In another version of the same experiment, the White testers presented themselves as ex-felons (having served eighteen months for possessing cocaine with intent to sell) and were teamed up with Latinx and Black applicants with no criminal records. Whites with criminal records still had more callbacks or job offers (17.2 percent) than did Latinx testers (15.4 percent) and Black testers (13 percent) with no criminal records. Though the discriminatory outcomes were clear, “few interactions between our testers and employers revealed signs of racial animus or hostility toward minority applicants.”23 In the absence of prejudiced remarks, would rejected Black applicants even be aware that discrimination was operating without being able to compare their results to those of their White and Latinx teammates? Maybe not. But as these studies demonstrate, getting to the point of an interview is a higher hurdle for Black applicants than White ones, and in the case of the last study, a higher hurdle than for Latinx applicants as well.
Goal-oriented affirmative action can help address this problem. At each step of the process, the question is asked: is our pool of qualified candidates diverse, and if not, have we cast our net wide enough? In this approach, more of those résumés of Black candidates would likely have been at least considered for the next step in the evaluation process. In a goal-oriented process, once the qualified pool of applicants has been identified, those in the pool who move the organization closer to its diversity hiring goals are likely to be favored. This doesn’t mean that underrepresented candidates would always be the ones selected (the consistently lower rates of White unemployment let us know that White people are still being hired), but some of the time candidates of color would prevail. The White candidates who are not selected are likely to feel disappointed and might even believe that they were better candidates than the ones selected, but such perceptions are by definition subjective.
I am reminded of a dialogue I had with one of my White female students about affirmative action. In an essay on the topic she wrote, “I am in favor of affirmative action except when it comes to my jobs.” I wrote in response, “Which jobs have your name on them?” Of course she wanted to get the jobs she applied for and did not want to lose out to anyone, especially on the basis of race, a factor over which she had no control. Yet she seemed to assume that because she wanted them, they belonged to her. She assumed that she would, of course, be qualified for the job and would therefore be entitled to it. What was she assuming about the candidates of color? She did not seem to take into account the possibility that one of them might be as qualified, or more qualified, than she was. The idea that she as a White woman might herself be the beneficiary of affirmative action was apparently not part of her thinking.
We have all heard someone tell a story about a friend or relative who lost a coveted job opportunity because a “less-qualified” person took that spot, almost always reported to be a person of color, usually Black, not a White woman. I always wonder how the speaker knows so much about the selected candidate’s résumé or what happened in the final interview. Can we really say with confidence any particular hiring decision was not the best choice for the organization and its goals?
Despite the attempts to ensure a fair process, without the clarity of a clear set of institutional diversity goals to guide their decision-making, too often well-intentioned search committees find the “best” person is yet another member of the dominant group. What goes wrong? Some answers may be found in the research of social psychologists.
Aversive Racism, Uncomfortable Egalitarians, and Color-Blindness
Psychologists Samuel Gaertner and John Dovidio, along with numerous colleagues, have been studying race relations, and its relevance to our question, for more than forty years. They argue that White opposition to affirmative action programs is largely rooted in a subtle but pervasive form of racism they call “aversive racism.” Aversive racism is defined as “an attitudinal adaptation resulting from an assimilation of an egalitarian value system with prejudice and with racist beliefs.” In other words, most Americans have internalized the espoused cultural values of fairness and justice for all at the same time that they have been breathing the smog of racial biases and stereotypes pervading popular culture. “The existence, both of almost unavoidable racial biases and of the desire to be egalitarian and racially tolerant, forms the basis of the ambivalence that aversive racists experience.”24 The key to understanding this framework is to recognize that the internalization of the biases and stereotypes of popular culture and continued segregation from (and therefore lack of familiarity with) Blacks leave many Whites feeling uneasy, uncomfortable, even perhaps fearful in the presence of Black people, often without their conscious awareness of these feelings. Consequently, interracial interactions may generate discomfort and lead White people to avoid or withdraw from such situations, finding them “aversive.” The idea that they might be considered prejudiced by anyone (including themselves) is also an “aversive” idea, hence the name “aversive racism.”25
Pointing to the findings of several impressive research studies, Dovidio and Gaertner argue that because so-called aversive racists see themselves as nonprejudiced and racially tolerant, they generally do not behave in overtly racist ways. When the norms for appropriate, nondiscriminatory behavior are clear and unambiguous, they “do the right thing,” because to behave otherwise would threaten the nonprejudiced self-image they hold. However, in situations when it is not clear what the “right thing” is, or if an action can be justified on the basis of some factor other than race, racial bias will reveal itself. In these ambiguous situations, an aversive racist can discriminate against Blacks and still preserve the racially tolerant self-image.
For example, in a 1989 study of hiring decisions, Dovidio and Gaertner asked White college students to review application materials and evaluate candidates for a peer-counseling position on campus. The application materials were arranged in three categories—those of a highly qualified candidate, those of a moderately qualified candidate, and those of a weakly qualified candidate—with half of the materials in each category identified as those of a White applicant and half identified as those of a Black candidate. When the decision was unambiguous (as in “hire the strong candidate”), there was no selection bias. Both White and Black “strongly qualified” candidates received positive recommendations from the students. Similarly, when the candidates were “weakly qualified,” there was no discrimination. The choice was unambiguous—“reject weakly qualified” applicants. However, when the candidates were “moderately qualified,” the decision to recommend or reject was less clear. In that instance, “moderately qualified” White candidates were recommended significantly more often than the “moderately qualified” Black candidates, even though the credentials were the same. Seventy-five percent of the moderately qualified Whites were recommended, compared to only 50 percent of the moderately qualified Black candidates. In 1999, ten years later, Dovidio and Gaertner repeated the same experiment with a new group of White undergraduate students and found almost identical results. Seventy-seven percent of the moderately qualified Whites were recommended compared to 40 percent of the Black candidates in that same category. As part of these studies, they also measured the students’ overt expressions of prejudice using a self-reported prejudice scale. The students in 1999 expressed less prejudice on the survey than the students from 1989, but the pro-White bias in their decision-making was virtually the same.26
In a subsequent study, participants were asked to help make admissions decisions for the university. Again, Black and White applicants were matched and described as either uniformly strong (high SAT scores and high grades), uniformly weak (low SAT scores and low grades), or unevenly strong (either grades were high or SAT scores were high, but not both). When applicants were clearly strong or clearly weak, there was no anti-Black bias. The right decision—to admit or reject—was unambiguous. However, when applicants were strong in one area but not both, differential treatment emerged. Black candidates were mo
re likely to be rejected based on the weak area (either weak grades or weak SATs), minimizing the strength in the other area, but White candidates were more likely to be accepted based on the area of strength, minimizing the weakness in the other area. In other words, the participants systematically changed how they weighed the criteria to justify their decisions on the basis of race. Unevenly qualified White candidates were given the benefit of the doubt in a way that Black candidates were not. Summarizing the variety of studies they conducted, Dovidio and Gaertner concluded, “Aversive racism—racism among people who are good and well-intentioned—can produce disparate outcomes.… Although the bias of aversive racists may be subtle and unintentional, its consequences may ultimately be as severe as old-fashioned racism—threats to the well-being of blacks and the restriction of opportunities.”27
The foundational work that Dovidio and Gaertner did in the twentieth century has been expanded upon by Mahzarin Banaji and Anthony Greenwald using twenty-first-century technology. With advances in our understanding of human cognition, psychologists now agree that much of human judgment and behavior is produced with little conscious thought. Our internalized stereotypes and biases are not always consciously known to us, but they can still influence our behavior. The implicit-association test (IAT) developed by Banaji and Greenwald was designed to measure the strength of associations between concepts (e.g., black people, gay people) and evaluations (e.g., good, bad) or stereotypes (e.g., athletic, clumsy) by tapping into thoughts and feelings outside of conscious awareness and control.28
As the researchers describe on the Project Implicit website, “The main idea is that making a response is easier when closely related items share the same response key. When doing an IAT you are asked to quickly sort words into categories” using computer keys. Since reaction times are recorded in milliseconds, responses are not subject to the kind of conscious control one might use when responding to a survey question. By timing their reactions, the researchers can see the pattern of associations that participants are making. (If you want to try out the Race IAT, you can do so by going to the website. It takes about ten minutes.) Most who take the Race IAT are faster at linking racial White images to pleasant words than at linking racial Black images to pleasant words. This pattern is described as showing “automatic preference for White relative to Black.”29 Does an “automatic White preference” on the Race IAT mean a person harbors deep-seated prejudices? Not necessarily. Banaji and Greenwald write that nothing about the IAT suggests that it taps into the kind of dislike, disrespect, or even hatred that we associate with strongly held prejudices. But now that millions of people have taken the test and many research studies have been done using the methodology, they have reached two important conclusions.
First, we now know that automatic White preference is pervasive in American society—almost 75 percent of those who take the Race IAT on the Internet or in laboratory studies reveal automatic White preference. This is a surprisingly high figure.…
Second, the automatic White preference expressed on the Race IAT is now established as signaling discriminatory behavior. It predicts discriminatory behavior even among research participants who earnestly (and, we believe, honestly) espouse egalitarian beliefs. That last statement may sound like a self-contradiction, but it’s an empirical truth. Among research participants who describe themselves as racially egalitarian, the Race IAT has been shown, reliably and repeatedly, to predict discriminatory behavior that was observed in the research.30
For example, it predicts results like the ones Dovidio and Gaertner found—in a simulated hiring situation, White applicants were judged more favorably than equally qualified Black applicants. It also predicts doctors’ differential treatment of Black and White patients—emergency room and resident physicians recommend the optimal treatment, thrombolytic (blood-clot-dissolving) therapy, less often for a Black patient than for a White one with the same acute cardiac symptoms.31 (The Institute of Medicine has concluded that racial and ethnic minorities receive less-effective care even when income levels and insurance benefits are the same, pointing to implicit bias as the cause.)32
Of the more than 1.5 million White Americans who have taken the Race IAT on the internet, approximately 40 percent regard themselves as egalitarian and still show the automatic White preference in their response times. This combination of egalitarian attitudes and unconscious bias is similar to what Dovidio and Gaertner described in their definition of aversive racism. Banaji and Greenwald agree with their theoretical understanding of the issue but suggest what might be described as a “kinder, gentler” label—“uncomfortable egalitarians.” They write,
We have some observations about these uncomfortable egalitarians. First, there are a lot of them.… Second, their differential behavior toward White and Black Americans can well be responsible for a substantial portion of the disadvantage experienced by Black Americans.… Third, and perhaps most needing explanation—is that uncomfortable egalitarians are extremely unlikely to notice that their differential behavior toward Whites and Blacks contributes in any way to the disadvantages experienced by Black Americans.… Uncomfortable egalitarians may be the prototypical “good people” who have hidden biases. They see themselves as helpful, but it turns out that their helpfulness is selective, caused in part by their discomfort in interracial interactions. Their discriminatory behavior consists of being selectively ready or able to help only or mostly those who are like them—in other words, those in the groups for which they have automatic preferences… unaware that their comfort and helpfulness in interactions with in-group members is not matched by similar levels of comfort and helpfulness toward out-group members.33
There is no malice in helping people like yourself—what Banaji and Greenwald call “in-group favoritism”—but as we saw in the Dovidio and Gaertner hiring simulations, it does mean that a lot more White people will get jobs ahead of equally qualified Blacks. At the conclusion of their book, Blindspot: Hidden Biases of Good People, Banaji and Greenwald review decades of social science research and come to this conclusion: “Black disadvantage exists.… The further conclusion—one demanded by the great weight of evidence—is indisputable. Some portion of Black disadvantage is attributable to the way people respond to Blacks just because they are Black.”34 Social science research is also conclusive that, while explicit bias is infrequent, implicit bias (automatic race preference) is pervasive and contributes to the racial discrimination against Black Americans.35
The next question many people ask is, “What can I do about an implicit preference that I don’t want?” Banaji, Greenwald, and their colleagues offer this advice on their Project Implicit website:
It is well-established that implicit preferences can predict behavior. But, there is not yet enough research to say for sure that implicit biases can be reduced, let alone eliminated.… Therefore, we encourage people not to focus on strategies for reducing bias, but to focus instead on strategies that deny implicit biases the chance to operate. One such strategy is ensuring that implicit biases don’t leak out in the first place. To do that, you can “blind” yourself from learning a person’s gender, race, etc. when you’re making a decision about them (e.g., having their name removed from the top of a resume). If you only evaluate a person on the things that matter for a decision, then you can’t be swayed by demographic factors. Another strategy is to try to compensate for your implicit preferences. For example, if you have an implicit preference for young people you can try to be friendlier toward elderly people. Although it has not been well-studied, based on what we know about how biases form we also recommend that people consider what gets into their minds in the first place. This might mean, for example, going out of our way to watch television programs and movies that portray women and minority group members in positive or counter-stereotypical ways.36
There is some evidence that repeated exposure to positive counterstereotypic images can be useful, but it is truly difficult to disrupt the stereotypes we learn e
arly and often in our lives.
Color-Blind Racial Ideology
Despite so much evidence that people are not color-blind even when they want to be, color-blind racial ideology has become commonplace among Whites in the United States, particularly in the twenty-first century. In fact, what psychologists Dovidio and Gaertner called “aversive racism,” Eduardo Bonilla-Silva, looking through a sociologist’s structural lens, has called “color-blind racism.” He describes color-blind racism as the dominant racial ideology of contemporary America, in which White people deny or minimize the degree of racial inequality or explain contemporary racial inequality as the result of factors unrelated to racial dynamics (such as Black cultural values or economic forces unrelated to race.)37
Color-blind racial ideology can be expressed in multiple ways. One is what Ruth Frankenberg calls “color evasion”—as when someone says, “I don’t see color; we are all the same,” for example. This emphasis on sameness is a way of denying or rejecting the idea of White racial superiority. In theory, this sounds good, but it overlooks the fact that people of color are not having the same experiences as White people. As the research discussed earlier demonstrates, their racial group membership is impacting their daily lives. Another expression of color blindness is what Frankenberg calls “power evasion,” as when someone minimizes the impact of racism, claiming that everyone has the same opportunities to succeed and those who don’t have only themselves to blame.38 In The Myth of Racial Color Blindness, the editors, Helen Neville, Miguel Gallardo, and Derald Wing Sue, explain, “To deny race and ignore the existence of racism actually causes harm to people of color because it a) falsely perpetuates the myth of equal access and opportunity, b) blames people of color for their lot in life, and c) allows Whites to live their lives in ignorance, naiveté, and innocence.”39
Why Are All the Black Kids Sitting Together in the Cafeteria? Page 25
