The New Face of Employment Discrimination: How Do Cases of Subtle Racism Play Out for Mock Jurors?
Vanessa A. Edkins and Lindsey M. Lee
Florida Institute of Technology
Author Note: The authors would like to thank Bharati Belwalkar for her contributions to data collection and data preparation.
When employers labor to protect their workplaces from the insidious effects of racism or discrimination, typical efforts involve enacting formal policies or procedures to combat instances of differential treatment (Sturm, 2001). But what happens when the discrimination is not intentional but is instead based on unconscious biases? Can an employer be liable? These are difficult issues to manage and the courts often vary in their interpretation of the statutes (e.g. see Dukes v. Wal-Mart Stores, Inc., 2010 and Wal-Mart Stores, Inc. v. Betty Dukes et al., 2011). However, as explained below, employers can be found liable for discrimination that results from unconscious motives.
Legally, racism is often defined as behavior stemming from hostile intent involving blatant incidents such as using racial slurs to create a hostile environment or refusing to hire minorities (Sturm, 2001). Plaintiffs who cannot prove hostile intent may have a more difficult time prevailing in court, but the law does leave room for unintentional racism (Banks, Eberhardt, & Ross, 2008). In fact, the United States Supreme Court has acknowledged, for both disparate impact (Griggs v. Duke Power, 1971) and disparate treatment (Desert Palace, Inc. v. Costa, 2003) cases, that discrimination may occur without intent or may be driven by unconscious bias or stereotypical thinking. This mirrors the last 30 years of social science research, demonstrating that today’s discrimination is expressed in more subtle ways, and the use of race as a motivating factor in a decision may be unconscious (Dovidio & Gaertner, 2004; McConahay, 1986; Sears, Henry, & Kosterman, 2000). Modern-day claims of employment discrimination are more likely to consist of a culmination of smaller, more subtle behaviors (Sturm, 2001). Although the courts may have allowed room for something subtle (and perhaps unconscious) to be recognized as racism, are the triers of fact in an employment discrimination case willing to make the same leap?
Court rulings aside, the common perception is that cases void of hostile intent are difficult, if not impossible, to win (King et al., 2011; Tolson, 2008). Possibly this is because the average person (read: “juror”) may not perceive the actions depicted as discriminatory in nature. Whereas African-Americans are more likely to view racism as institutionalized and systemic, present in many everyday experiences, research indicates that Whites are less likely to view these subtle incidents as racism (Bobo, 2001).
King et al. (2011) further suggest that the difficulty targets experience in prevailing in their claims of racial discrimination may be explained by system justification theory (Jost & Banaji, 1994). System-justifying beliefs (SJBs) are culturally established beliefs that individuals may endorse in order to explain why their society is set up in a hierarchical manner. These beliefs differ from culture to culture, but in the U.S., many of the SJBs fall under one umbrella: meritocracy. The meritocratic worldview (MWV) is the cultural belief that hard work and industriousness lead to success (Protestant work ethic) and that people are responsible for their own fate (belief in a just world)—they have the ability to succeed, if they are willing to persevere, and they are responsible for their failures (Kaiser & Major, 2006). The internalization of failure that this perspective breeds encourages the perception that inequalities exist not because of systemic discrimination but because individuals (or entire social groups) have not worked hard enough. Thus, one of the reasons that people may be less likely to see racism in ambiguous events is because these underlying meritocracy beliefs may allow fact finders to rationalize discriminatory behaviors as nondiscriminatory, misattributing negative target outcomes to the target’s failure to persevere.
The Effects of Discrimination on a Target
If Whites are less likely to acknowledge the presence of systemic racism, might this also mean that they are less likely to see it as harmful at the individual level? Research on target experiences of discrimination is lagging behind research on perpetrators of discrimination, but we are beginning to understand the effects of discrimination. For example, the experience of discrimination has been associated with negative self-reported physical health (Schulz et al., 2000), elevated blood pressure (Guyll, Mathews, & Bromberger, 2001), diagnosis of depression (Clark, Anderson, Clark, & Williams, 1999), and impaired psychological well-being (Williams, Yu, Jackson, & Anderson, 1997). If being a target of discrimination means experiencing such harms, could the display of the harm make an outsider more likely to perceive that discrimination occurred?
Research investigating perceptions of sexist behavior has taken a closer look at this proposition (Swim, Scott, Sechrist, Campbell, & Stangor, 2003). Swim and colleagues investigated how harm and intent influenced perceptions of an act as prejudiced. What they found was that if the intent of an action was obviously discriminatory (i.e., a man excluding women because he believed women were inferior) then whether or not harm occurred had very little impact; overt, blatant discrimination is judged as such, regardless of its effects on the target. On the other hand, if intent is more ambiguous, participants found information on harm to be helpful in judgments of discrimination. Incidents of subtle discrimination are, by their very nature, lacking in intent, suggesting that these types of cases may have jurors seeking out (or putting more emphasis on) information outlining the harm caused by the actions.
We performed a study, hypothesizing that, as prior research suggests, individuals would be more likely to rule in favor of the plaintiff in employment discrimination claims with overt racism and in favor of the defendant in claims featuring covert racism. We further expected that the addition of information highlighting the impact of the actions on the plaintiff will increase verdicts for the plaintiff, especially in cases of covert discrimination in which intent in ambiguous.
Testing the Effect of Harm
The Swim et al. (2003) research investigating harm implications on judgments of discrimination focused on everyday instances of sexism, not cases that would necessarily be relevant in the legal arena. To keep our research relevant to employment discrimination, we created a disparate treatment accusation in a case of failure to promote.
With regard to employment discrimination claims, evidence of employee well-being may be introduced in the form of employee performance evaluations, and in fact, plaintiff attorneys contend that evidence portraying performance-related information is very important for an employee in litigation (Bisom-Rapp, 1999). The performance-related documentation that plaintiff attorneys feel is important includes things like a history of poor evaluations or poor evaluations that began, presumably as retaliation, as soon as an employee complained about discriminatory treatment. The attorneys overwhelmingly agree that such information presented at a trial could be very persuasive: Consistently poor evaluations may bolster the defendant’s claim that a firing was not race related, but evaluations that dropped dramatically at the time of complaint-filing may bolster a plaintiff’s contention that discrimination was occurring. However, we were unable to find any literature investigating how a fact finder may respond to actual decreases in performance occurring at the time of the discriminatory action in question, not just those recorded by a supervisor in retaliation to a complaint filing. Based on the previous literature discussing the negative effects of discrimination on the target (Clark et al., 2007; Williams et al., 1997), it seems entirely feasible that a discriminatory act may result in impaired performance and in decreased psychological well-being; thus, our study used this type of evidence to demonstrate alleged harm.
One hundred and six undergraduate students (58 men, 46 women, 2 failed to indicate) participating to gain course credit were randomly assigned to read one of four vignettes, differing by type of discrimination (overt vs. covert) and presence of harm (no mention vs. performance impact); two were excluded from analyses for not following directions. Participant age ranged from 18 to 52, with a mean age of 20.24 years old (SD = 3.02) and a median of 19. Fifty-eight of our participants identified as White, 14 as Black/African-American, 9 as Latino/Hispanic American, 3 as Asian American, 21 indicated “other;” and 1 participant did not indicate. Thirty-one of the participants responding were not American citizens, but their decisions on verdicts did not differ from the 73 American citizen respondents (χ2 (1, N = 102) < .01, p = .99) and so analyses were collapsed across citizenship. The same was true for gender (χ2 (1, N = 103) = 2.89, p = .09). All participants read a vignette describing an employee’s allegation of discriminatory treatment on the basis of his employer’s failure to promote; the employee was depicted as having an excellent work history, prior to the events, and as an asset to the company. Participants were asked to act as though they were jurors deciding the case and rendered a verdict for the plaintiff or for the defendant, following instructions that mirrored state law for discrimination claims based on disparate treatment. Using a scale developed by Gilbert and Lownes-Jackson (2005), we also obtained ratings of the plaintiff as an employee (based on the information provided in the given scenario) along six dimensions (competence, a = .67; potential, a = .74; collegiality, a = .86; emotional stability, a = .92; seriousness about work, a = .84; and demeanor, a = .86) to explore how perceptions of the plaintiff influenced verdicts.1
1Materials are available from the authors upon request.
Verdict preferences (see Table 1) supported our first hypothesis: 63.4% of those in the overt conditions decided for the plaintiff but only 36.6% of those in the covert conditions decided for the plaintiff (χ2 (1, N = 104) = 4.20, p = .04), lending credence to the belief that subtle discrimination cases are less likely than cases of blatant discrimination to result in a verdict favorable to the plaintiff. Although this is unsurprising, the analyses for our second hypothesis take a completely different turn. We predicted that adding performance impact to the covert condition (C-P) would result in significantly higher verdicts than covert racism with no performance impact (C-NP). The difference between these two groups was not significant (χ2 (1, N = 52) = 1.20, p = .27) and the trend was in the opposite direction than that which we predicted. Individuals in the C-NP condition decided for the plaintiff at a rate of 36.0%, but the addition of performance impact information (C-P) dropped this to a rate of 22.2%. Although again not significant, the same trend was also noted in the overt racism conditions: verdicts for the plaintiff dropped from 59.3% to 40% (χ2 (1, N = 33) = 1.46, p = .23) with the addition of information on performance impact. In fact, when we collapsed across type of discrimination, the presence of performance impact acted counter to what we hypothesized, showing a rate of plaintiff verdicts of 30.8% compared to 48.1% without performance impact (χ2 (1, N = 104) = 3.26, p = .07). Harm, it seems, is taken into consideration regardless of information on intent, and has a negative impact on a plaintiff’s case.
Rates of Verdicts by Discrimination Type and Presence or Absence of Performance Impact
[Table 1 goes here]
Determining Verdict Influences
Using a logistic regression to predict verdict, our independent variables (and their interaction) were entered on the first step and the ratings of the plaintiff as an employee were entered on the second step (forward selection was used in Step 2 because we were taking an exploratory approach). Although individuals in the overt conditions had decided for the plaintiff at a higher rate than those in the covert conditions, when performance impact and the interaction between condition and performance were included as predictors, the difference between the overall model and the constant only approached significance (χ2 (3, N = 99) = 7.19, p = .07). The addition of the ratings of the plaintiff as an employee resulted in much better model fit: χ2 (5, N = 99) = 56.80, p < .001 and explained between 43.7% (Cox and Snell R square) and 59.3% (Nagelkerke R squared) of verdict decisions. Overall classification significantly increased (χ2 (2, N =99) = 49.61, p < .001) from 64.6% in Step 1 to 87.9%; verdicts for the defendant were classified correctly approximately 92% of the time and for the plaintiff 82% of the time. Our exploratory model increased the prediction of plaintiff verdicts by 42.5%, and perhaps what was most interesting is that this significant increase was due solely to the addition of two variables: perceptions of the plaintiff’s emotional stability and perceptions regarding how serious the plaintiff is about his work. Table 2 shows the contribution of the individual predictors to the overall model. In the final model, emotional stability and work seriousness were the only variables to reliably enhance model prediction at p < .05. According to the results, perceiving the plaintiff as emotionally stable made an individual about 2.4 times more likely to decide in his favor, and perceiving him as serious about his work, 1.9 times more likely to decide in his favor. Rather than evaluating whether or not the behavior of the defendant was discriminatory, the focus was on how well the plaintiff could withstand the alleged discrimination.
Summary of Stepwise Regression Analysis for Variables Predicting Verdicts (N = 99)
[Table 2 goes here]
Note. All regression coefficients are unstandardized. OR = odds ratio; CI = confidence interval; Cox and Snell R2 = 0.07 for Step 1 and R2 = 0.44 for Step 2. *p <.05; **p < .01
Theoretical Implications and Future Directions
Why are our mock jurors so focused on the attributes of the plaintiff and not of the actions of the defendant? Kaiser and Major (2006) have suggested that fact finders charged with deciding a case of employment discrimination may make their determination based not on the actions of the defendant but instead on characteristics of the plaintiff that may justify the actions. In our study, we handed the characteristics right to the mock jurors: emotional stability and seriousness about work. What we depicted in our scenario to demonstrate the psychological effects of discrimination (i.e., increased sick leave, seeking out stress-management classes) may have been interpreted as signs of emotional instability and a lack of seriousness about work, giving those deciding the case a reason for the promotion denial that was internal to the plaintiff. If you are unsure about the emotional stability of your employee, why would you promote him or her to a higher position? Even though our employee was depicted as a model worker up until the failure to promote, and the manifestations of psychological distress occurred only after this incident, internal attributions made about the employee guided decisions of whether or not the employer acted in a discriminatory manner when the company denied the promotion.
This focus on the target may also be why our results differed from the Swim et al. (2003) research. In their vignette that most closely resembled an employment discrimination situation, the researchers demonstrated harm in terms of outcomes that were more objective in nature (e.g., women not invited to computer training were later fired for lacking those skills) and that occurred to multiple women—it is presumably much more difficult to attribute blame to the target when multiple targets are involved. As mentioned previously, current research suggests that what may be driving the tendency to look toward the alleged victim in discrimination cases may have something to do with system-justifying beliefs (Kaiser, Dyrenforth, & Hagiwara, 2006). In our case, individuals may be using the reliance on a meritocratic worldview as a mask for discrimination (Kaiser et al., 2006). This idea echoes theories of aversive racism (Dovidio & Gaernter, 2004), which hold that individuals may behave in an outwardly egalitarian manner unless behaving in a discriminatory manner can be attributed to a valid reason that has nothing to do with race. Theoretically, a meritocratic worldview may make this rationalization of discriminatory behavior easy: If we believe that an individual was not trying his or her hardest, then deciding against him or her has nothing to do with our views and everything to do with his or her ability (not race).
Although the research on SJBs is still in its burgeoning stages, at least one study has found that endorsement of a meritocratic worldview predicted the derogation of minority race members (Kaiser et al., 2006), and it seems possible that in our study we have an example of meritocratic worldview in action. Individuals are turning away from the actions of the defendant and toward the attributes of the plaintiff to explain the incident in question. Our participants placed greater importance on the emotional stability and seriousness toward work measures—measures that directly reflect a person’s ability to get ahead on the basis of merit—than on the alleged discriminatory behavior. Although factors like competence and potential (two other measures collected) seem to make more sense as verdict influences from a meritocracy perspective, participants were unable to explain defendant verdict preferences in these terms because the information provided to the participants emphasized a hard-working and skilled employee. In order to justify deciding for the defendant more, our mock jurors decided that the most important information was not the objective measures of competence or potential but the subjective, personal attributes encompassed by emotional stability and seriousness about work. Our future research will seek to first replicate the results found here (with a larger sample in order to increase power) and also determine the role that meritocratic worldview may play in explaining these unexpected findings.
In the end, this paper seems to have raised more questions than it answered. One thing that is apparent is that more research is needed on this new face of employment discrimination. The topic is both pertinent and complicated, with no easy answers for either side. Businesses will need to be more vigilant with their policies and possibly educate managers and supervisors on the many forms that discrimination may take. If our meritocratic beliefs impede us from seeing discrimination in a court of law where the issues are salient, then it seems unlikely to be acknowledged in the everyday working lives of employees. Therefore, employers must be more proactive in educating their workforce on modern expressions of employment discrimination. On the flip side, employees, who already have less than a 25% success rate in employment discrimination claims that reach the level of litigation (Zink & Gutman, 2004), are likely to find themselves with an even more difficult battle: Not only are triers of fact less likely to see the acts as discriminatory, but they may seek out information about the plaintiff to explain the incident. It seems that this new face is not so pretty.
Banks, R., Eberhardt, J., & Ross, L. (2008). Race, crime and antidiscrimination. In E. Borgida & S.T. Fiske (Eds.), Beyond common sense: Psychological science in the courtroom. Malden, MA: Blackwell.
Bisom-Rapp, S. (1999). Bulletproofing the workplace: Symbol and substance in employment discrimination law practice. Florida State University Law Review, 26, 959–1049.
Bobo, L. D. (2001). Racial attitudes and relations at the close of the twentieth century. In N. J. Smelser, W. J. Wilson, & F. Mitchell (Eds.), America becoming: Racial trends and their consequences (Vol. 1, pp. 264-301). Washington, DC: National Academy Press.
Clark, R., Anderson, N. B., Clark, V. R., & Williams, D. R. (1999). Racism as a stressor for African-Americans: A biopsychosocial model. American Psychologist, 54(10), 805–816. doi:10.1037/0003-066X.54.10.805.
Desert Palace, Inc v. Costa, 2003, 539 U.S. 90.
Dovidio, J. F., & Gaertner, S. L. (2004). Aversive racism. Advances in Experimental Social Psychology, 36, 1–51. doi:10.1016/S0065-2601(04)36001-6.
Dukes v. Wal-Mart Stores, Inc., 2010, 603 F.3d 571.
Gilbert, J. A., & Lownes-Jackson, M. (2005). Blacks, whites, and the new prejudice: Does aversive racism impact employee assessment? Journal of Applied Social Psychology, 35(7), 1540–1553. doi:10.1111/j.1559-1816.2005.tb02183.x.
Griggs v. Duke Power Co., 1971, 401 U.S. 424.
Guyll, M., Matthews, K., & Bromberger, J. (2001). Discrimination and unfair treatment: Relationship to cardiovascular reactivity among African-American and European American women. Health Psychology, 20(5), 315–325. doi:10.1037/0278-618.104.22.1685.
Jost, J. T., & Banaji, M. R. (1994). The role of stereotyping in system justification and the production of false consciousness. British Journal of Social Psychology, 33, 1–27.
Kaiser, C. R., Dyrenforth, P. S., & Hagiwara, N. (2006). Why are attributions to discrimination interpersonally costly? A test of system- and group-justifying motivations. Personality and Social Psychology Bulletin, 32(11), 1423–1536. doi:10.1177/0146167206291475.
Kaiser, C. R., & Major, B. (2006). A social psychological perspective on perceiving and reporting discrimination. Law & Social Inquiry, 31(4), 801–830. doi:10.1111/j.1747-4469.2006.00036.x.
King, E. B., Dunleavy, D. G., Dunleavy, E. M., Jaffer, S., Botsford Morgan, W., Elder, K., Graebner, R. (2011). Discrimination in the 21st century: Are science and the law aligned? Psychology, Public Policy, and Law, 17(1), 54–75. doi:10.1037/a0021673.
McConahay, J. B. (1986). Modern racism, ambivalence, and the modern racism scale. In J. F. Dovidio & S. L. Gaertner (Eds.), Prejudice, discrimination, and racism (pp. 91–125). Orlando, FL: Academic Press.
Schulz, A., Israel, B., Williams, D., Parker, E., Becker, A., & James, S. (2000). Social inequalities, stressors and self-reported health status among African-American and white women in the Detroit metropolitan area. Social Science & Medicine, 51(11), 1639–1653. doi:10.1016/S0277-9536(00)00084-8.
Sears, D. O., Henry, P. J., & Kosterman, R. (2000). Egalitarian values and the origins of contemporary American racism. In D. O. Sears, J. Sidanius, & L. Bobo (Eds.), Racialized politics: The debate about racism in America (pp. 75–117). Chicago, IL: University of Chicago Press.
Sturm, S. (2001). Second generation employment discrimination: A structural approach. Columbia Law Review, 101, 458–569.
Swim, J. K., Scott, E. D., Sechrist, G. B., Campbell, B., & Stangor, C. K. (2003). The role of intent and harm in judgments of prejudice and discrimination. Journal of Personality and Social Psychology, 84, 944–959. doi:10.1037/0022-3522.214.171.1244
Tolson, F. (2008). The boundaries of litigating unconscious discrimination: Firm-based remedies in response to a hostile judiciary. Delaware Journal of Corporate Law, 33, 347–421.
Wal-Mart Stores, Inc. v. Betty Dukes et al., 2011, 131 S. Ct. 2541.
Williams, D., Yu, Y., Jackson, J., & Anderson, N. (1997). Racial differences in physical and mental health: Socioeconomic status, stress, and discrimination. Journal of Health Psychology, 2(3), 335–351. doi:10.1177/135910539700200305.
Zink, D. L., & Gutman, A. (2004). Statistical trends in private sector employment discrimination suits. In F. J. Landy (Ed.), Employment discrimination litigation: Behavioral, quantitative, and legal perspectives. San Francisco, CA: Jossey-Bass.