key: cord-0693083-mrxpmqw8
authors: Skinner-Dorkenoo, Allison L.; Sarmal, Apoorva; Rogbeer, Kasheena; André, Chloe; Patel, Bhumi; Cha, Leah
title: Highlighting COVID-19 racial disparities can reduce support for safety precautions among White U.S. residents
date: 2022-03-29
journal: Soc Sci Med
DOI: 10.1016/j.socscimed.2022.114951
sha: 7d56d735898042c0578abcd65fe082172f4b9aa9
doc_id: 693083
cord_uid: mrxpmqw8

U.S. media has extensively covered racial disparities in COVID-19 infections and deaths, which may ironically reduce public concern about COVID-19. In two preregistered studies (conducted in the fall of 2020), we examined whether perceptions of COVID-19 racial disparities predict White U.S. residents’ attitudes toward COVID-19. Utilizing a correlational design (N = 498), we found that those who perceived COVID-19 racial disparities to be greater reported reduced fear of COVID-19, which predicted reduced support for COVID-19 safety precautions. In Study 2, we manipulated exposure to information about COVID-19 racial disparities (N = 1,505). Reading about the persistent inequalities that produced COVID-19 racial disparities reduced fear of COVID-19, empathy for those vulnerable to COVID-19, and support for safety precautions. These findings suggest that publicizing racial health disparities has the potential to create a vicious cycle wherein raising awareness reduces support for the very policies that could protect public health and reduce disparities.

7 support for COVID-19 safety precautions. In Study 1 our preregistered hypothesis was that perceiving COVID-19 racial disparities to be greater would be associated with reduced fear of COVID-19 and support for COVID-19 safety precautions. In Study 2 our preregistered hypothesis was that exposure to information about COVID-19 racial disparities would reduce fear of COVID-19 and support for COVID-19 safety precautions. We included an exploratory measure of empathic concern in Study 2, as past research shows that empathy is particularly relevant in predicting willingness to make personal sacrifices for the good of others (Batson et al., 2015) . Data collected from several Western countries has provided evidence that empathic concern is predictive of motivation to adhere to COVID-19 safety recommendations (Pfattheicher et al., 2020) . In both studies, we accounted for knowledge of systemic contributors to COVID-19 racial disparities-which was anticipated to predict increased fear of and support for safety precautions. We also report exploratory analyses for both studies that include political orientation as a statistical control. In addition to this, we also assessed and preregistered hypotheses related to associations between perceived COVID-19 racial disparities and racial biases and social distancing. Although our hypotheses were generally supported (see supplemental materials), similar findings have already been reported elsewhere (see Lu et al., 2021) and, given the space constraints and scope of the journal, we do not report them here.

Both studies were approved by the [BLINDED FOR REVIEW] IRB and complied with all ethical regulations.

In Study 1, we assessed White U.S. residents' (a) perceptions of COVID-19 racial disparities, (b) systemic knowledge of contributors to COVID-19 racial disparities, (c) fear of and (d) support for COVID-19 safety precautions. We predicted that among White U.S. residents, perceiving COVID-19 racial disparities to be greater would be associated with reduced fear of COVID-19 and decreased support for COVID-19 safety precautions. We also tested an exploratory mediation model to examine whether fear of COVID-19 statistically mediated the relation between perceptions of COVID-19 racial disparities and support for COVID-19 safety precautions.

We recruited 500 participants based on a G*Power analysis indicating that a sample of 496 was needed to detect a small effect size (linear regression, f 2 = 0.02) with 80% power.

English-speaking, White U.S. adults (18 years or older) were recruited from MTurk between September 5th and 8th, 2020 in exchange for $2.50. After excluding participants who failed more than one of our four data quality checks (n = 2; see Open Science Framework, OSF), our final sample of 498 participants (from 48 different states; Mage = 40.56, SDage = 12.08) consisted of 58% men and 41% women (three non-binary participants) with a median income of $40,000 to $49,999.

Participants accessed the study online and provided informed consent before answering two data quality check questions (used to screen for bots and people who were not reading the questions). Next, participants reported their perceptions of racial disparities in COVID-19 and their systemic knowledge of contributors to COVID-19 racial disparities. Participants were then presented with the following set of measures in a randomized order: fear of COVID-19, support for COVID-19 safety precautions, intentions to socially avoid people of different races, and racial attitudes. Before being debriefed, participants responded to demographic questions and J o u r n a l P r e -p r o o f two additional data quality measures. See Open Science Framework for full materials (including exploratory measures of colorblindness, news consumption, and behavioral changes in response to , data, and statistical syntax.

Perceived COVID-19 Racial Disparities. Using a scale ranging from 1 (the impact on White people is worse than the impact on people of color) to 7 (the impact on people of color is worse than the impact on White people), participants reported their perceptions of the impact of COVID-19 with respect to "COVID-19 infection rates" and "COVID-19 related deaths" (r = .67). The two items were averaged to create a composite measure of perceived racial disparities.

Systemic Knowledge. Using a six-item measure ranging from 1 (strongly disagree) to 7 (strongly agree), participants indicated their agreement with true statements about the systemic and structural contributors to COVID-19 racial disparities. Items included: "Longstanding racial health disparities in the U.S. have led people of color to be harder hit by COVID-19;" "Relative to White people in the U.S., people of color are disproportionately employed as 'essential workers'-increasing their risk of exposure to COVID-19;" "As a result of structural inequalities, people of color in the U.S. are more likely to become infected with COVID-19;" "Relative to White people in the U.S., people of color tend to have more limited access to quality healthcare, increasing the impact of COVID-19 on people of color;" "People of color in the U.S. are less likely to have access to drive-through testing services (relative to White people), which increases the spread of COVID-19 among people of color;" and "People of color are less likely to have access to protective gear (masks, gloves, hand sanitizers) than White people in the U.S., increasing the likelihood of spreading COVID-19 among people of color." Responses to these J o u r n a l P r e -p r o o f six items were averaged (α = .93) to create a composite measure of knowledge of the systemic and structural contributors to COVID-19 racial disparities.

Fear of Contracting COVID-19. Using Ahorsu and colleagues' (2020) seven-item measure, participants indicated their fear of contracting COVID-19 (e.g., "I cannot sleep because I am worrying about getting COVID-19") on a scale of 1 (strongly disagree) to 5 (strongly agree). To assess participants' fear of friends and relatives contracting COVID-19, we added the following two-items adapted from Ahorsu and colleagues' scale: "I am afraid of family members or friends losing their lives because of COVID-19" and "My heart races or palpitates when I think about family and friends getting COVID-19." We averaged the seven items from the original scale (α = .92) and the two additional items (r = .57) separately to create two composite scores indexing fear for oneself and fear for loved ones.

was measured using two sets of items designed to capture support for (a) government-issued safety measures and (b) the adoption of safety precautions by the public. The first set of items measured participants support (1 = completely unsupportive to 5 = completely supportive) for the following public health policies: "limiting the maximum capacity of public spaces," "public mask requirements," "international travel restrictions," "closing non-essential businesses," and "stay at home order" (α = .90). The second set of items assessed participants' level of personal agreement (1 = strongly disagree to 5 = strongly agree) that the U.S. public should abide by the following public health recommendations: "social distancing (staying 6 feet away from other people)," "wearing masks in public spaces," "avoiding non-essential travel," "avoiding large/crowded public gatherings," and "staying at home/avoiding public spaces" (α = .93).

J o u r n a l P r e -p r o o f

Those who are knowledgeable about systemic causes of COVID-19 racial disparities could be expected to show greater concern about COVID-19 and increased support for safety precautions, but they are necessarily aware of COVID-19 racial disparities themselves. This means that by including knowledge of the systemic causes of COVID-19 racial disparities in our models, we are able to estimate the association between perceptions of COVID-19 disparities and concern about COVID-19 (e.g., fear, support for safety precautions) that is unrelated to knowledge of systemic causes of disparities. Bivariate correlations between all variables and results without systemic knowledge included as a covariate (which looked markedly different) are reported in the Supplemental Materials.

A multivariate analysis was used to examine the association between the two independent variables (perceived COVID-19 racial disparities and systemic knowledge) and fear of COVID-19 (for self and for loved ones). As predicted, those who perceived COVID-19 racial disparities 

We used a multivariate analysis to test the association between the two independent variables and support for COVID-19 safety precautions (government-issued safety measures and adoption of safety precautions by the public). Individual differences in perceived COVID-19 racial disparities did not predict support for COVID-19 safety precautions (p = .410). However, greater systemic knowledge predicted increased support for COVID-19 safety precautions (b = 0.30, SE = 0.03, 95% CI [0.24, 0.36], β = 0.47), t(495) = 9.84, p < .001.

When political orientation was added to the model, perceptions of COVID-19 racial disparities still did not predict support for COVID-19 safety precautions (p = .239). Increased conservativism predicted reduced support for COVID-19 safety precautions (B = -0.23, SE = 0.02), t(494)= -9.86, p < .001.

We speculated that perceiving COVID-19 racial disparities to be greater may only reduce support for safety precautions if it leads to a reduction in fear of COVID-19. To test this, we 

Discussion Study 1 provided initial support for our preregistered hypotheses. We found that White U.S. residents who perceived there to be greater racial disparities in COVID-19 were less fearful of COVID-19. In addition, an exploratory mediation analysis indicated that this reduced fear of COVID-19 may decrease support for safety precautions. Thus, these findings suggest that publicizing racial disparities in COVID-19 could possibly lead the public to be less supportive of safety measures that could limit the spread of COVID-19. In contrast, systemic knowledge of contributors to COVID-19 racial disparities predicted increased fear and support for COVID-19

safety precautions-such that greater systemic knowledge seemed to reflect increased awareness and concern about systemic inequalities and racism.

In Study 2, we used a between-subjects design to manipulate exposure to information about COVID-19 racial disparities among White U.S. residents. Given known tendencies for people to perceive the status quo as just (Kay et al., 2009 ) and recent evidence of racist interpretations of COVID-19 racial disparities in the U.S. (Crist & Schlegel, 2021; Perry et al., 2021 )-indicating that they are a result of biological susceptibility or irresponsible behaviorwe chose to test the effect of two slightly different messages about racial disparities in COVID-19. One message framed COVID-19 racial disparities as a product of persistent health inequalities (providing an explanation for COVID-19 racial disparities) whereas the other did not provide any explanations for COVID-19 racial disparities. Communications scholars have highlighted the potential benefits of framing racial health disparities in terms of structural contributors (Niederdeppe et al., 2008) . Thus, given that systemic knowledge of the contributors to COVID-19 racial disparities was associated with more fear of COVID-19 and support for J o u r n a l P r e -p r o o f safety precautions in Study 1, we reasoned that framing COVID-19 racial disparities as a product of persistent health inequalities could alter the impact of exposure to information about COVID-19 racial disparities. We also included an exploratory measure of empathy for those vulnerable to COVID-19, as this has previously been found to predict motivation to adhere to COVID-19 safety recommendations (Pfattheicher et al., 2020) . In addition, we tested an exploratory mediation model to see if fear of COVID-19 or empathy for those vulnerable to statistically mediated the relation between experimental conditions and support for COVID-19 safety precautions.

We tripled the Study 1 sample size, which provided sufficient power to reliably detect a small effect size (f = 0.07) with 80% power and used the same inclusion criteria and recruitment methods as Study 1 (excluding participants who completed Study 1). A total of 1,515 White U.S. residents completed the study between October 25th, 2020 and November 3rd, 2020 in exchange for $1.75. After exclusions (n = 10), our final sample of 1,505 (representing all 50 states and Washington D.C.; Mage = 41.58, SDage = 12.83) consisted of 46% men and 54% women (7 non-binary participants) with a median income of $50,000-$59,999.

After consenting to the study, participants were randomly assigned to read one of three articles about the impact of COVID-19 on the U.S. public, which were based on COVID-19 public health data from the months leading up to data collection (see OSF materials). The 

All scales showed similar levels of internal consistency to Study 1 (see Supplemental

Empathy. We used Pfattheicher and colleagues' (2020) three-item empathy scale (α = .94) to measure participants' ability to relate to those most vulnerable to COVID-19 (e.g., "I am very concerned about those most vulnerable to coronavirus (COVID-19)") on a scale from 1 (strongly disagree) to 7 (strongly agree).

Statistical models were parallel to those used in Study 1, but instead of having individual differences in perceived disparities predict our outcomes, experimental condition was included as the key independent variable (see Table 1 Experimental condition no longer significantly predicted fear of COVID-19 (p = .117) and did not interact with political orientation (p = .433). However, experimental condition remained a significant predictor of support for COVID-19 safety precautions (p < .001), and there was a marginally significant interaction between political orientation and experimental condition (p = .053). Simple slopes analyses indicated that increased conservatism predicted reduced support for COVID-19 safety precautions among those in the control condition (p < .001) and among those who received the persistent inequalities framing for racial disparities (p = .016). In contrast, among those who just read about COVID-19 racial disparities, political orientation was unrelated to support for COVID-19 safety precautions (p = .324). Experimental condition also remained a significant predictor of empathy (p = .015) and political orientation did not moderate the effect of experimental condition (p = .263).

We conducted an exploratory mediation analysis to examine whether reduced fear and empathy could statistically explain the effect of priming persistent inequalities on support for COVID-19 safety precautions (see Figure 2 ). To do this, we created composite measures of fear and support for safety precautions (by averaging the two scores that made up each of these measures for the primary multivariate analyses) and standardized all continuous variables. 

In Study 2, we replicated the systemic knowledge findings from Study 1 and observed initial evidence that informing White U.S. residents of COVID-19 racial disparities can causally reduce fear of COVID-19 and support for safety precautions. Previous research had shown a similar pattern of results among White U.S. residents who reported relatively high racial biases (Harell & Lieberman, 2021; Stephens-Dougan, 2021 ), but our findings-which were largely J o u r n a l P r e -p r o o f unmoderated by political orientation-suggest a more general phenomenon. We had speculated that framing COVID-19 racial disparities as a product of persistent health inequalities might serve as a buffer against this tendency, consistent with individual differences in systemic knowledge. However, we observed even less concern about COVID-19 when racial disparities were framed as a result of systemic inequalities. In addition, our exploratory mediation analysis provided suggestive evidence that the reduced fear and empathy for those vulnerable to COVID-19 that resulted from framing COVID-19 racial disparities as a product of persistent health inequalities, may have reduced support for safety precautions. Taken together, these findings point to the potential drawbacks of highlighting COVID-19 racial disparities among White U.S.

Our studies examined the association between White U.S. residents' perceptions of COVID-19 racial disparities and their attitudes toward COVID-19. We hypothesized that those who perceived racial disparities in COVID-19 to be greater would be less concerned about COVID-19-as indicated by reduced fear and support for safety precautions. Indeed, both studies provided some evidence of this tendency. In Study 1, we observed that those who perceived COVID-19 racial disparities in the U.S. to be greater were less fearful of COVID-19, and our exploratory mediation was consistent with the notion that this may lead to reduced support for safety precautions. Our experimental manipulation (Study 2) provided evidence that informing White U.S. residents of COVID-19 racial disparities can reduce fear, empathy, and support for COVID-19 safety precautions. We postulated that framing COVID-19 racial disparities as a product of persistent health inequalities could undermine the tendency to justify the status quo (Kay et al., 2009) and help increase concern about COVID-19. However, our findings indicate that this framing further reduced fear and support for safety precautionsperhaps because this information signaled that these disparities were not just transitory epidemiological trends which could potentially shift and disproportionately impact White people in the future. In fact, our exploratory mediation analysis provided suggestive evidence that both reduced fear of COVID-19 and empathy for those most vulnerable may contribute to the observed reduction in support for safety precautions. Emerging evidence is consistent with this interpretation: White U.S. residents who perceived COVID-19 racial disparities to be a result of genetic susceptibility reported lower adherence to CDC COVID-19 safety recommendations (Crist & Schlegel, 2021) . Our findings are also in line with trends observed in public opinion polls indicating that White U.S. residents increased their support for reopening the economy-in spite of public health risks-between late March and early June of 2020 (Cox, 2020), which corresponds with the release of data indicating racial health disparities in COVID-19.

Overall, individual differences in systemic knowledge of the contributors to COVID-19 racial disparities predicted greater fear of COVID-19, support for safety precautions, and empathy for those most vulnerable. Given the parallels between the systemic knowledge measure and the persistent inequalities framing condition, it may be surprising that these factors had opposing effects on our outcomes. However, the persistent inequalities article merely framed racial disparities as a product of persistent inequalities and did not provide extensive education on the systemically biased policies and practices that gave rise to persistent inequalities. The systemic knowledge measure, in contrast, captures a deeper understanding of systemic racism, consistent with prior work on theories of racial ignorance (e.g., Bonam et al., 2019; Mueller, 2018; Nelson et al., 2012; Salter et al., 2018) . Furthermore, supplementary J o u r n a l P r e -p r o o f analyses indicated that across both studies systemic knowledge was strongly correlated (rs > .70) with acknowledgement of systemic racism in the U.S. (Neville et al., 2000) .

Examining the impact of efforts to address and educate people about population health disparities is critically important (Lewis et al., 2020) , and our work is not the first to conclude that informing people of racial health disparities has the potential to backfire (Harell & Lieberman, 2021; Lee et al., 2017; Niederdeppe et al., 2008; Stephens-Dougan, 2021) .

Experimental investigation of reactions to systemic inequality in an entirely different domainthe criminal justice system-has shown a pattern similar to our study, such that alerting White U.S. citizens of racial disparities in the criminal justice system reduced support for criminal justice reform (Hetey & Eberhardt, 2014; 2018) . Indeed, psychologists and communications scholars have argued that educating the public about systemic inequalities in a way that motivates (rather than depresses) health and social policy changes could be critical to addressing inequalities (Hetey & Eberhardt, 2018; Niederdeppe et al., 2008) . Previous scholarship has recommended the use of narratives and visual images in public communication about population health disparities; thus, incorporating these factors in future research may lead to more optimal outcomes (Niederdeppe et al., 2008) .

Future work should also examine the generality of these findings across U.S. residents and messages about racial disparities. Although our data included White participants from all U.S. states and was fairly representative of the White U.S. adult population, our participants tended to be younger (residents 65+ were underrepresented) and had lower incomes than the median U.S. household (U.S. Census Bureau, 2019) . It is also worth considering how discussions of disparities develop over the course of a public health crisis. For instance, the emergence of COVID-19 vaccines in early 2021-which were disproportionately received by J o u r n a l P r e -p r o o f White U.S. residents during the initial rollout (DiRago et al. in press; Sacarny & Daw, 2021 )may have further reduced fear of COVID-19 and support for safety precautions beyond what we observed in our samples. Future work will also be critical for determining the extent to which the findings observed here generalize to other public health issues and messages about disparities. Even if the effects observed here do not generalize to all messages about racial disparities, the fact that some such messages can reduce public concern about health crises and dampen support for precautions among White U.S. residents should raise serious concerns.

Future research that examines a broader stimulus set (Yarkoni, 2020) and considers how to publicly address racial and ethnic disparities without spurring these unintended consequences will be essential.

Our findings suggest that alerting the public to racial health disparities (such as those observed with COVID-19) has the potential to perpetuate systemic racial inequalities-by reducing White U.S. residents' concerns about public health issues and support for policies aimed at protecting public health. Political and social power remains concentrated among White U.S. residents; thus, their attitudes toward public health issues have considerable implications for the establishment of public health policy. Our findings show that when COVID-19 racial disparities are highlighted, it can decrease White U.S. residents' willingness to engage in safety precautions themselves and support for policies protecting public health. Thus, publicizing COVID-19 racial disparities could reduce support for the very policies that aim to limit the toll of the pandemic, creating a vicious cycle wherein raising awareness of systemic racial health disparities results in public responses that actually exacerbate these disparities. Given these findings, we argue that care should be taken when considering how we talk about racial J o u r n a l P r e -p r o o f Note. Means in the same row with different subscripts significantly differ (p < .05) from one another.

J o u r n a l P r e -p r o o f Note. bc is the indirect effect of X on Y and bc' is the direct effect of X on Y when the path through the mediator is included in the model, **p < .01, ***p < .001 

• Tested relations with White US residents' awareness of COVID-19 racial disparities • Perceiving greater COVID-19 racial disparities predicted reduced fear of COVID-19

• Exposure to information about COVID-19 racial disparities reduced empathy • Reduced empathy and fear of COVID-19 predict reduced support for safety precautions • Highlighting racial disparities may paradoxically perpetuate racial inequalities

Incorporating immersive virtual environments in health promotion campaigns: A construal level theory approach. Health Communication

The Fear of COVID-19 Scale: Development and initial validation

Filipinos make up 4% of nurses in the US, but 31.5% of nurse deaths from COVID-19

The color of coronavirus: COVID-19 deaths by race and ethnicity in the

Using empathy to improve intergroup attitudes and relations

Oxford library of psychology. The Oxford perceptions of others' pain

Wisconsin chief justice sparks backlash by saying covid-19 outbreak is among meatpacking workers, not 'the regular folks

Race, ethnicity, and age trends in persons who died from COVID-19 -United States

Intergroup differences in the sharing of emotive states: Neural evidence of an empathy gap

How information about race-based health disparities affects policy preferences: Evidence from a survey experiment about the COVID-19 pandemic in the United States

Differential occupational risk for COVID-19 and other infection exposure according to race and ethnicity

Introduction to mediation, moderation, and conditional process analysis: A regression-based approach

Racial disparities in incarceration increase acceptance of punitive policies

The numbers don't speak for themselves: Racial disparities and the persistence of inequality in the criminal justice system

Assumptions about life hardship and pain perception

Healthcare workers of color nearly twice as likely to contract Covid-19 -study. The Guardian

Inequality, discrimination, and the power of the status quo: Direct evidence for a motivation to see the way things are as the way they should be

The behavior of same-race others and its effects on Black patients' attention to publicly presented HIV-prevention information. Health Communication

Residential segregation

Effects of upward and downward social comparison information on the efficacy of an appearance-based sun protection intervention: a randomized, controlled experiment

Neural basis of extraordinary empathy and altruistic motivation

Racial bias in pain perception and response: Experimental examination of automatic and deliberate processes

Advancing a sociology of ignorance in the study of racism and racial nonknowing

Communities in Action: Pathways to Health Equity

Message design strategies to raise public awareness of social determinants of health and population health disparities

The Marley Hypothesis: Denial of racism reflects ignorance of history

Construction and initial validation of the Color-Blind Racial Attitudes Scale (CoBRAS)

Risk of COVID-19 among frontline healthcare workers and the general community: A prospective cohort study. The Lancet: Public Health

Prejudice and pandemic in the promised land: How white Christian nationalism shapes Americans' racist and xenophobic views of COVID-19

The emotional path to action: Empathy promotes physical distancing and wearing of face masks during the COVID-19 pandemic

Leveraging upward social comparison in social media to promote healthy parenting. Health Communication. Advance online publication

Children's evaluations of individually and structurally based inequalities: The role of status

So it is, so it shall be: Group regularities license children's prescriptive judgments

Group presence, category labels, and generic statements influence children to treat descriptive group regularities as prescriptive

Children's descriptive-toprescriptive tendency replicates (and varies) cross-culturally: Evidence from China

The psychology of American racism

Inequities in COVID-19 vaccination rates in the 9 largest US cities

Racism in the structure of everyday worlds: A cultural-psychological perspective

Knowing how they feel: Perceiving emotions felt by outgroups

How microaggressions reinforce and perpetuate systemic racism in the United States

White Americans' reactions to racial disparities in COVID-19

Social comparison: Why, with whom, and with what effect?

People represent their own mental states more distinctly than those of others

Many White Americans are ready to reopen the economy

Construal-level theory of psychological distance

American Community Survey 5-Year Estimates

Too far to care? Measuring public attention and fear for Ebola using Twitter

A superhumanization bias in Whites' perceptions of Blacks

Social comparison in adjustment to breast cancer

The generalizability crisis