Differential Effects of Personal-Level vs Group-Level Racial Discrimination on Health among Black Americans

Abstract

Objective

Racial/ethnic minorities in the United States not only experience discrimination personally but also witness or hear about fellow in-group members experiencing discrimination (ie, group-level discrimination). The objective of our study was to examine whether the effects of group-level discrimination on mental and physical health are different from those of personal-level discrimination among Black Americans by drawing upon social psychology research of the Personal/Group Discrimination Discrepancy.

Design and Setting

We conducted a secondary analysis of cross-sectional survey data from a larger study.

Participants

One hundred and twenty participants, who self-identified as Black/African Americans during the laboratory sessions (57.5% women, mean age = 48.97, standard deviation = 8.58) in the parent study, were included in our analyses.

Main Predictor Measures

Perceived personal-level discrimination was assessed with five items that were taken from two existing measures, and group-level racial discrimination was assessed with three items.

Main Outcome Measures

Self-reported physical and mental health were assessed with a modified version of SF-8.

Results

Perceived personal-level racial discrimination was associated with worse mental health. In contrast, perceived group-level racial discrimination was associated with better mental as well as physical health.

Conclusions

Perceived group-level racial discrimination may serve as one of several health protective factors even when individuals perceive personal-level racial discrimination. The present findings demonstrate the importance of examining both personal- and group-level experiences of racial discrimination as they independently relate to health outcomes for Black Americans.

Introduction

Experiencing racial discrimination personally (ie, personal-level discrimination) is stressful—people not only report psychological distress 1-4 but also experience biological stress reactions following the experience of discrimination.5-10 Because stress, both daily hassles and major life events, negatively affects mental and physical health,11-16 the experience of personal-level discrimination, as a stressor, can have profound effects on health.17-20 It is now widely hypothesized that the experience of personal-level discrimination is one major factor contributing to the pervasive and prevalent health disparities that many racial/ethnic minorities face in the United States.21-25

However, racial/ethnic minorities not only experience discrimination personally, but also witness or hear about fellow in-group members experiencing discrimination (ie, group-level discrimination). This is particularly true in the current society, where more and more people are exposed to information about other people’s experiences with discrimination (eg, unarmed Black Americans being unfairly treated by the police, homosexual couples being refused service at restaurants), possibly because sharing information is easier than ever with the help of advanced technology (eg, smart phone) and social networking websites (eg, Facebook).26,27

Does group-level discrimination have similar effects on individuals’ health as personal-level discrimination? No study to date, to our knowledge, has empirically addressed this question. However, according to social psychology research of the Personal/Group Discrimination Discrepancy (PGDD), which refers to the individual’s tendency to perceive and report greater amount of discrimination against their social group than themselves personally as members of that group,28,29 personal- and group-level discrimination are hypothesized to have differential effects on psychological well-being, such as self-esteem and affect.30-32 For example, it has been shown that, among Belgian women and African immigrants in Belgium, those who reported more, as opposed to less, personal-level discrimination had significantly lower self-esteem, whereas those who reported more, as opposed to less, group-level discrimination had significantly higher self-esteem than those who reported less group-level discrimination.31 These opposite effects of personal- vs group-level discrimination on psychological well-being have been further replicated by an experimental study in which the experience of personal- vs group-level discrimination was experimentally induced,32 suggesting a causal negative relation between personal-level discrimination and self-esteem and a causal positive relation between group-level discrimination and self-esteem. Taken together, these studies provide strong evidence that group-level discrimination may buffer the negative psychological consequences of personal-level discrimination.

Our study built on this PGDD and psychological well-being literature and investigated the separate effects of personal- vs group-level discrimination on mental and physical health. We hypothesized that there are independent effects of personal- and group-level discrimination on health. Specifically, we predicted that Black Americans who report higher levels of perceived personal-level discrimination would report poorer mental and physical health than Black Americans who report lower levels of perceived personal-level discrimination. It was also predicted that Black Americans who report higher levels of perceived group-level discrimination would report better mental and physical health than Black Americans who report lower levels of perceived group-level discrimination. We also explored whether or not there would be interactive effects of personal- and group-level discrimination on mental and physical health. However, we did not make a specific prediction because the findings in the current PGDD literature are mixed, with some finding an interacting effect30 and others finding no evidence of such effect.31,32

Methods

Participants

One-hundred and thirty individuals who met the following eligibility criteria were recruited in the metropolitan Richmond, Virginia area through flyers or purposive referral sampling: 1) aged ≥ 35 years; 2) self-identified as Black/African American; 3) was able to come into the lab for one hour; and 4) had a permanent address and a personal phone number. Of those 130 individuals, nine individuals indicated their race as other than Black/African American (5 American Indian/Alaska native, 2 Asian/Pacific Islander, 3 Caucasian/White) during the laboratory session even though they self-identified as Black/African American during the screening. These individuals were excluded from all analyses. Additionally, one participant did not report their age. Because age was included as a control variable in the main analyses, this resulted in 120 analyzable cases (57.5% women, mean age = 48.97, standard deviation = 8.58).

Procedure

The data analyzed in our study came from a larger study approved by the Virginia Commonwealth University IRB and conducted between October, 2013 and December, 2014. In the original study, up to two participants reported to each laboratory session that was run by one of nine Black research assistants. Participants first underwent a baseline vital check. Next, they completed a computer-based questionnaire, which included the measures analyzed in our study. Participants then underwent another vital check and finally participated in a physical feature assessment (eg, skin tone, nose shape, lip thickness).33 Upon completion of the study, participants were debriefed and received $40 for their participation.

The questionnaire consisted of a series of measures designed to assess their beliefs and attitudes, experiences of discrimination, and mental and physical health. In the following section, only the measures that are the focus of our study are discussed. All procedures were in accordance with the ethical standards of the responsible committee on human experimentation (institutional and national) and with the Helsinki Declaration of 1975, as revised in 2000. Informed consent was obtained from all participants being included in the study.

Measures

Perceived Personal-level Discrimination

Five items from two separate measures34,35 were selected to assess perceived personal-level discrimination. Example items include “I feel like I am personally a victim of society because of my race” and “I personally have been a victim of racial discrimination.” Participants’ responses to the five items on a scale ranging from 1 (Strongly disagree) to 5 (Strongly agree) were averaged to compute a composite score (α = .84). Higher numbers indicate more experience of personal-level racial discrimination. It should be noted that this measure was chosen over more frequently used measures in health disparities research, which tend to result in many zero values,36-38 in order to better capture variability in perceived personal-level discrimination among Blacks. The composite scores were normally distributed with skewness = -.12 (SE = .21) and kurtosis = -.58 (SE = .42).

Perceived Group-level Discrimination

Perceived group-level discrimination was assessed with three items: “Other members of my race experience discrimination,” “My racial group is discriminated against,” and “My racial group has been victimized by society.” Participants’ responses to the three items on a scale ranging from 1 (Strongly disagree) to 5 (Strongly agree) were averaged to compute a single score (α = .84). Higher numbers indicate greater perceptions of group-level discrimination. The composite scores for perceived group-level discrimination also had a normal distribution with skewness = -.65 (SE = .21) and kurtosis = -.24 (SE = .42).

Self-reported Mental and Physical Health

An 8-item modified version of SF-8 Health Survey39 was used to assess participants’ self-reported physical and mental health during the past 4 weeks. SF-8 has been found to be reliable and valid40,41 and is designed to assess the eight subscales from widely-used SF-36: general health; physical functioning; role limitations due to physical problems; pain; energy; social functioning; emotional well-being; and role limitations due to emotional problems. Following the standard guidelines, the first four items were averaged to compute a composite score for physical health (α = .85), and the latter four items were averaged to compute a composite score for mental health (α = .76).39 The scale ranged from 0 to 100, with higher numbers indicating better health.

Statistics

For each mental and physical health, a multiple regression with a main effect of perceived personal-level discrimination, a main effect of perceived group-level discrimination, and an interaction-term between the two as predictors was conducted. Before being entered into the model, both perceived personal- and group-level discrimination were grand-mean-centered. Age, household income, and education were included in each regression as control variables because age was associated with mental health (r= .18, P = .05), household income was associated with both physical (r= .21, P = .02) and mental health (r= .23, P = .01), and education was associated with physical health (r = .25, P = .01). Sex was excluded from control variables because it was not associated with mental or physical health and the inclusion of gender as a control variable did not change the main results. However, it should be noted that, consistent with prior research,42-44 men reported more perceived personal-level discrimination (M = 3.56, SD = .95) than women (M = 2.90, SD = .86), t(119) = 4.00, SE = .16, P < .0001.

Results

Table 1 summarizes descriptive statistics for major variables. Consistent with previous research, participants reported higher perceived group-level discrimination (M = 3.86, SD = .85) than perceived personal-level discrimination (M = 3.18. SD = .96) on average. Additionally, correlation coefficients indicate that participants who reported higher levels of personal-level discrimination were also likely to report higher levels of group-level discrimination (r = .49, P < .001) and that participants who reported better mental health were also likely to report better physical health (r = .67, P < .001).

Table 1. Means, standard deviations, and correlations among all major variables (N=120).

	1	2	3	4
Personal-level discrimination	--
Group-level discrimination	.49b	--
Mental health	-.22a	.06	--
Physical health	-.02	.18 a	.67b	--
M	3.18	3.86	68.85	69.89
SD	.96	.85	21.33	21.91

a. P < .05

b. P< .001

Self-Reported Mental Health

Table 2 summarizes the findings of both mental and physical health. The overall model predicting mental health was significant, adjusted R² = .11, F(6,113) = 3.52, MSE = 403.61, P = .003. Consistent with our prediction, both the main effects of perceived personal-level discrimination (B = -7.33, SE = 2.35, P = .002) and perceived group-level discrimination (B = 5.97, SE = 2.85, P = .04) were significant. Specifically, Black Americans who reported personally experiencing a greater degree of racial discrimination had poorer mental health than those who reported experiencing less racial discrimination. In contrast, Black Americans who perceived a greater, as opposed to lesser, degree of group-level discrimination had better mental health. Finally, the interaction between perceived personal- and group-level discrimination was not significant (B = 1.99, SE = 2.52, P = .43).

Table 2. Multiple regressions predicting mental and physical health.

	Mental Health				Physical Health
	B	SE	β	t	B	SE	β	t
Intercept	38.44	12.06		3.19	45.25	12.61		3.59
Covariates
Age	.45	.22	.18	2.04a	.14	.23	.06	.62
Income	3.21	1.63	.19	1.96a	2.51	1.71	.14	1.47
Education	.14	1.28	.01	.11	2.02	1.34	.15	1.51
Personal-level discrimination	5.97	2.85	.24	2.09a	6.28	2.98	.25	2.11a
Group-level discrimination	-7.33	2.35	-.33	-3.13b	-3.49	2.45	-.15	-1.42
Personal X discrimination	1.99	2.52	.08	.79	3.95	2.63	.15	1.50

a. P < .05

b. P< .001

Self-Reported Physical Health

A multiple regression revealed that the overall model predicting physical health was significant, adjusted R² = .08, F(6,113) = 2.72, MSE = 441.69, P = .02. Consistent with our prediction, there was a significant positive association between perceived group-level discrimination and physical health (B = 6.28, SE = 2.98, P = .04). This indicates that Black Americans who perceive that fellow Black Americans experience racial discrimination to a greater degree had better physical health than those who perceive that other Black Americans experience racial discrimination to a lesser degree. In contrast, inconsistent with our prediction, the main effect of perceived personal-level discrimination did not reach statistical significance. However, it should be noted that the direction of the association between perceived personal-level discrimination and physical health was consistent with the prediction, such that greater experiences of personal-level discrimination was associated with worse physical health (B = -3.49, SE = 2.45, P = .16). Finally, the interaction between perceived personal- and group-level discrimination did not reach statistical significance (B = 3.95, SE = 2.63, P = .14).

Discussion

The negative health consequences of the experience of discrimination is well-documented.19,20,23,45,46 However, prior research has generally focused on the effects of personal-level discrimination on mental health, and relatively little is known about the effects of witnessing/hearing about fellow in-group members’ experience with discrimination on mental and physical health. Our study contributes to the growing literature of the discrimination-health disparities literature by drawing upon social psychology research of PGDD and investigating the independent effects of personal- and group-level discrimination on mental and physical health among Black Americans.

The findings of our study were generally consistent with our predictions. Black Americans who reported higher levels of perceived personal-level discrimination had poorer mental health than those who reported lower levels of perceived personal-level discrimination. Although the effects of perceived personal-level discrimination on physical health did not reach statistical significance, the direction of the regression coefficient was consistent with the prediction, such that Black Americans who reported higher levels of perceived personal-level discrimination tended to have poorer physical health than those who reported lower levels of perceived personal-level discrimination. The null finding of self-reported physical health might be due to small sample size. It might be also due to a difference in the amount of time it takes for stress to manifest in mental vs physical health. The experience of stress that is psychological in nature, such as the experience of personal racial discrimination, is likely to be more proximal to mental health than to physical health; therefore, it may manifest in mental health much faster than in physical health, particularly in our relatively young sample. Indeed, the cumulative effect of the biological mechanisms thought to underlie the stress-physical health (eg, increased allostatic load,47-49 upregulation of pro-inflammatory genes,50-53 alterations in the DNA methylation patterns54-57) may not manifest until decades of exposure.58

Our hypotheses about the effects of perceived group-level discrimination were also supported by our findings. More specifically, Black Americans who reported higher levels of perceived group-level discrimination had better mental and physical health than those who reported lower levels of perceived group-level discrimination. These findings are consistent with prior research of the PGDD showing that greater reports of group-level discrimination is associated with better psychological well-being.30-32 Finally, there was no evidence for an interactive effect of personal- and group-level discrimination on mental and physical health. That is, Black Americans who reported higher levels of perceived group-level discrimination had better mental and physical health than those who reported lower levels of perceived group-level discrimination, regardless of whether they reported higher or lower levels of perceived personal-level discrimination. These findings suggest that perceived group-level discrimination could serve as one of several health proactive factors even when Black Americans perceived discrimination against themselves.

While our study cannot address why group-level discrimination had differential effects of personal-level discrimination on health, several well-established social psychology theories point to potential explanations. First, according to the discounting hypothesis,59 an attribution of unfair treatment to racial discrimination, rather than to internal, personal characteristics, such as flaws in personality or abilities (eg, “I did not get a job not because I did not meet the qualifications but because the interviewer was a racist”) can serve to protect psychological well-being.60,61 If individuals perceive that their fellow in-group members are also experiencing the same kinds of unfair treatment, then it is easier for them to attribute the unfair treatment to discrimination rather than to their personal characteristics. Second, according to the relative deprivation theory,62 subjective judgments that one is better off than others is associated with positive personal outcomes,63 and people often engage in downward comparisons to maintain and/or boost their self-esteem and positive affect.64-66 Given that Black Americans in the current sample perceived more discrimination happening to fellow in-group members than to themselves on average, reflected in a higher mean of perceived group-level discrimination than that of perceived personal-level discrimination (3.86 vs 3.18, respectively), this is suggestive of downward comparison. Supporting the positive effects of downward comparison, a post-hoc analysis using between personal- and group-level discrimination difference scores as a predictor (ie, positive numbers indicating more group- than persona-level discrimination and thus more downward comparison) revealed that Black Americans who engaged in downward comparison to a greater degree had better mental health (B = 6.21, SE = 2.07, P = .003). Although it did not reach statistical significance, the direction of the relation between difference scores and physical health was also consistent with the downward comparison (B = 3.14, SE = 2.18, P = .15). Finally, perceptions of group-level discrimination may allow individuals to feel that they are not alone in their struggle.31,32 Future research systematically investigating these potential mechanisms is strongly encouraged.

Our findings should be interpreted in light of study limitations. The relatively small sample size may have limited our ability to detect significant associations for physical health and interactions between personal and group-level discrimination. Other limitations include the exclusion of other racial/ethnic minority groups for which racial discrimination is a salient stressor, lack of regional diversity, and limited focus on older Black Americans. The findings from our study should be replicated in larger, diverse samples of racial/ethnic minorities.

Conclusions

Our study adds to both applied health disparities research and social psychology research testing the predictions of the PGDD. Disparities research on the discrimination-health link has generally focused on the effects of the experience of personal-level discrimination. Our present findings demonstrate the importance of examining both personal- and group-level experiences of discrimination as they independently relate to health outcomes for Black Americans. It should be noted that we are not arguing that perception of group-level discrimination is good for health among Black Americans. Rather, an examination of the effects of group-level discrimination would provide additional information about factors that make Black Americans more resilient against the well-documented negative health consequences of the experience of personal-level discrimination. Additionally, social psychology research on the PGDD has primarily focused on psychological well-being (eg, self-esteem and affect), which are relatively malleable. The findings from our study extends this work to other measures of mental health and physical health, and suggests that PGDD is related to these broader health indicators as well.