057
t score	−2.850**	−2.900**	−2.940**	−2.740**	−3.080**	−2.850**	−2.850**	−2.850**	−2.850**	−2.850**
Explicit
Factor mean disparity^d	−.599	−.599	−.599	−.599	−.599	−.570	−.617	−.599	−.612	−.609
SD	(.085)	(.085)	(.085)	(.085)	(.085)	(.083)	(.080)	(.081)	(.084)	(.077)
t score	−7.080***	−7.080***	−7.080***	−7.080***	−7.080***	−6.870***	−7.760***	−7.370***	−7.310***	−7.910***
χ²	603.100	607.870	614.130	612.510	610.120	544.070	529.480	568.920	617.540	445.900
df	106	105	105	105	105	105	105	105	105	103
RMSEA	.054	.055	.055	.055	.055	.051	.050	.053	.055	.046
CFI	.989	.989	.989	.989	.989	.990	.991	.990	.989	.992

Note: This table reflects the following order on the releasing constraints between men and women: (1) factor means, factor variances, correlations between observed errors, and errors for observed terms are allowed to vary; (2) unconstrain the factor loading for “treated with less courtesy”; (3) unconstrain factor loading for “treated with less respect”; (4) unconstrain factor loading for “poor service”; (5) unconstrain factor loading for “think you are not smart”; (6) unconstrain factor loading for “afraid of you”; (7) unconstrain factor loading for “called names or insulted”; (8) unconstrain factor loading for “threatened or harassed”; (9) unconstrain factor loadings for “are you followed”; and (10) unconstrain factor loadings for “afraid of you,” “called names or insulted,” and “threatened or harassed.” Statistical tests are two tailed (difference > 0). CFI = comparative fit index; DAS = Detroit Area Study; NSAL = National Survey of American Life; RMSEA = root mean square error of approximation.

Results in this table are from the NSAL (2001–2003) (n = 1,120 men, n = 2,073 women).

For all comparisons, the top row presents the factor loading for men, and the bottom row presents the factor loading for women.

This is the disparity for the latent factor mean, between men and women, for everyday implicit discrimination.

This is the disparity for the latent factor mean, between men and women, for everyday explicit discrimination.

p < .01. ***p < .001.

Tables 4a and 4b show that model fit is significantly improved when women and men are allowed to have different factor loadings for major life discrimination (without attribution to race) in both the DAS (Δχ² = 19.66, p = .000) and NSAL (Δχ² = 36.93, p = .000). The specific items associated with improved model fit in the DAS are “unfairly fired” (Δχ² = 3.92, p = .048), “not been hired” (Δχ² = 8.64, p = .003), and being “stopped . . . or abused by the police” (Δχ² = 17.45, p = .000). The items associated with improved model fit for the NSAL are “not been hired” (Δχ² = 23.64, p = .000) and “stopped . . . or abused by the police” (Δχ² = 31.71, p = .000). In both cases, questions assessing the perception of discrimination in employment and with the police are primarily responsible for improved model fit, suggesting that these experiences are the most gendered.

Table 4a.

Multiple-group Confirmatory Factor Analysis for Major Life Discrimination, DAS (1995).^a

Variable	1	2	3	4	5	6	7
Neighbors made life difficult for you and your family?	—	—	—	—	—	—	—
Unfairly fired or denied a promotion?^b	2.110*	2.150*	2.010*	1.980*	2.290*	2.160*	1.800*
	(.910)	(.953)	(.875)	(.818)	(1.030)	(.961)	(.774)
	—	.865*	—	—	—	—	3.05
	—	(.397)	—	—	—	—	(3.200)
Not been hired for unfair reasons?	3.950*	4.000*	3.210*	3.500*	4.220*	3.770*	2.810*
	(1.980)	(2.040)	(1.450)	(1.590)	(2.140	(1.840)	(1.230)
	—	—	1.160*	—	—	—	3.810
	—	—	(.525)	—	—	—	(4.000)
Unfairly stopped by the police?	1.560*	1.550*	1.460*	1.330*	1.670*	1.560*	1.180*
	(.672)	(.683)	(.634)	(.550)	(.745)	(.690)	(.504)
	—	—	—	1.680**	—	—	3.00
	—	—	—	(.651)	—	—	(3.080)
Unfairly discouraged from continuing education?	.984*	1.020*	1.000*	1.040**	1.210	.979*	.963**
	(.433)	(.472)	(.455)	(.401)	(.625)	(.432)	(.321)
	—	—	—	—	.586	—	—
	—	—	—	—	(.281)	—	—
Unfairly prevented from moving into a neighborhood?	2.320*	2.390*	2.360*	2.350**	2.410*	2.350*	1.960*
	(.969)	(1.050)	(1.040)	(.902)	(1.020)	(1.090)	(.927)
	—	—	—	—	—	.790*	—
	—	—	—	—	—	(.326)	—
Factor mean disparity^c	−.818	−1.090	−1.000	−.273	−.892	−.830	.277
SD	(.334)	(.480)	(.422)	(.148)	(.385)	(.341)	(.747)
t score	−2.450*	−2.270*	−2.370*	−1.850	−2.320*	−2.430*	.370
χ²	32.760	28.450	24.820	13.650	32.890	32.880	12.000
df	22	21	21	21	21	21	19
RMSEA	.041	.035	.025	.000	.044	.044	.000
CFI	.957	.970	.985	1.000	.953	.953	1.000

Note: This table reflects the following order on the releasing constraints between men and women: (1) factor means, factor variances, correlations between observed errors, and errors for observed terms are allowed to vary; (2) unconstrain the factor loading for “denied promotion”; (3) unconstrain factor loading for “not hired”; (4) unconstrain factor loading for “the police”; (5) unconstrain factor loading for “continuing education”; (6) unconstrain factor loading for “moving into neighborhood”; and (7) unconstrain factor loadings for “denied promotion,” “not hired,” and “the police.” Statistical tests are two tailed (difference > 0). CFI = comparative fit index; DAS = Detroit Area Study; NSAL = National Survey of American Life; RMSEA = root mean square error of approximation.

Results in this table are from the DAS (1995) (n = 180 men, n = 393 women).

For all comparisons, the top row presents the factor loading for men, and the bottom row presents the factor loading for women.

This is the disparity for the latent factor mean, between men and women, for major life discrimination (not attributed to any particular personal attribute).

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

Table 4b.

Multiple-group Confirmatory Factor Analysis for Major Life Discrimination, NSAL (2001–2003).^a

Variable	1	2	3	4	5	6	7	8	9	10
Neighbors made life difficult for you and your family?	—	—	—	—	—	—	—	—	—	—
Have you ever been unfairly fired?^b	1.010***	.984***	1.000***	1.000***	.923***	1.000***	1.010***	1.010***	1.020***	.952***
	(.172)	(.170)	(.174)	(.172)	(.149)	(.173)	(.172)	(.174)	(.174)	(.159)
	—	.374***	—	—	—	—	—	—	—	—
	—	(.084)	—	—	—	—	—	—	—	—
Ever not been hired for a job?	1.640***	1.640***	1.540***	1.600***	1.540***	1.600***	1.650***	1.650***	1.640***	1.510***
	(.274)	(.275)	(.259)	(.267)	(.249)	(.268)	(.276)	(.277)	(.275)	(.244)
	—	—	.472***	—	—	—	—	—	—	.882***
	—	—	(.120)	—	—	—	—	—	—	(.198)
Unfairly fired or denied a promotion?	1.610***	1.620***	1.550***	1.560***	1.440***	1.570***	1.610***	1.630***	1.620***	1.500***
	(.265)	(.267)	(.257)	(.255)	(.222)	(.259)	(.265)	(.268)	(.266)	(.241)
	—	—	—	.469***	—	—	—	—	—	—
	—	—	—	(.109)	—	—	—	—	—	—
Unfairly stopped by the police?	1.100***	1.100***	1.090***	1.090***	1.070***	1.090***	1.090***	1.110***	1.100***	1.080***
	(.189)	(.190)	(.191)	(.189)	(.178)	(.189)	(.188)	(.192)	(.190)	(.182)
	—	—	—	—	.819***	—	—	—	—	.713***
	—	—	—	—	(.158)	—	—	—	—	(.165)
Unfairly discouraged from continuing education?	1.090***	1.090***	1.090***	1.090***	1.050***	1.110***	1.090***	1.090***	1.100***	1.060***
	(.180)	(.181)	(.185)	(.181)	(.159)	(.193)	(.179)	(.181)	(.182)	(.165)
	—	—	—	—	—	.376***	—	—	—	—
	—	—	—	—	—	(.091)	—	—	—	—
Unfairly prevented from moving into a neighborhood?	1.360***	1.360***	1.330***	1.330***	1.350***	1.320***	1.330***	1.360***	1.360***	1.350***
	(.234)	(.236)	(.235)	(.231)	(.215)	(.229)	(.239)	(.235)	(.235)	(.222)
	—	—	—	—	—	—	.559***	—	—	—
	—	—	—	—	—	—	(.127)	—	—	—
Unfairly denied a bank loan?	.964***	.962***	.950***	.954***	.934***	.950***	.963***	1.000***	.967***	.941***
	(.169)	(.169)	(.171)	(.168)	(.149)	(.168)	(.169)	(.184)	(.169)	(.154)
	—	—	—	—	—	—	—	.381***	—	—
	—	—	—	—	—	—	—	(.089)	—	—
Worse repair service than other people get?	1.170***	1.180***	1.170***	1.170***	1.100***	1.170***	1.170***	1.180***	1.180***	1.120***
	(.198)	(.200)	(.203)	(.199)	(.169)	(.199)	(.197)	(.200)	(.205)	(.180)
	—	—	—	—	—	—	—	—	.475***	—
	—	—	—	—	—	—	—	—	(.099)	—
Factor mean disparity^c	−2.230	−2.340	−3.090	−2.380	−.719	−2.390	−2.150	−2.360	−2.210	−1.040
SD	(.562)	(.601)	(.928)	(.604)	(.246)	(.642)	(.557)	(.619)	(.553)	(.417)
t score	−3.970***	−3.900***	−3.330***	−3.950***	−2.920**	−3.720***	−3.850***	−3.820***	−4.000***	−2.500*
χ²	146.590	145.920	118.430	146.360	99.020	147.980	147.770	148.240	148.070	95.800
df	61	60	60	60	60	60	60	60	60	59
RMSEA	.030	.030	.025	.030	.020	.030	.030	.030	.030	.020
CFI	.970	.970	.980	.970	.986	.970	.970	.969	.970	.987

Note: This table reflects the following order on the releasing constraints between men and women: (1) factor means, factor variances, correlations between observed errors, and errors for observed terms are allowed to vary; (2) unconstrain the factor loading for “unfairly fired”, (3) unconstrain factor loading for “not hired”; (4) unconstrain factor loading for “denied a promotion”; (5) unconstrain factor loading for “the police”; (6) unconstrain factor loading for “continuing education”; (6) unconstrain factor loading for “moving into neighborhood”; (7) unconstrain factor loading for “denied loan”; (8) unconstrain factor loading for “worse repairs”; and (9) unconstrain factor loadings for “not hired” and “the police.” Statistical tests are two tailed (difference > 0). CFI = comparative fit index; DAS = Detroit Area Study; NSAL = National Survey of American Life; RMSEA = root mean square error of approximation.

Results in this table are from the NSAL (2001–2003) (n = 1,120 men, n = 2,073).

For all comparisons, the top row presents the factor loading for men, and the bottom row presents the factor loading for women.

This is the disparity for the latent factor mean, between men and women, for major life discrimination (not attributed to any particular personal attribute).

p < .05. **p < .01. ***p < .001.

Tables 5a and 5b assess improvements in model fit for major life discrimination attributed to race. The results show that allowing select items to have different factor loadings for women and men improves the overall model fit for both the DAS (Δχ² = 7.81, p = .020) and the NSAL (Δχ² = 45.72, p = .000). Specifically, findings from the DAS show that allowing the factor loading to differ between women and men for “unfairly fired” (Δχ² = 3.62, p = .057) and being “stopped . . . or abused by the police” (Δχ² = 7.49, p = .006) significantly improved model fit. Findings from the NSAL show that more than half of the questions suffer from gender-related measurement error, including “not been hired” (Δχ² = 28.23, p = .000), “stopped . . . or abused by the police” (Δχ² = 27.54, p = .000), “prevented from moving into a neighborhood” (Δχ² = 10.21, p = .001), “denied a bank loan” (Δχ² = 3.34, p = .068), and “worse repair service” (Δχ² = 7.00, p = .008). These findings suggest that the experience of discrimination is more gendered when it is attributed to race, as model improvements are associated with a wider range of questions than when compared with major life discrimination without the race attribution. In addition, overall model fit is best when perceived discrimination is operationalized as major life discrimination attributed to race.

Table 5a.

Multiple-group Confirmatory Factor Analysis for Major Life Discrimination, Attributed to Race, DAS (1995).^a

Variable	1	2	3	4	5	6	7
Neighbors made life difficult for you and your family?	—	—	—	—	—	—	—
Unfairly fired or denied a promotion?^b	1.850*	1.630*	1.790*	1.630*	1.850*	1.770*	1.630*
	(.898)	(.795)	(.870)	(.716)	(.923)	(.850)	(.719)
	—	.705	—	—	—	—	1.970
	—	(.606)	—	—	—	—	(1.320)
Not been hired for unfair reasons?	2.200*	2.020*	2.110*	2.030*	2.210*	2.360*	2.020*
	(1.060)	(.984)	(1.010)	(.884)	(1.090)	(2.000)	(.905)
	—	—	.577	—	—	—	—
	—	—	(.412)	—	—	—	—
Unfairly stopped by the police?	2.240	2.450	2.230	2.320	2.250	2.090*	2.320
	(1.170)	(1.380)	(1.180)	(1.200)	(1.210)	(1.060)	(1.200)
	—	—	—	1.130	—	—	1.140
	—	—	—	(.626)	—	—	(.682)
Unfairly discouraged from continuing education?	1.150	1.140	1.150	1.180*	1.160	1.110	1.180*
	(.674)	(.707)	(.683)	(.571)	(.756)	(.648)	(.569)
	—	—	—	—	.550	—	—
	—	—	—	—	(.411)	—	—
Unfairly prevented from moving into a neighborhood?	1.390*	1.380*	1.380*	1.420*	1.390*	1.320*	1.410*
	(.672)	(.697)	(.672)	(.597)	(.683)	(.656)	(.599)
	—	—	—	—	—	.571	—
	—	—	—	—	—	(.382)	—
Factor mean disparity^c	−1.970	−2.780	−2.180	−.275	−1.990	−1.870	−.253
SD	(1.520)	(2.500)	(1.800)	(.483)	(1.700)	(1.470)	(.638)
t score	−1.290	−1.110	−1.210	−.570	−1.170	−1.270	−.400
χ²	25.380	22.090	24.340	17.310	25.780	25.360	17.320
df	22	21	21	21	21	21	20
RMSEA	.025	.014	.025	.000	.030	.029	.000
CFI	.979	.993	.979	1.000	.970	.973	1.000

Note: This table reflects the following order on the releasing constraints between men and women: (1) factor means, factor variances, correlations between observed errors, and errors for observed terms are allowed to vary; (2) unconstrain the factor loading for “denied promotion”; (3) unconstrain factor loading for “not hired”; (4) unconstrain factor loading for “the police”; (5) unconstrain factor loading for “continuing education”; (6) unconstrain factor loading for “moving into neighborhood”; and (7) unconstrain factor loadings for “not hired” and “the police.” Statistical tests are two tailed (difference > 0). CFI = comparative fit index; DAS = Detroit Area Study; NSAL = National Survey of American Life; RMSEA = root mean square error of approximation.

Results in this table are from the DAS (1995) (n = 152 men, n = 347 women).

For all comparisons, the top row presents the factor loading for men, and the bottom row presents the factor loading for women.

This is the disparity for the latent factor mean, between men and women, for major life discrimination (not attributed to any particular personal attribute).

p < .05.

Table 5b.

Multiple-group Confirmatory Factor Analysis for Major Life Discrimination, Attributed to Race, NSAL (2001–2003).¹

Variable	1	2	3	4	5	6	7	8	9	10
Neighbors made life difficult for you and your family?	—	—	—	—	—	—	—	—	—	—
Have you ever been unfairly fired?²	.797***	.847***	.811***	.795***	.739***	.798***	.786***	.792***	.808***	.775***
	(.139)	(.155)	(.148)	(.138)	(.118)	(.140)	(.137)	(.138)	(.142)	(.136)
	—	.466***	—	—	—	—	—	—	—	—
	—	(.132)	—	—	—	—	—	—	—	—
Ever not been hired for a job?	1.370***	1.380***	1.190***	1.380***	1.170***	1.370***	1.360***	1.370***	1.380***	1.120***
	(.276)	(.277)	(.238)	(.278)	(.223)	(.277)	(.277)	(.277)	(.276)	(.224)
	—	—	.835***	—	—	—	—	—	—	1.630**
	—	—	(.277)	—	—	—	—	—	—	(.568)
Unfairly fired or denied a promotion?	1.310***	1.330***	1.360***	1.330***	1.050***	1.320***	1.280***	1.290***	1.340***	1.250***
	(.233)	(.236)	(.251)	(.239)	(.170)	(.236)	(.227)	(.230)	(.240)	(.235)
	—	—	—	.807***	—	—	—	—	—	—
	—	—	—	(.214)	—	—	—	—	—	—
Unfairly stopped by the police?	.729***	.744***	.786***	.721***	.748***	.739***	.689***	.703***	.768***	.749***
	(.141)	(.145)	(.157)	(.140)	(.142)	(.144)	(.133)	(.137)	(.150)	(.149)
	—	—	—	—	1.407***	—	—	—	—	.992**
	—	—	—	—	(.334)	—	—	—	—	(.352)
Unfairly discouraged from continuing education?	1.280***	1.280***	1.300***	1.270***	1.220***	1.330***	1.270***	1.270***	1.280***	1.260***
	(.234)	(.235)	(.252)	(.232)	(.204)	(.260)	(.234)	(.234)	(.235)	(.235)
	—	—	—	—	—	.605***	—	—	—	—
	—	—	—	—	—	(.159)	—	—	—	—
Unfairly prevented from moving into a neighborhood?	1.290***	1.280***	1.250***	1.290***	1.300***	1.270***	1.090***	1.300***	1.270***	1.050***
	(.233)	(.232)	(.236)	(.233)	(.218)	(.230)	(.201)	(.237)	(.229)	(.198)
	—	—	—	—	—	—	1.100***	—	—	1.280**
	—	—	—	—	—	—	(.273)	—	—	(.433)
Unfairly denied a bank loan?	1.200***	1.200***	1.180***	1.200***	1.180***	1.200***	1.200***	1.090***	1.200***	1.030***
	(.236)	(.236)	(.245)	(.236)	(.205)	(.237)	(.237)	(.225)	(.237)	(.220)
	—	—	—	—	—	—	—	.954***	—	1.160**
	—	—	—	—	—	—	—	(.231)	—	(.394)
Worse repair service than other people get?	1.200***	1.210***	1.240***	1.190***	1.100***	1.190***	1.180***	1.190***	1.390***	1.260
	(.224)	(.226)	(.241)	(.220)	(.190)	(.222)	(.220)	(.222)	(.279)	(.258)
	—	—	—	—	—	—	—	—	.485***	.801
	—	—	—	—	—	—	—	—	(.151)	(.304)
Factor mean disparity³	−1.220	−1.390	−2.630	−1.150	.275	−1.350	−.910	−.980	−1.720	−.560
SD	(.510)	(.582)	(1.180)	(.482)	(.271)	(.557)	(.436)	(.448)	(.698)	(.686)
t score	−2.400*	−2.390*	−2.230*	−2.380*	1.020	−2.420*	−2.080*	−2.190*	−2.460*	−.820
χ²	157.800	159.120	126.960	159.200	121.980	159.690	146.440	154.690	147.980	95.440
df	61	60	60	60	60	60	60	60	60	56
RMSEA	.032	.032	.026	.032	.025	.032	.030	.031	.030	.021
CFI	.959	.958	.972	.958	.974	.958	.964	.960	.963	.983

Note: This table reflects the following order on the releasing constraints between men and women: (1) factor means, factor variances, correlations between observed errors, and errors for observed terms are allowed to vary; (2) unconstrain the factor loading for “unfairly fired”; (3) unconstrain factor loading for “not hired”; (4) unconstrain factor loading for “denied a promotion”; (5) unconstrain factor loading for “the police”; (6) unconstrain factor loading for “continuing education”; (6) unconstrain factor loading for “moving into neighborhood”; (7) unconstrain factor loading for “denied loan”; (8) unconstrain factor loading for “worse repairs”; and (9) unconstrain factor loadings for “not hired” and “the police.” Statistical tests are two tailed (difference > 0). CFI = comparative fit index; DAS = Detroit Area Study; NSAL = National Survey of American Life; RMSEA = root mean square error of approximation.

Results in this table are from the NSAL (2001–2003) (n = 1,120 men, n = 2,073 women).

For all comparisons, the top row presents the factor loading for men, and the bottom row presents the factor loading for women.

This is the disparity for the latent factor mean, between men and women, for major life discrimination (not attributed to any particular personal attribute).

p < .05. **p < .01. ***p < .001.

Does Gendered Measurement Bias Account for the Gender Gap?

We now focus on our primary and previously unexplored research question: the degree to which gendered differences in model fit, or gender bias in measurement, is responsible for the gender gap in perceived discrimination. We find mixed evidence depending upon the data set analyzed and the type of discrimination considered. In the DAS, adjusting the model for gender-related measurement error results in a complete reduction of the gender gap in implicit discrimination (from MΔ = −2.04, p < .100, in model 1 to MΔ = −1.57, ns, in model 9). Measurement error is also responsible for 25 percent of the gap in explicit discrimination in the DAS (from MΔ = –.862, p < .001, in model 1, to MΔ = –.644, p < .010, in model 9). These reductions in the factor mean difference are supported by five of eight items, excluding “treated with less courtesy” and “treated with less respect.” For the NSAL, there are no reductions in either implicit or explicit discrimination when the base model and full models are compared. There is, however, a 5 percent reduction associated with allowing women and men to have different factor loadings for “people act as if they are afraid of you.” Because reductions found in the DAS are not replicated in the NSAL, the role gender-related measurement error plays in the gender gap in perceived everyday discrimination is not clear. Because findings are weaker in the national sample, our results suggest that measurement differences likely do not explain gender differences in perceived everyday discrimination. Although this observation is not strictly in line with the subordinate male target hypothesis on its face, it is more in line with this hypothesis than it is with intersectionality theory. We reflect more on this finding below.

In line with expectations set forth in intersectionality theory, gender-related measurement error is responsible for a significant portion of the gap in major life discrimination (Tables 4a and 4b). For major life discrimination without attribution to race, gender-related measurement error fully accounts for the gender gap in the DAS and accounts for 53 percent of the gender gap in the NSAL. In both cases, reductions in the gender gap, or factor mean difference, are driven by measurement error associated with the question that references unfair treatment by the police. Allowing women and men to have different factor loadings for this item alone is associated with a 67 percent reduction in the gender gap in both the DAS and the NSAL. Intersectionality theory and the expectation that measurement error is largely responsible for gender differences in perceived discrimination is most strongly supported by our operationalization of major life discrimination attributed to race (Tables 5a and 5b). Measurement error associated with the indicator for being “stopped . . . or abused by the police” fully accounts for the gender gap in major life discrimination attributed to race in the NSAL.

Taken together, these findings suggest that although differences in everyday discrimination are generally not due to gender-related measurement error, the way we measure major life discrimination (particularly when attributed to race) has important implications for our understanding of differential rates between women and men.

Discussion and Conclusion

Previous research shows that although African American women and men perceive similar levels of group discrimination, African American men report having more experiences with discrimination than do African American women. In this study, we tested for the role of gender-related measurement bias in explaining some or all gender differences in reports of perceived discrimination among African Americans. This study provides mixed and domain-specific support for our expectations regarding the gendered nature of perceived (racial) discrimination. Findings for major life discrimination are most consistent with the view provided by intersectionality theory (Baca Zinn and Thornton Dill 1996; Cho et al. 2013; Collins 2000, 2004; Crenshaw 1989; King 1988; Weber 2010). However, findings for everyday discrimination are somewhat mixed but appear to lean toward ideas found in the targeted male hypothesis, particularly in the national sample (e.g., Ferguson 2001; Navarrete et al. 2010).

Adjusting for gender-related measurement error provides for a partial explanation of the gender gap in major life discrimination and a full explanation for the gender gap in major life discrimination attributed to race. In both cases, discrimination in employment and unfair or mistreatment by the police played a central role in the production of gender-related measurement error. That is, discrimination in employment and mistreatment by the police better reflect the ways in which African American men perceive their experiences with (racial) discrimination. Adjusting measures of everyday discrimination for gender-related measurement error did not result in an improved model fit or any appreciable reduction in the gender gap. This suggests that, as currently measured, the ways in which scholars currently conceptualize the experience of discrimination in the form of daily hassles appears to be more gender balanced. This is likely associated with the fact that this scale was informed by a study that focused on the experiences of Black women (Essed 1991).

Limitations

Our analysis is limited and caution should be taken when interpreting these results. As noted previously, we have relied on retrospective accounts of perceived discrimination, and there is a potential discrepancy between subjective self-reports of perceived discrimination and objective experiences with discrimination. Importantly, research shows that this relationship is itself influenced by gender (Essed 1991; King 2003). Our study is also restricted to African Americans, so our findings may or may not apply to other racial or ethnic groups. Finally, although we think considering the intersection of race and gender is a key contribution of the current study, we also recognize that sexuality, social class, age, (dis)ability, and other social inequalities are likely to also influence experiences with discrimination.

Despite these limitations, the present study clearly demonstrates the value of analyzing discrimination within an intersectional framework. Although our conceptualization and measurement of everyday discrimination appears to be less affected by gender-related measurement bias than measures of major life discrimination, these measures may still be incomplete in ways that marginalize the experiences of African American women. Future studies should consider the social and institutional contexts in which African American women may be more likely to experience discrimination than African American men. For instance, given that women often function as primary caregivers, the potential for discrimination in social services and health care might uniquely shape the lives of African American women. Future research might also consider the inclusion of additional items that concern sexual harassment. For instance, “cat calling,” or unwanted sexual comments and/or advances made by men toward women, represents a potential source of gendered discrimination. Finally, we note that measures of discrimination tend to emphasize active forms of discrimination, such as being called names or insulted or being fired from a job. Experiences with more passive forms of discrimination, such as being ignored, overlooked, dismissed, or not taken seriously, are generally not included in measures of discrimination. These omissions likely produce downward bias in the prevalence of discrimination among women.

Intersectionality theorizes the interplay among race, gender, and other social statuses, and our analysis shows that African American women and men may experience or at least perceive discrimination at levels that are more similar to one another than previous studies suggest. Therefore, studies that neglect to account for gendered measurement bias risk presenting an inaccurate view, whereby the gender gap in perceived discrimination is appreciably inflated, if not wholly created, by the gendered nature of survey questions.

Footnotes

Notes

Author Biographies

Mosi Adesina Ifatunji is an Assistant Professor in the Department of Sociology and at the Carolina Population Center at the University of North Carolina at Chapel Hill. His research interests are in race, culture, biology and social stratification. Dr. Ifatunji also holds a position at the Inter-university Consortium for Social and Political Research (ICPSR) at the Institute for Social Research at the University of Michigan at Ann Arbor where he teaches a course at the Summer Program in Quantitative Methods of Social Research which focuses on Race, Ethnicity and Quantitative Methodologies. For more information, please visit: .

Catherine E. Harnois is an Associate Professor in the Departments of Sociology and Gender, Women’s, and Sexuality Studies at Wake Forest University. Her current work brings an intersectional theoretical perspective to the study of identity, consciousness, and perceived discrimination.

References

Amaro

Hortensia

Russo

Nancy Felipe

Johnson

Julie

. 1987. “Family and Work Predictors of Psychological Well-being among Hispanic Women Professionals.” Psychology of Women Quarterly 11(4):505–21.

Anderson

Elijah

. 2008. Against the Wall: Poor, Young, Black, and Male. Philadelphia: University of Pennsylvania Press.

Baca Zinn

Maxine

Dill

Bonnie Thornton

. 1996. “Theorizing Difference from Multiracial Feminism.” Feminist Studies 22(2):321–31.

Barnes

Lisa L.

Menendes-DeLeon

Carlos F.

Wilson

Robert S.

Bennett

David A.

Evans

Denis A.

2004. “Racial Differences in Perceived Discrimination in a Community Population of Older Blacks and Whites.” Journal of Aging & Health 16(3):315–37.

Beale

Frances

. 1970. “Double Jeopardy: To Be Black and Female.” Pp. 90–100 in The Black Woman: An Anthology, edited by Bambara

Toni Cade

. New York: New American Library.

Blank

Rebecca M.

Dabady

Marilyn

Citro

Constance

, eds. 2004. Measuring Racial Discrimination. Washington, DC: National Academies Press.

Blumer

Herbert

. 1958. “Race Prejudice as a Sense of Group Position.” Pacific Sociological Review 1(1):3–7.

Bobo

Lawrence D.

Fox

Cybelle

. 2003. “Race, Racism and Discrimination: Bridging Problems, Methods and Theory in Social Psychological Research.” Social Psychology Quarterly 66(4):319–32.

Bobo

Lawrence D.

Tuan

Mia

. 2006. Prejudice in Politics: Group Position, Public Opinion, and the Wisconsin Treaty Rights Dispute. Cambridge, MA: Harvard University Press.

10.

Bollen

Kenneth A.

1989. Structural Equations with Latent Variables. New York: John Wiley.

11.

Eduardo

Bonilla-Silva

. 1997. “Rethinking Racism: Toward a Structural Interpretation.” American Sociological Review 62(3):465–80.

12.

Bowleg

Lisa

. 2008. “When Black + Lesbian + Woman ≠ Black Lesbian Woman: The Methodological Challenges of Qualitative and Quantitative Intersectionality Research.” Sex Roles 59(5–6):312–25.

13.

Broman

Clifford L.

1996. “The Health Consequences of Racial Discrimination: A Study of African Americans.” Ethnicity and Disease 6(1–2):148–153.

14.

Broman

Clifford L.

Mavaddat

Roya

Hsu

Shu-yao

. 2000. “The Experience and Consequences of Perceived Racial Discrimination: A Study of African Americans.” Journal of Black Psychology 26(2):165–80.

15.

Brown

Tony N.

2001. “Measuring Self-perceived Racial and Ethnic Discrimination in Social Surveys.” Sociological Spectrum 21(3):377–92.

16.

Cheung

Gordon W.

Rensvold

Roger B.

2002. “Evaluating Goodness-of-fit Indexes for Testing Measurement Invariance.” Structural Equation Modeling: A Multidisciplinary Journal 9(2):233–55.

17.

Cho

Sumi

Crenshaw

Kimberlé Williams

McCall

Leslie

. 2013. “Toward a Field of Intersectionality Studies: Theory, Applications, and Praxis.” Signs 38(4):785–810.

18.

Cole

Elizabeth R.

2009. “Intersectionality and Research in Psychology.” American Psychologist 64(3):170–80.

19.

Collins

Patricia Hill

. 2000. Black Feminist Thought: Knowledge, Consciousness and the Politics of Empowerment. New York: Routledge.

20.

Collins

Patricia Hill

. 2004. Black Sexual Politics: African Americans, Gender and the New Racism. New York: Routledge.

21.

Crenshaw

Kimberlé Williams

. 1989. “Demarginalizing the Intersection of Race and Sex: A Black Feminist Critique of Antidiscrimination Doctrine, Feminist Theory, and Antoracist Politics.” University of Chicago Legal Forum 1989:139–68.

22.

Dill

Bonnie Thornton

. 1988. “Our Mothers’ Grief: Racial Ethnic Women and the Maintenance of Families.” Journal of Family History 13(4):415–31.

23.

Din-Dzietham

Rebecca

Nembhard

Wendy N.

Collins

Rakale

Davis

Sharon K.

2004. “Perceived Stress Following Race-based Discrimination at Work Is Associated with Hypertension in African Americans. The Metro Atlanta Heart Disease Study, 1999–2001.” Social Science and Medicine 58(3):449–61.

24.

Essed

Philomena

. 1991. Understanding Everyday Racism: An Interdisciplinary Theory. Newbury Park, CA: Sage.

25.

Ferguson

Ann Arnett

. 2001. Bad Boys: Public Schools in the Making of Black Masculinity. Ann Arbor: University of Michigan Press.

26.

Forman

Tyrone A.

Ifatunji

Mosi A.

2006. “Cohorts, Discrimination and Causal Attributions: System Blame Ideology among African Americans.” Presented at the annual meeting of the American Sociological Association, Montreal, Canada.

27.

Forman

Tyrone A.

Williams

David R.

Jackson

James S.

1997. “Race, Place and Discrimination.” Perspectives on Social Problems 9:231–61.

28.

Gary

Lawrence E.

1995. “African American Men’s Perceptions of Racial Discrimination: A Sociocultural Analysis.” Social Work Research 19(4):207–17.

29.

Hancock

Ange-Marie

. 2007. “When Multiplication Doesn’t Equal Quick Addition: Examining Intersectionality as a Research Paradigm.” Perspectives on Politics 5(1):63–79.

30.

Hardt

Jochen

Rutter

Michael

. 2004. “Validity of Adult Retrospective Reports of Adverse Childhood Experiences: Review of the Evidence.” Journal of Child Psychology and Psychiatry 45(2):260–73.

31.

Harnois

Catherine E.

2013. Feminist Measures in Survey Research. Thousand Oaks, CA: Sage.

32.

Harnois

Catherine E.

Ifatunji

Mosi

. 2011. “Gendered Measures, Gendered Models: Toward an Intersectional Analysis of Interpersonal Racial Discrimination.” Ethnic and Racial Studies 34(6):1006–28.

33.

Hausmann

Leslie R. M.

Jeong

Kwonho

Bost

James E.

Ibrahim

Said A.

2008. “Perceived Discrimination in Health Care and Health Status in a Racially Diverse Sample.” Medical Care 46(9):905–14.

34.

Herring

Cedric

Thomas

Melvin E.

Durr

Marlese

Horton

Hayward D.

1998. “Does Race Matter? The Determinants and Consequences of Self-reports of Discrimination Victimization.” Race & Society 1(2):109–23.

35.

Ifatunji

Mosi A.

Forman

Tyrone A.

2006. “Education and Discrimination: A Test of Some Mediating Mechanisms.” Presented at the annual meeting of the Association of Black Sociologists, Montreal, Canada.

36.

Jackson

James S.

Williams

David R.

1995. “Detroit Area Study: Social Influences on Health (Stress, Racism and Health Protective Services).” Ann Arbor: University of Michigan.

37.

Jackson

James S.

Torres

Myriam

Caldwell

Cleopatra H.

Neighbors

Harold W.

Nesse

Randolph M.

Taylor

Robert Joseph

Trierweiler

Steven J.

Williams

David R.

2004. “The National Survey of American Life: A Study of Racial, Ethnic and Cultural Influences on Mental Disorders and Mental Health.” International Journal of Methods in Psychiatric Research 13(4):196–207.

38.

Jones

Camara

. 2000. “Levels of Racism: A Theoretic Framework and a Gardener’s Tale.” American Journal of Public Health 90(8):1212–15.

39.

Kane

Emily W.

Kyyro

Elise K.

2001. “For Whom Does Education Enlighten? Race, Gender, Education and Beliefs about Social Inequality.” Gender & Society 15(5):710–33.

40.

Kessler

Ronald C.

Mickelson

Kristin D.

Williams

David R.

1999. “The Prevalence, Distribution and Mental Health Correlates of Perceived Discrimination in the United States.” Journal of Health & Social Behavior 40(3):208–30.

41.

King

Deborah K.

1988. “Multiple Jeopardy, Multiple Consciousness: The Context of a Black Feminist Ideology.” Signs 14(1):42.

42.

King

Kimberly R.

2003. “Do You See What I See? Effects of Group Consciousness on African American Women’s Attributions to Prejudice.” Psychology of Women Quarterly 27(1):17–30.

43.

Krieger

Nancy

. 1999. “Embodying Inequality: A Review of Concepts, Measures and Methods for Studying Health Consequences of Discrimination.” International Journal of Health Services 29(2):295–352.

44.

Krieger

Nancy

Sidney

Stephan

. 1996. “Racial Discrimination and Blood Pressure: The CARDIA Study of Black and White Adults.” American Journal of Public Health 86(10):1370–78.

45.

Krieger

Nancy

Smith

Kevin

Naishadham

Deepa

Hartman

Cathy

Barbeau

Elizabeth M.

2005. “Experiences of Discrimination: Validity and Reliability of a Self-report Measure for Population Health Research on Racism and Health.” Social Science & Medicine 61(7):1576–96.

46.

Lewis

Tené T.

Barnes

Lisa L.

Bienias

Julia L.

Lackland

Daniel T.

Evans

Denis A.

Mendes De Leon

Carlos F.

2009. “Perceived Discrimination and Blood Pressure in Older African American and White Adults.” Journals of Gerontology Series A: Biological Sciences and Medical Sciences 64(9):1002–1008.

47.

Lewis

Tené T.

Cogburn

Courtney D.

Williams

David R.

2015. “Self-reported Experiences of Discrimination and Health: Scientific Advances, Ongoing Controversies, and Emerging Issues.” Annual Review of Clinical Psychology 11:407–40.

48.

Maughan

Barbara

Rutter

Michael

. 1997. “Retrospective Reporting of Childhood Adversity: Issues in Assessing Long-term Recall.” Journal of Personality Disorders 11(1):19–33.

49.

May

Vivian

. 2015. Pursuing Intersectionality. New York: Routledge.

50.

McCall

Leslie

. 2005. “The Complexity of Intersectionality.” Signs 30(3):1771–1800.

51.

Molenaar

Peter C. M.

Nesselroade

John R.

2012. “Merging the Idiographic Filter with Dynamic Factor Analysis to Model Process.” Applied Developmental Science 16(4):210–19.

52.

Muthén

Linda K.

Muthén

Bengt O.

2012. Mplus User’s Guide. Los Angeles, CA: Muthén & Muthén.

53.

Navarrete

Carlos David

McDonald

Melissa M.

Molina

Ludwin E.

Sidanious

Jim

. 2010. “Prejudice at the Nexus of Race and Gender: An Outgroup Male Target Hypothesis.” Journal of Personality and Social Psychology 98(6):933–45.

54.

Pager

Devah

Shepherd

Hana

. 2008. “The Sociology of Discrimination: Racial Discrimination in Employment, Housing, Credit, and Consumer Markets and Minorities.” Annual Review of Sociology 34:181–209.

55.

Paradies

Yin

. 2006. “A Systematic Review of Empirical Research on Self-reported Racism and Health.” International Journal of Epidemiology 35(4):888–901.

56.

Pascoe

Elizabeth A.

Richman

Laura Smart

. 2009. “Perceived Discrimination and Health: A Meta-analytic Review.” Psychological Bulletin 135(4):531–54.

57.

Pérez

Debra Joy

Fortuna

Lisa

Alegria

Margarita

. 2008. “Prevalence and Correlates of Everyday Discrimination among US Latinos.” Journal of Community Psychology 36(4):421.

58.

Phinney

Jean S.

Madden

Tanya

Santos

Lorena J.

1998. “Psychological Variables as Predictors of Perceived Ethnic Discrimination among Minority and Immigrant Adolescents.” Journal of Applied Social Psychology 28(11):937–53.

59.

Pierce

Chester M.

Pierce-Gonzalez

Willis

1978. “An Experiment in Racism: TV Commercials.” Pp. 62–88 in Television and Education, edited by Pierce

Chester M.

Beverly Hills, CA: Sage.

60.

Pieterse

Alex L.

Carter

Robert T.

2007. “An Examination of the Relationship between General Life Stress, Racism-related Stress, and Psychological Health among Black Men.” Journal of Counseling Psychology 54(1):101–109.

61.

Reskin

Barbara F.

2012. “The Race Discrimination System.” Annual Review of Sociology 38:17–35.

62.

Rogers

Reuel

. 2001. “Black Like Who?: Afro-Caribbean Immigrants, African Americans and the Politics of Group Identity.” Pp. 163–92 in Islands in the City: West Indian Migration to New York, edited by Foner

Berkeley, Los Angeles and London: University of California Press.

63.

Sass

Daniel A.

Schmitt

Thomas A.

Marsh

Herbert W.

2014. “Evaluating Model Fit with Ordered Categorical Data within a Measurement Invariance Framework: A Comparison of Estimators.” Structural Equation Modeling: A Multidisciplinary Journal 21:167–80.

64.

Schulz

Amy J.

Gravlee

Clarence C.

Williams

David R.

Israel

Barbara A.

Mentz

Graciela

Rowe

Zachary

. 2006. “Discrimination, Symptoms of Depression, and Self-rated Health among African American Women in Detroit: Results from a Longitudinal Analysis.” American Journal of Public Health 96(7):1265–70.

65.

Sellers

Robert M.

Shelton

J. Nicole

. 2003. “The Role of Racial Identity in Perceived Racial Discrimination.” Journal of Personality & Social Psychology 84(5):1079–92.

66.

Shih

Johanna

. 2002. “‘. . . Yeah, I Could Hire This One, but I Know It’s Gonna Be a Problem’: How Race, Nativity and Gender Affect Employers’ Perceptions of the Manageability of Job Seekers.” Ethnic and Racial Studies 25(1):99–119.

67.

Sidanius

Jim

Pratto

Felicia

. 1999. Social Dominance: An Intergroup Theory of Social Hierarchy and Oppression. Cambridge, UK: Cambridge University Press.

68.

Sidanius

Jim

Veniegas

Rosemary C.

2000. “Gender and Race Discrimination: The Interactive Nature of Disadvantage.” Pp. 47–69 in Reducing Prejudice and Discrimination, edited by Oskamp

Stuart

. Mahwah, NJ: Lawrence Erlbaum.

69.

Sigelman

Lee

Welch

Susan

. 1991. “The Sources of Blacks’ Perceptions of Racial Inequality.” Pp. 67–84 in Black Americans’ Views of Racial Inequality: The Dream Differed. Cambridge, MA: Cambridge University Press.

70.

Slatton

Brittany

Spates

Kamesha

. 2014. Hyper Sexual, Hyper Masculine? Gender, Race and Sexuality in the Identities of Contemporary Black Men. Farnham, UK: Ashgate.

71.

Stucky

Brian D.

Gottfredson

Nisha C.

Panter

A. T.

Daye

Charles E.

Allen

Walter R.

Wightman

Linda F.

2011. 2011. “An Item Factor Analysis and Item Response Theory-based Revision of the Everyday Discrimination Scale.” Cultural Diversity & Ethnic Minority Psychology 17(2):175–85.

72.

Sue

Derald Wing

Capodilupo

Christina M.

Torino

Gina C.

Bucceri

Jennifer M.

Holder

Aisha

Nadal

Kevin L.

Esquilin

Marta

. 2007. “Racial Microaggressions in Everyday Life: Implications for Clinical Practice.” American Psychologist 62(4):271.

73.

Taylor

John

Turner

R. Jay

. 2002. “Perceived Discrimination, Social Stress, and Depression in the Transition to Adulthood: Racial Contrasts.” Social Psychology Quarterly 65(3):213–25.

74.

Timberlake

Jeffrey M.

Estes

Sarah Beth

. 2007. “Do Racial and Ethnic Stereotypes Depend on the Sex of Target Group Members? Evidence from a Survey-based Experiment.” Sociological Quarterly 48(3):399–433.

75.

Veenstra

Gerry

. 2013. “The Gendered Nature of Discriminatory Experiences by Race, Class, and Sexuality: A Comparison of Intersectionality Theory and the Subordinate Male Target Hypothesis.” Sex Roles 68(11–12):646–59.

76.

Waters

Mary C

. 1999. Black Identities: West Indian Immigrant Dreams and American Realities. Cambridge, Massachusetts and London, England: Harvard University Press.

77.

Weber

Lynn

. 2010. Understanding Race, Class, Gender, and Sexuality : A Conceptual Framework. New York: Oxford University Press.

78.

Weitzer

Ronald

Tuch

Steven A.

1999. “Race, Class and Perceptions of Discrimination by the Police.” Crime & Delinquency 45(4):494–507.

79.

Welch

Susan

Sigelman

Lee

Bledsoe

Timothy

Combs

Michael

. 2001. “Perceptions of Racial Discrimination.” Pp. 71–93 in Race and Place: Race Relations in an American City. Cambridge, UK: Cambridge University Press.

80.

Williams

David R.

Chung

An-Me

. 1996. “Racism and Health.” In Health in Black America, edited by Neighbors

Harold W.

Jackson

James S.

Thousand Oaks, CA: Sage.

81.

Williams

David R.

Yan

Jackson

James S.

1997. “Racial Differences in Physical and Mental Health: Socioeconomic Status, Stress and Discrimination.” Journal of Health Psychology 2(3):335–51.

82.

Wingfield

Adia Harvey

. 2007. “The Modern Mammy and the Angry Black Man: African American Professionals’ Experiences with Gendered Racism in the Workplace.” Race, Gender and Class 14(2):196–212.