AuthorSaiger, Aaron

Introduction 1004 I. Some Points of Agreement 1008 II. Competing Frames for the Exam School 1010 A. Exam Schools as State-Sponsored Contests 1011 B. Exam Schools as "Meritocratic" Prizes 1013 i. Public Resources 1018 ii. Peer Effects 1022 iii. Signaling and Status 1025 C. Exam Schools as Sites of Differentiated Instruction 1027 D. Exam Schools as Pork 1029 III. Second-Order Diversity 1031 A. First- and Second-Order Diversity 1031 B. The Case of New York City 1034 Conclusion 1040 INTRODUCTION

A public school admits students based, at least in substantial part, on an entrance exam. The student body that results looks, with respect to race, entirely different than the district it serves. Res ipsa loquitur, many say: Behold a racist admissions policy. Admissions schemes that yield titanically disparate racial impacts are intolerable. (1) Certainly they are unacceptable in a public school--managed by the state, staffed by government employees, and funded by tax dollars.

In several high-profile cases across the country, this sort of reasoning has led to efforts to eliminate public school admissions tests--and then to backlash. In 2021, school boards in both Boston and Fairfax County, Virginia, dropped exam scores from the vector of requirements for admission to their prestigious Boston Latin School and Thomas Jefferson School for Science and Technology (respectively). Both decisions are being litigated in federal court. (2)

The same year, the San Francisco school board embarked upon what proved to be an ephemeral experiment at its famed Lowell High School. The board, citing racial equity, replaced Lowell's longstanding academic admissions criteria with a lottery system. This led, in rapid succession, to litigation, the bitter and very public resignation of Lowell's principal, the successful (and even more bitter) recall of several school board members, and the reinstatement by the newly constituted school board of Lowell's original admission process. (3)

In New York City, meanwhile, exam-only admissions at the City's eight "exam schools"--three of which are required to use the exam by state statute--have been a perennial font of controversy. (4) Two of those three schools, Stuyvesant High School and the Bronx High School of Science, have in recent years admitted classes where between one and three percent of students are Black, in a city where one in four public school students is Black. (5) Advocates filed a high-profile complaint with the federal Department of Education, (6) and New York City's then-Mayor Bill de Blasio aggressively took on Albany and some angry citizens in an effort to kill the test. (7) Both efforts failed. Test-only admissions continue at Stuyvesant and its sister New York schools, (8) with the (hedged) support of de Blasio's successor (9) and the chastened acquiescence of other local elected officials. (10)

Educational regulators in New York have simultaneously undertaken a "sweeping" repudiation of the de Blasio push to abolish the use of any academic measures for admission across the public school system; many City schools other than the exam schools had used not a single examination but a range of academic criteria to make their admissions decisions. (11) Such academic "screens" have been restored to their central role in those schools' admissions practices as well. (12)

The litigation in Boston and Virginia, the electoral recall and policy reversal in San Francisco, and the ongoing battles in New York suggest that the thing does not speak for itself. Test-based admissions (again, using "test-based" also to describe admissions schemes that use tests in conjunction with other academic measures) have found considerable political and intellectual support. Efforts to do away with them have met with stalwart and often effective resistance, including high profile political campaigns. As in the higher-education cases now before the Supreme Court, some Asian Americans have taken a leading role in wondering why systems that benefit their success should be understood to be racial problems, rather than racial victories. (13) The tests, they and others argue, are the same for everyone. They benefit the diligent and the brilliant. And aren't grit, intelligence, and accomplishment--often lumped together as "merit"--things that we, as a society, need, value, and want to encourage? (14)

For social and educational critics who view the exam schools through the prism of racism and anti-racism--among whom one finds many legal advocates who view the issue exclusively in terms of federal constitutional and civil rights law--the answer to this question is that the answer does not much matter. Test-based admission is formally neutral--it is not a facially racial criterion--but creates deeply disparate impacts. (15) There can be no claim that the state has a compelling interest in sponsoring elite public schools; plenty of jurisdictions get along fine without them. (16) So the only legal question is the perennial one of how to analyze disparate impact. If the law of equality should reject practices that have disparate impacts without compelling justification, test-based admissions ought to fall. If it should confine itself to blocking only practices that impose facial racial criteria or that are animated by racial animus, the schools are lawful.

This analysis is inadequate because the framing of the question is itself inadequate. The test schools do not pose only questions under federal civil rights law. States bear a positive duty to educate all children. (17) They have concomitant leeway to structure the schools they provide--a massive government expenditure (18)--in line with their citizens' preferences. (19) No such system can require zero disparate impact, and certainly not one structured by our educational federalism. Therefore, neither the societal debate nor the legal debate over exam schools has been, or should be, only about disparate impact. It is also necessarily one about educational policy: Is the public preference for exam schools, as realized through the actions of elected officials, important enough, or plausible enough, to justify the disparate impact that it causes?

This debate is usually framed as a clash between racial fairness and educational "meritocracy." (20) This frame is too narrow. This Article seeks to expand it, in two ways. First, it argues that the reasonableness or desirability of academically selective schools does not depend only on whether it is proper to reward "merit," either at all or in the ways that admissions tests define it, given disparate impact. It suggests that, in addition to considering selective public school admissions as academic rewards, there are (at least) three other ways to think about them. They might be public contests, not primarily designed to reward. They might be a project of differentiated instruction, an effort to provide a learning environment to a particular kind of student that will best serve their educational needs. And they might be pork, one non-public good among many that politicians spread around favored constituencies for political reasons.

These possibilities lead to the Article's second argument: The normative appeal of test-based admissions policies depends not only on the policies themselves but upon the structure of the entire local school system in which the test school is embedded. A system that has special schools for strong test-takers but offers no corresponding kinds of programs for other kinds of students must meet a higher standard of justification than a system with programs that select for a wide range of needs and attributes. The analysis draws upon the concept of "second-order diversity," introduced by Heather Gerken in 2004, to develop these claims. (21)

Part I of the Article makes some brief introductory remarks regarding what is (or should be) a consensus about structural racism and geography-based school assignment. Part II then develops the framing argument, and Part III the argument about second-order diversity.


    Two propositions about the exam schools, and the racial unrepresentativeness of the populations they serve, are not controversial--or at least should not be, among reasonable people.

    First, everyone should agree that the racially skewed results of admissions by test are evidence of serious racial problems in American education. Reasonable people can disagree about what kind of problem the evidence suggests. Some will see evidence of bad tests, skewed notion of merit, and an obstinate willingness of society to double down on the consequences of structural racism. (22) Others, comfortable with the tests and the importance of what they measure, will see evidence of underlying social and structural disparities that lead to racial variability in academic performance and that need amelioration. (23) Controversy swirls over which of these two groups has the better account of the problem. But reasonable people cannot look at the results of admission by testing and see no problem at all.

    Second, school assignment by test is undeniably better policy than school assignment by geography--which is the modal method of student assignment in use in the United States. Every criticism of testing--that it does not predict success, that it does not measure merit, that it misunderstands merit, that it tests for the wrong things, that it is racially and culturally biased, that it magnifies the advantages of wealth--applies a fortiori to the geographical assignment of students to schools. To whatever extent testing is a poor measure of potential, merit, knowledge, or desert, geography is a worse measure. To whatever extent testing disadvantages particular racial or cultural groups, the educational costs that America's segregated housing system imposes upon such groups are orders of magnitude larger. (24) The wealth that allows some but not all to purchase test-preparation services or...

To continue reading

Request your trial

VLEX uses login cookies to provide you with a better browsing experience. If you click on 'Accept' or continue browsing this site we consider that you accept our cookie policy. ACCEPT