The Promise of Diversity
December 1, 2000
With Responses From
Dec 1, 2000
8 Min read time
Test score legitimate existing class and race inequality.
Sturm and Guinier argue that affirmative action merely nibbles at the margins of inherently unfair structures of opportunity because the primary methods for assessing applicants for admission to higher education, entry-level jobs, and promotion are neither functional nor fair. As prediction mechanisms, standardized tests are pretty ineffective in reading the future performance of prospective students or employees. And, of course, they tell us nothing about drive, motivation, or the values an individual brings to her career and future contribution to society. As methods for ensuring fairness, standardized tests behave even worse: test scores are highly correlated with socioeconomic status and reflect the cumulative advantages of class rather than the inherent abilities of the candidate. As an alternative to testing, Sturm and Guinier propose a daring shift in our approach to selection: right now we are busy trying to assess the talents of an individual before they begin school or begin a job. But what if we were to simply reverse the order? Suppose we distribute theopportunity to perform through more direct and fair methods and determine merit of the candidate based on the quality of their actual performance in school or on the job itself?
We think this is a novel and potentially exciting way to widen the structure of opportunity. And for the most part, we agree with the authors’ critique of standardized testing. But does the robust critique of standardized testing offered by Sturm and Guinier also spell doom for the policy of affirmative action? Are these two issues necessarily joined at the hip? Or is it possible to decouple the issues–to reject standardized testing while recognizing the need for affirmative action policies?
We think it is wise to separate these two issues from one another. We doubt that the alternative vision proposed by the authors would, on its own, produce fair outcomes and create diversity in schools and workplaces. On the contrary, we believe that, in the absence of the legal framework of affirmative action, the Sturm and Guinier approach to selection would have a regressive impact on the very goals they aspire to achieve.
The argument that much standardized testing is of questionable value in predicting future success, and largely reflects the past cultural and economic advantages of the test-taker, is hardly new. But given the testing mania currently sweeping the nation, it is certainly worthy of broad dissemination. Focusing on the inherent flaws of standardized testing to assess the value of affirmative action is, however, a grave mistake. We agree with the premise that our society falls far short of a true meritocracy, but find it hard to agree with the premise that our society functions as a true testocracy either. How many jobs really use standardized paper-and-pencil tests as the key method for hiring? A recent study of a nationally selective sample of US firms found that only about 10 percent required an intelligence test to screen applicants for employment.1 Probably the most common employment test is the simple typing test, one form of testing that is, arguably, both functional and fair. But even this form of assessment is never used alone, and is typically paired with educational credentials (evidence of past performance) and a face-to-face interview to assess appearance, demeanor, attitude, and so forth.
In educational admissions standardized testing clearly plays a greater role. But even in undergraduate, law, and medical school admissions, tests are only one of a number of factors taken into account. The authors’ own example of University of Pennsylvania Law School admissions shows that only half of the law school class was chosen automatically based on LSAT and college grades. If tests really determined admission, we could fire all of our admissions committees tomorrow and have computers select incoming classes.
We think it is important to remember that standardized tests were instituted to increase fairness and open opportunities in the face of selection processes that operated via subjective assessments and closed networks bound by class, race, ethnicity, and gender. While tests are not perfectly fair, Sturm and Guinier vastly oversimplify their effects. Tests are a double-edged sword. The use of SATs made it possible for Jews educated in public schools to compete with moneyed applicants from prep schools for entry into elite universities.2 Standardized tests enable women police officers to legally challenge the decisions of their superiors when they repeatedly pass over the higher scoring females to promote male candidates.3 A survey of private firms in Los Angeles, Detroit, Chicago, and Boston in the late 1980s and early 1990s shows that when tests were used as important screening devices in hiring decisions, African American employment rates were higher than when interviews were used.4
Sturm and Guinier are correct to point out that test scores today more often legitimate existing class and race inequality, and that small and essentially meaningless differences in test scores are used to differentiate among candidates who are functionally identical in skill, talent, and ability. The authors propose to replace these flawed tests with different strategies for leveling the playing field. Take, for instance, the idea that we determine admission to our best colleges by taking the top 1 percent of graduates from every high school in the country and then doing a weighted lottery. The lottery would be weighted to ensure representation from major racial and ethnic groups and to ensure equal numbers of men and women.
What would be the result? Many youngsters from poor schools with weaker educational backgrounds would win admission to our top colleges, and perhaps some would not succeed. But such a group would most definitely include individuals from diverse racial and socioeconomic backgrounds who have demonstrated the drive, determination, and ambition to succeed. To be the valedictorian of the high school in Wellesley or Weston takes a great deal of effort and intelligence, but to be the valedictorian in Chelsea or East Boston takes all of that, plus an enormous determination to overcome considerable obstacles. Who will do better after four years in one of our best universities? The student who did what his parents, teachers, and neighbors supported and expected of him, or the student who succeeded despite overcrowded classrooms, indifferent adults, violent communities, chaotic learning environments, and the added pressure of after-school jobs or family responsibilities? As a selection method, this approach seems a superior way to tap into the drive, talent, and ambition of all our young citizens rather than those from already privileged homes.
Another problem is that gross categories like race and gender have a considerable heterogeneity within them. A great deal of evidence suggests that under current affirmative action programs recent immigrants and the children of immigrants are fulfilling many targets for racial diversity, leaving behind African Americans and Puerto Ricans. Affirmative action has also noticeably failed in taking into account class origins. Both working-class African Americans and working-class whites are woefully under-represented in our elite schools and workplaces.
Devising alternate methods for giving members of these disadvantaged groups the opportunity to acquire and demonstrate skills would begin to broaden the reach of affirmative action into these difficult-to-reach segments of society. In addition to changing requirements for college admissions, Sturm and Guinier suggest an alternate selection strategy for employment decisions. They propose that above a certain floor of minimum qualifications for a job there should be a weighted lottery that assigns applicants to existing openings. These new hires would then have a probationary period in which they would be able to grow into the job, learn new skills, and demonstrate the aptitudes and abilities to meet the demands of the position. Sturm and Guinier argue that this new method would open positions to individuals who would previously not make the cut based on standardized testing.
But it is foolish to ignore the historical reason so many groups fought to implement standardized testing in the first place. When managers, college admissions officers, and bosses assess and evaluate the skills, talents, and abilities of individuals, there is a powerful opening for the operation of conscious or unconscious biases rooted in the racial, ethnic, class, and gender identities of both parties. The plain fact is that there is no silver bullet that removes socially embedded preferences from the decisions of human beings. The Sturm-Guinier proposal moves the crucial hiring decision to a post-hire evaluation based on job performance during the period of probation. But what is to keep a male manager from deciding that a female trainee did not grow into the job as effectively as the male trainee who, in his view, seems more suited to the job? Sturm and Guinier imply that somehow work evaluation is inherently more objective than decisions based on interviews. Although they acknowledge that who does the evaluation matters, and state that it is important to have diversity in those doing the evaluations, they do not explain how, absent affirmative action, this diversity will be achieved. If the premise of the new model is that these probationary hires will be evaluated by diverse teams of managers, it is hard to see how this would be ensured, if not through current legal and cultural values of affirmative action.
The promise of the authors’ approach comes if it is seen not as a replacement for, but as an addition to, current affirmative action programs. Their proposal needs affirmative action in order to work because at its heart their proposal necessitates subjective assessments of work and education quality. Without diversity among managers, administrators, and teachers, the situation for women and minorities could actually be worse under the Sturm and Guinier proposal than it is now. Combining their proposal with the existing legal framework of affirmative action, however, might actually be better able to deliver on the original promise of affirmative action programs–increasing both opportunities for individuals and true diversity in organizations.
1 Peter V. Marsden, "Selection Methods in US Establishments," Acta Sociologica 37 (1994): 287-301.
2 David Karen, "Achievement and Ascription in Admission to an Elite College: A Political-Organizational Analysis," Sociological Forum 6 (1991): 349-380.
3 Kevin Cullen, "Breaking the Thin Blue Line of Bias: Policewoman is Likely Next Chief in Springfield," Boston Globe, 16 January, 1996, Metro/Region p. 15.
4 Philip Moss and Chris Tilly, "Soft Skills and Race," Work and Occupations 23 (1996): 252-276. See also Kathryn Neckerman and Joleen Kirshcenman, "Hiring Strategies, Racial Bias, and Inner-City Workers: An Investigation of Employers’ Hiring Decisions," Social Problems 38 (1991): 433-447.
While we have you...
...we need your help. Confronting the many challenges of COVID-19—from the medical to the economic, the social to the political—demands all the moral and deliberative clarity we can muster. In Thinking in a Pandemic, we’ve organized the latest arguments from doctors and epidemiologists, philosophers and economists, legal scholars and historians, activists and citizens, as they think not just through this moment but beyond it. While much remains uncertain, Boston Review’s responsibility to public reason is sure. That’s why you’ll never see a paywall or ads. It also means that we rely on you, our readers, for support. If you like what you read here, pledge your contribution to keep it free for everyone by making a tax-deductible donation.
December 01, 2000
8 Min read time