The Psychology of Fairness
This post as well as part 1 and part 3 examine the recent Supreme Court ruling on affirmative action in the context of social psychological research on fairness and justice. What do psychologists have to say about why so many people see affirmative action as unfair? The most relevant area of research is the work on social justice perceptions. The perceived fairness of distributed outcomes (e.g., job offers, college admissions, compensation) depends on the social norms that are most important in the situation (Deutsch, 1975).
Distribute outcomes based on contributions. The norm of equity or merit requires a distribution of outcomes commensurate with the contributions individuals have made. Thus, those who do the most work or otherwise bring to the situation the most personal resources (e.g., skills, past performances) receive the most rewards. Equity becomes important where maximizing economic productivity is the primary objective (e.g., producing successful graduates in a school or making a profit in a company), the resources to be distributed are relatively scarce, and it is believed that rewarding based on merit yields the highest level of performance.
Distribute outcomes based on need. Another norm is need which dictates that those who are the most disadvantaged should receive the most. This is a salient norm when the group is primarily concerned with the welfare or development of members. A family group is an example. In most families (perhaps not all) children are given things they did not earn because they need more. In other words, even though they may contribute little to the “performance” of the family (e.g., they don’t cook the meals, bring in a paycheck, repair the car), parents bestow on them love and attention to help them mature into healthy adults. The philosopher John Rawls (1999) stressed need in his Difference Principle of social justice when he stated that inequalities cannot be avoided in a society, but that social and economic inequalities should work for the good of the least advantaged. He addressed educational opportunities when he wrote that “Resources for education are not to be allotted solely or necessarily mainly according to their return as estimated in productive trained abilities, but also according to their worth in enriching the personal and social life of citizens, including here the less favored. As a society progresses the latter consideration becomes increasingly more important” (p. 92).
Distribute outcomes equally. If the norm is for equality, a just distribution of outcomes is giving everyone the same amount. Equality is most salient when there are sufficient resources to allow an equal distribution among supplicants and the primary concern is maintaining good relations among participants. For instance, a team of friends may decide to divide a reward equally to maintain cohesion rather than attempting to reward for merit or need. Another instance where equality may become dominant is where the relevant inputs and outcomes cannot be measured. What constitutes the criteria for judging the success of an organization is always to some degree subjective but in some cases it may be impossible to quantify or ascertain the relative contributions to achieving success. Rather than risk judgements of unfairness as the consequence of using invalid measures, the decision is to simply treat members of the group as if there were no differences.
Distribute outcomes in a just manner. Apart from whether the distribution meets or violates what is most valued in the situation, perceptions of justice also depend on at least three other aspects of how outcomes are distributed (Colquitt, 2001). These include the extent to which the procedures in distributing the outcomes are consistent, unbiased, participative, ethical, and could be appealed (procedural justice), recipients are treated with dignity and respect (interpersonal justice), and an adequate explanation and justification for the distribution is provided (informational justice).
The perceived procedural fairness of a selection decision in employment situations and in college admissions depends to some extent on the method that is used and the attribute that is measured. A study of employment practices in 15 European countries, the United States, and South Africa revealed that interviews, work samples (e.g., a typing test), resumes, and references were perceived most favorably. Less favorably received were cognitive ability tests (i.e., IQ tests), personality inventories, and scored biographical measures. The least favorable were personal contacts, honesty tests, and graphology (Anderson et al, 2010). The small amount of research that has been conducted shows that sex, age, and ethnicity do not affect the perceived favorability of selection procedures used in employment (Hausknecht, Day & Thomas, 2004). The procedural fairness of college admissions also depends on whether the method and attribute measured are appropriate. A USA Today/ Suffolk University Poll conducted in 2019 reported that fewer than 20% of respondents thought that university admission decisions were fair. By a 3 -1 margin, respondents thought that universities favored the rich and powerful in their admission decisions. The overwhelming sentiment was that admission procedures should not favor the children of parents giving large donations to the university, athletes, and minorities. Although older respondents were more favorable toward the standardized tests such as the SAT than younger respondents, most respondents favored their use.
The self-serving nature of fairness perceptions. In an actual selection or admission procedure, the elephants in the room are whether the applicants score well on tests and other selection instruments and whether they are accepted or rejected (Van Vianen, Taris, Scholten & Schinkel, 2004). A favorable outcome is likely to lead to perceptions that the procedures are fair. Consequently, fairness is less of an issue when outcomes are favorable (i. e., you are admitted to college, you are hired) than when they are not favorable (i.e., you are rejected for admission, you do not get a job offer). There is some evidence of a two-step process in which there is an assessment of whether outcome was favorable or unfavorable followed by an assessment of whether the distribution of outcomes is fair or unfair. Recipients seem more likely to become concerned with whether the distribution is fair or unfair when they receive an unfavorable outcome (see Brockner & Wiesenfeld, 1996). I am reminded of when a classmate and I interviewed for a scholarship during our senior year in high school. When a Hispanic candidate received the scholarship, the first reaction in our 17-year-old brains was to believe the committee was guilty of bias in favor of minorities. As I matured, I realized that a more likely explanation was that I had given the worst interview performance known to humankind. Along these lines, Miguel Unzueta and colleagues (2008) concluded from their research that believing quotas were being used to benefit minorities was a way that white men protected their self-esteem when they performed poorly on an intelligence test in a selection situation.
The predominance of equity norms. For the most part, applied psychology has been focused on how to fulfill an equity norm and has tended to neglect need and equality as ways of achieving distributive justice. Maintaining equity by rewarding people proportional to their contributions is preferred over equality and need in western culture and especially the United States. This is not surprising given the centrality of capitalism in western societies and the importance of the work ethic. An entire subdiscipline of psychology (Industrial and Organizational Psychology) emerged in the early 1900s using scientific methods to identify the most valid and reliable ways of selecting job applicants who provided the best fit to job requirements. The underlying assumption was that employers needed to distribute employment opportunities based on objective qualifications shown in research to be valid predictors of important outcomes. In other words, personnel selection served an equity norm in which those most likely to succeed on the job were given preference over those less likely.
Retributive and restorative justice. The discussion so far deals with the immediate reaction of recipients to a distribution of resources in which they are advantaged or disadvantaged. In the typical study, there is no history of past injustice involved. What happens when recipients believe that an unfair distribution of resources has occurred repeatedly, and the unfair outcome was only one manifestation of a historical pattern? There are two notable exceptions to this neglect. In work on retributive justice, fairness is achieved when those who were responsible for past injustices are punished (Wenzel & Okomoto, 2016). This was one response to the Nazi holocaust, and it led to the conviction and punishment of a small portion of the Germans who were responsible for the holocaust (see Mary Fulbrook, 2017). This is more effective in restoring a sense of justice where specific wrongdoers are identifiable and the punishment occurs soon after the criminal act. Although it may feel good to see someone punished, with the passage of time guilty parties are harder to find, memory fades, and more people see the past wrongs as irrelevant to present concerns. For instance, during Reconstruction and the Jim Crow era, there were individuals who were clearly guilty of lynchings and other racial hate crimes, but they were not prosecuted. Who do we now punish?
Restorative justice is achieved when good relations are reinstituted between the perpetrator of the injustice and the victim (Cohen, 2016). Germany may have fallen short on retributive justice in letting so many of the murderers responsible for the holocaust escape punishment, but they have been exemplary in their efforts at restorative justice (see Feldman, 2012). People are likely to see a situation as unfair if the guilty go unpunished, the punishment does not fit the crime, or the innocent are punished. Moreover, restorative justice requires the admission of guilt and is likely to be resisted if the accused is convinced of his or her innocence.
White people are likely to rejective affirmative action when they see it as an attempt to punish or redeem them for the past wrongs of their racial group. Research has shown that when people perceive their ingroup to be guilty of past misdeeds they are more accepting of remedial action, including preferential treatment (Swim & Miller, 1999). Those who do not feel collective guilt, however, are more likely to see affirmative action as an unfair assignment of guilt or punishment for crimes they did not commit. Such a reaction appears likely given the small percentages of white respondents in national surveys who admit to feeling guilty over past discrimination against African Americans (see Hawley, 2017).
The Norms for Justice Applied to the AA Ruling
Affirmative action can clash with all the norms that psychologists have shown to influence perceptions of fairness. Although much of the majority opinion in the recent Supreme Court case on affirmative action builds on legal reasoning and precedent, we see all the norms for fairness at work.
Distributive justice – equity: the justices seem to argue that the affirmative action program was unjust because Black and Hispanic applicants were not selected for their competence. This is most clearly stated in the part of the opinion in which the majority allowed the use of essays in which the applicant discussed how race affected their life but added that any judgment “must be tied to that student’s unique ability to contribute to the university.” The affirmative action program of the two universities was described as “placing some into more competitive institutions than they otherwise would have attended.” Consequently, Blacks and Hispanics were admitted who were “less likely to succeed academically relative to their peers.” It seems from these statements that justices believed not only that equity in admission required that more qualified applicants are preferred over less qualified, but also that the Black and Hispanic applicants who were chosen at these schools were less qualified than the whites who were rejected.
Distributive justice – equality: The justices saw the affirmative action program as violating the norm that all citizens regardless of race and ethnicity are equal and should be afforded the same opportunities. In their view, any consideration of race led to inequality in opportunity. In other words, Black and Hispanic applicants had a better chance of being admitted than white or Asian applicants.
Distributive justice -need: Justice Thomas stated that affirmative action is unfair because it is biased against the neediest minority applicants. In his words, “these programs are overinclusive, providing the same admissions bump to a wealthy black applicant given every advantage in life as to a black applicant from a poor family with seemingly insurmountable barriers to overcome. In doing so, the programs may wind up helping the most well-off members of minority races without meaningfully assisting those who struggle with real hardship.”
Procedural justice: Affirmative action was also perceived as violating procedural justice to the extent that it was an arbitrary act of bureaucrats with no right of appeal or redress. At Harvard and UNC readers can and do consider race in both the initial and final readings to ensure that the rate of admission of minorities did not fall from previous years. Chief Justice Roberts said that this was unfair because it was not color-blind and further argued that “Eliminating racial discrimination means eliminating all of it.” Moreover, the justices said that the admission procedures “lack sufficiently focused and measurable objectives warranting the use of race” and that the universities had failed to show that increasing diversity improved the educational experience of students. Thus, the admission process violated procedural justice because it involved consideration of an invalid and biased factor – race – in pursuit of an unproven goal – diversity.
Interpersonal justice: In the majority opinion, affirmative action violates interpersonal justice by degrading the dignity of applicants. “When a university admits students ‘on the basis of race, it engages in the offensive and demeaning assumption that [students] of a particular race, because of their race, think alike.” The majority said that an affirmative action program could benefit or harm minority applicants but in either case it demeaned them by assuming that all members of the racial or ethnic category were the same. Thus, some Asian American applicants were stereotyped as lacking social skills and were harmed. Black applicants benefitted because they were seen as bringing something to the university that a white applicant could not, but the stereotyping involved in this preferential treatment was offensive and demeaning. The beneficiaries are “stamped with a badge of inferiority” that harms them in their subsequent careers.
Informational justice: In the view of the majority opinion, the universities did not adequately justify their affirmative action programs. Clarence Thomas argued in his opinion that those who use race as a factor must show “an actual link between racial discrimination and educational benefits” but that the defendants in this case had not been able “to establish a compelling reason to racially discriminate.” The research that had been offered as justification was described by Thomas as “tenuous, unspecific, and stereotypical.” He argued that universities must provide measurable goals and then “articulate a meaningful connection between the measure they employ and the goals they pursue.” He accused the universities of “self-proclaimed righteousness” in their unwillingness to justify what they had done in the affirmative action program. Roberts in his opinion stated that the only response of the universities to requests for information was to say “trust us.”
Retributive justice: Justice can be restored according to a retributive justice norm when those guilty of injustice are punished. When punishment is inflicted on those who bear no responsibility, however, it is unfair. Thomas argues that the affirmative action at the two universities was punishment for past discrimination, but those punished were not guilty. “Whatever their skin color, today’s youth simply are not responsible for instituting the segregation of the 20th century, and they do not shoulder the moral debts of their ancestors. Our Nation should not punish today’s youth for the sins of the past.”
Restorative justice: Affirmative action aims to restore relations between racial and ethnic groups and to the extent that it accomplishes this, the program can be judged fair and just. But according to Justice Thomas, affirmative action at the two universities is not likely to restore race relations between white and minority people. To the contrary, they prolong racial discrimination and assign collective guilt to all white people, and in doing so they spawn interracial conflict and worsen race relations.
A Little Bit of Framing Makes the AA Go Down?
Although they did not state this in their opinions, the majority justices seem to point to all the norms for judging fairness in their decision against university affirmative action. Based on national polls, a substantial portion of the public appears to share their views. So, what can be done to ensure the perceived fairness of efforts to remedy past and present discrimination? Some social science research has focused on how to frame the program. In the American Psychological Association’s Dictionary of Psychology, framing is “the process of defining the context or issues surrounding a question, problem, or event in a way that serves to influence how the context or issues are perceived and evaluated.” Given the ambivalent feelings associated with affirmative action programs, whether people perceive them as fair or unfair seems quite vulnerable to how they are presented.
One type of framing is to call an affirmative action program something other than affirmative action. Research shows that most people believe something should be done to eliminate discrimination and redress the problems created by racism. Nonetheless, the mention of “affirmative action” can evoke opposition and the false impression that there is a preferential policy and the use of quotas even when there are none (see Unzueta et al, 2008). Four types of affirmative action programs are distinguishable. The approach that is most favorably received is opportunity enhancement (also see Haley & Sidanius, 2006). Here no demographic group is targeted, and the program instead focuses on broadening the applicant pool. The emphasis is typically on recruitment, compensatory training, or outreach to the minority community. Equal opportunity means the elimination of discrimination and consists of prohibitions against demographic bias in selection and other personnel decisions. The least favorably received are programs that give preference to minorities. With the tiebreak or weak preferential treatment option, preference is given to unrepresented minority groups only if their qualifications are otherwise equivalent. In the strong preferential treatment option, preference is given to an unrepresented minority group even if their qualifications are lower than the nonminority.
Another approach to framing is to stress the benefits of the affirmative action program (see White, Charles & Nelson, 2008). In a corporation this may mean reminding employees of the importance of diversifying the customer base. In a university it could mean emphasizing the importance of increasing the diversity of the student body to enhance the educational experience. The caveat here is that there is little research in either workplaces or universities on the effects of affirmative action on performance in achieving organizational goals. There is somewhat more evidence that demographic diversity (i.e., racial, ethnic, sexual) benefits performance on some tasks, but contrary to some of the arguments in the dissenting opinions, the findings are far from clear. A stronger case can be made that diversity is most beneficial to organizational performance when there is also an overall climate that promotes not only racial and gender diversity but also what has come to be known as “deep diversity” in the form of differences that are not as visible such as skills, values, religion, and attitudes (Holmes et al, 2021). Nonetheless, contrary to what Justice Thomas said in his concurring opinion, there is no evidence that affirmative action necessarily lowers the productivity of a workplace or the academic excellence of a university. The lack of detrimental effects, combined with potential benefits such as the enrichment of the educational experience through diversity, can provide strong selling points in the framing of an affirmative action program. An even better approach would be to stress the intrinsic value of diversity in a democratic society rather than attempting to make it a means to some end. As Justice Sotomayor states in her dissenting opinion, “Diversity is now a fundamental American value, housed in our varied and multicultural American community that only continues to grow.” In other words, diversity should be the primary goal rather than merely a means to other goals.
A variety of other effects of framing have been found to influence perceptions of affirmative action. In a survey, asking respondents to indicate their approval of affirmative action for women before they are asked about racial minorities has been shown to increase acceptance of affirmative action for racial minorities (Wilson, Moore, McKay & Avery, 2008). White respondents are less likely to reject racial quotas in college admissions when the arguments against are framed as giving benefits to African Americans that they do not deserve than when they are framed as reverse discrimination against whites (Fine 1992). Students who are given the proportion of African Americans and white persons likely to be accepted and rejected for college admission perceive race-conscious affirmative action programs as fairer than students given the numbers of expected accepts and rejects (Friedrich, Lucas & Hodell, 2005). There is also evidence that presenting an affirmative action program in positive terms led to more acceptance of the program than presenting in negative terms (Gamliel, 2007). In other words, when told there was only one vacancy and one of the two applicants had to be rejected, they were less accepting of an affirmative action program than when told there was one vacancy and one of the two applicants had to be accepted.
References
Brockner, J. & Wiesenfeld, B.M. (1996). An integrative framework for explaining reactions to decisions: Interactive effects of outcomes and procedures. Psychological Bulletin, 120, 189– 208.
Cohen, R. L. (2016). Restorative justice. In C. Sabbagh & M. Schmitt (Eds.), Handbook of social justice theory and research (pp. 257–272). New York: Springer.
Colquitt, J. (2001). On the dimensionality of organizational justice: A construct validation of a measure. Journal of Applied Psychology, 86(3), 386-400.
Deutsch, M. (1975). Equity, equality, and need: What determines which value will be used as the basis for distributive justice. Journal of Social Issues, 31, 137-150.
Feldman, L. (2012). Germany’s Policy of Reconciliation: From Enmity to Amity. Lanham, Maryland: Rowman & Littlefield.
Fine, T. S. (1992). The impact of issue framing on public opinion: Toward affirmative action programs. The Social Science Journal, 29(3), 323–334.
Friedrich, J., Lucas, G. & Hodell, E. (2005). Proportional reasoning, framing effects, and affirmative action: Is six of one really half a dozen of the other in university admissions? Organizational Behavior and Human Decision Processes, 98, 195-215.
Gamliel, E. (2007). To accept or to reject: The effect of framing on attitudes toward affirmative action. Journal of Applied Social Psychology, 37(4), 683-702.
Haley, H., & Sidanius, J. (2006). The positive and negative framing of affirmative action: A group dominance perspective. Personality and Social Psychology Bulletin, 32(5), 656-668.
Hawley, J. (2017). How many people actually feel ‘White guilt’? The American Conservative.
Rawls, J. (1999). A Theory of Justice (revised ed). Cambridge, MA: Harvard University Press.
Swim, J. K., & Miller, D. L. (1999). White guilt: Its antecedents and consequences for attitudes toward affirmative action. Personality and Social Psychology Bulletin, 25, 500 – 515.
Uluğ, Ö. M., & Tropp, L. R. (2021). Witnessing racial discrimination shapes collective action for racial justice: Enhancing awareness of privilege among advantaged groups. Journal of Applied Social Psychology, 51(3), 248–261.
Unzueta, M. M., Lowery, B. S., & Knowles, E. D. (2008). How believing in affirmative action quotas protects white men’s self-esteem. Organizational Behavior and Human Decision Processes, 105, 1–13.
Van Vianen, A. E. M., Taris, R., Scholten, E. & Schinkel, S. (2004). Perceived fairness in personnel section: Determinants and outcomes in different of the assessment center. International Journal of Selection and Assessment, 12(1/2), 149-159.
Wenzel, M., & Okimoto, T. G. (2016). Retributive justice. In C. Sabbagh & M. Schmitt (Eds.), Handbook of social justice theory and research (pp. 237–256). New York: Springer.
White, F. A., Charles, M. A., & Nelson, J. K. (2008). The role of persuasive arguments in changing affirmative action attitudes and expressed behavior in higher education. Journal of Applied Psychology, 93(6), 1271-1286.
Wilson, D. C., Moore, D. M., McKay, P. F., & Avery, D. R. (2008). Affirmative action programs for women and minorities. Public Opinion Quarterly, 72(3), 514-522.
Menu: