Listening to the debate on reforming law school admissions preferences

Denver University Law Review

Volumn 88, Issue 4, 2011, Pages 889-953

  Sander, Richard H ^a  

^a UNIVERSITY OF CALIFORNIA   (United States)

Author keywords

[No Author keywords available]

Indexed keywords

EID: 84866457610     PISSN: 08839409     EISSN: None     Source Type: Journal    
DOI: None     Document Type: Conference Paper

References (258)

1. Richard H. Sander, Class in American Legal Education, 88 DENV. U. L. REV. 631 (2011) [hereinafter CALE].
2. Hopwood v. Texas, 78 F.3d 932 (5th Cir. 1996).
3. For example, in the years since the Supreme Court's holding in Grutter v. Bollinger, 539 U.S. 306 (2003)
4. Law school racial preferences have become larger and even more mechanical than before.
5. Richard H. Sander, Why Strict Scrutiny Requires Transparency: The Practical Effects of Bakke, Gratz, and Grutter, in NEW DIRECTION IN JUDICIAL POLITICS (Kevin McGuire, ed., forthcoming 2012).
6. EUGENE ROBINSON, DISINTEGRATION: THE SPLINTERING OF BLACK AMERICA (2010).
7. Americans self-identifying as multiracial numbered 3.9 million in 2000, and 5.3 million in 2009, thus growing at four times the rate of the United States population as a whole, and growing faster in proportionate terms than any other racial group.
8. UNITED STATES BUREAU OF THE CENSUS, STATISTICAL ABSTRACT OF THE UNITED STATES 10 tbl.6 (2011).
9. ROBINSON, at 139-62.
10. These ideas were earlier developed by William Julius Wilson in THE TRULY DISADVANTAGED (1987), and are empirically investigated and elaborated
11. Yana Kucheva & Richard H. Sander, The Misunderstood Consequences of Shelley v. Kramer, (July 16, 2010) (unpublished manuscript) (on file with author).
12. ROBINSON, at at 208-14.
13. ELLIS COSE, THE RAGE OF A PRIVILEGED CLASS: WHY ARE MIDDLE-CLASS BLACKS ANGRY? (1993).
14. ELLIS COSE, THE END OF ANGER: A NEW GENERATION'S TAKE ON RACE AND RAGE (2011) [herinafter THE END OF ANGER].
15. Susan Saulny & Jacques Steinberg, On College Forms, a Question Of Race, or Races, Can Perplex, N.Y. TIMES, June 16, 2011, at A1.
16. David Leonhart, Top Colleges, Largely for the Elite, N.Y. TIMES, May 25, 2011, at B1.
17. Chris L. Jenkins, Theola Labbe-DeBose & Peyton M. Craighill, Class, Not Race, a Divider For Many in D.C., WASH. POST, Jun. 19, 2011, available at http://www.washingtonpost.com/local/dc-residents-see-class-not-race-as-citys- greatdivider/ 2011/06/17/AGZdU9bH-story.html.
18. See COSE, THE END OF ANGER 20, at 12.
19. MARGERY AUSTIN TURNER, MICHAEL E. FIX & RAYMOND J. STRUYK, URBAN INSTITUTE, OPPORTUNITIES DENIED, OPPORTUNITIES DIMINISHED: RACIAL DISCRIMINATION IN HIRING (1991).
20. June O'Neill, The Role of Human Capital in Earnings Differences Between Black and White Men, 4 J. ECON. PERSP. 25 (1990).
21. O'Neill observes, "The black-white hourly earnings ratio is 82.9 percent before adjusting for any characteristics ⋯ after adjusting for region, schooling and potential work experience ⋯ the ratio rises to 87.7 percent.
22. The addition of the AFQT raises the ratio to 95.5 percent, at which point close to three-quarters of the gap is explained. Adding actual work experience virtually closes the gap. at 40.
23. Note that perhaps the most in-depth study of racial discrimination patterns ever conducted- the National Housing Discrimination Study of 2000 conducted by The Urban Institute, found very low levels of net discrimination.
24. THE URBAN INSTITUTE, THE HOUSING DISCRIMINATION STUDY 2000 HDS (2000).
25. The "net" rate of discrimination experienced by blacks seeking to establish the availability of rental housing was about 4%; the net rate of "segregation" steering by real estate agents towards black testers was also about 4%. In other, more subtle or subjective aspects of housing search, net discrimination rates are higher, but the general pattern is one of astonishing progress, even relative to the prior national discrimination audit studies of 1989.
26. MARGERY AUSTIN TURNER, STEPHEN L. ROSS, GEORGE C. GALSTER & JOHN YINGER, THE URBAN INSTITUTE, DISCRIMINATION IN METROPOLITAN HOUSING MARKETS: NATIONAL RESULTS FROM PHASE I HDS 2000 3-2, 6-7 (2002).
27. Rob M.A. Nelissen & Marijn H.C. Meijersa, Social Benefits of Luxury Brands as Costly Signals of Wealth and Status, 32 J. EVO. & HUM. BEH. 343 (2011).
28. One defect in this study is that it was not "double-blind"-that is, the people seeking donations or survey participants presumably knew whether they were wearing a logo, and that may have influenced their behavior in subtle ways.
29. See Eli Wald, The Visibility of Socioeconomic Status and Class-Based Affirmative Action: A Reply to Professor Sander, 88 DENV. U. L. REV. 861, 865 (2011).
30. Timothy A. Judge & Daniel M. Cable, The Effect of Physical Height on Workplace Success and Income: Preliminary Test of a Theoretical Model, 89 J. APPLIED PSYCH. 428 (2004).
31. Note, however, that some commentators think that at least part of the difference in height outcomes is attributable to differences in cognitive ability, and others think that height effects are mediated by adolescent experiences related to height.
32. N. Persico, A. Postlewaite & D. Silverman, The Effect of Adolescent Experience on Labor Market Outcomes: The Case of Height, 112 J. POL. ECON. 1019 (2004)
33. Anne Case & Christina Paxson, Stature and Status: Height, Ability, and Labor Market Outcomes, 116 J. POL. ECON. 499 (2008). 37.
34. John Cawley, The Impact of Obesity on Wages, 39 J. HUM. RESOURCES 2 (2004)
35. Dalton Conley & Rebecca Glauber, Gender, Body Mass, and Socioeconomic Status: New Evidence from the PSID, 17 ADVANCES IN HEALTH ECON. & HEALTH SERVS. RES. 253 (2007).
36. Daniel S. Hammermesh & Jeff E. Biddle, Beauty and the Labor Market, 84 AMER. ECON. REV. 1174 (1994)
37. Markus Mobius & Tanya Rosenblat, Why Beauty Matters, 96 AMER. ECON. REV. 222 (2006).
38. See Mobius & Rosenblat 39, for experimental evidence on the causal role of physical attractiveness.
39. For one particularly relevant example, see Richard H. Sander, A Systemic Analysis of Affirmative Action in American Law Schools, 57 STAN. L. REV. 367, 463-64 (2004) [hereinafter Systematic Analysis], and accompanying text.
40. Richard H. Sander, The Racial Paradox of the Corporate Law Firm, 84 N.C. L. REV. 1755 (2006)
41. Sander, 456-68 (2004)
42. Linda Loury & David Garman, College Selectivity and Earnings, 13 J. LAB. ECON. 289 (1995).
43. Jane Yakowitz, Marooned: An Empirical Investigation of Law School Graduates Who Fail the Bar Exam, 60 J. LEG. EDUC. 3 (2010).
44. Differences in black and white incomes often appear much starker when one does not control for family composition, but it is important to do so, since the issue here is how human capital is rewarded in the labor market. The reasons for the very large proportion of unmarried blacks are undoubtedly complex, ranging from the high incarceration rates of black men to the attitudes of black women towards interracial marriage.
45. RALPH RICHARD BANKS, IS MARRIAGE FOR WHITE PEOPLE? HOW THE AFRICAN AMERICAN MARRIAGE DECLINE AFFECTS EVERYONE (2011).
46. Jay D. Teachman, Kathleen M. Paasch, Randal D. Day & Karen P. Carver, Poverty During Adolescence and Subsequent Educational Achievement, in CONSEQUENCES OF GROWING UP POOR 382-418 (Duncan & Brooks-Gunn, eds., 1997), and author's analysis of data from the National Educational Longitudinal Study.
47. Richard H. Sander & Jane Yakowitz, The Secret of My Success: How Status, Prestige, and School Performance Shape Legal Careers, J. EMP. LEGAL STUDIES (forthcoming 2011).
48. This article substantially elaborates on my earlier work on the interaction of law school grades and school eliteness on post-law-school earnings.
49. Peter Arcidiacono et al., What Happens After Enrollment? An Analysis of the Time Path of Racial Differences in GPA and Major Choice, (Duke University, Working Paper, 2011).
50. Richard H. Sander, A Reply to Critics, 57 STAN L. REV. 1963, 1972-73 (2005).
51. The data reported in Tables 2 and 3 show that the GPA improvement for black "second-choice" students closely mirrors the reduction in the credential gap between those students and others in the same tier.
52. Lawrence D. Bobo & Ryan A. Smith, From Jim Crow Racism to Laissez-Faire Racism: The Transformation of Racial Attitudes, in BEYOND PLURALISM: THE CONCEPTION OF GROUPS AND GROUP IDENTITIES IN AMERICA 199 (Katkin, Landsman & Tyree eds., 1995).
53. This article presents 1990 survey evidence showing that 56% of whites rated blacks as less intelligent than whites, and over 60% rated blacks as lazier than whites.
54. The survey method had respondents rate various racial groups on a series of scales, and then compared the average scalar ratings, rather than asking respondents to make direct comparisons. Bobo and Smith also note that whites have moved sharply away from "biological" explanations for racial differences and towards "cultural" explanations.
55. Richard O. Lempert, David L. Chambers & Terry K. Adams, "From the Trenches and Towers": Michigan's Minority Graduates in Practice: The River Runs Through Law School, 25 LAW & SOC. INQUIRY 395 (2000).
56. In their in-depth analysis of affirmative action at the University of Michigan Law School (UMLS), the authors nowhere mention the abysmal academic performance of UMLS blacks, 60% of whom had GPAs in the bottom tenth of their class during the period they studied. In The Shape of the River, Bowen and Bok to their credit do discuss the problem of poor academic performance of minorities in college.
57. But they misleadingly imply that low grades are mostly connected to some sort of racial underperformance, rather than the use of preferences by colleges.
58. Even more misleadingly, when they talk about minority performance they only give "average" class rank (the 23rd percentile for elite college blacks, according to them).
59. Given the highly skewed distribution of minority GPAs towards the bottom of the distribution, an average is very misleading-an "average" 23rd percentile probably translates to a 10th percentile median.
60. See WILLIAM G. BOWEN & DEREK BOK, THE SHAPE OF THE RIVER: LONG-TERM CONSEQUENCES OF CONSIDERING RACE IN COLLEGE AND UNIVERSITY ADMISSIONS 72 (1998).
61. Claude Steele & Joshua Aronson, Stereotype Threat and the Intellectual Test Performance of African Americans, 69 J. OF PERSONALITY AND SOC. PSYCH. 797 (1995).
62. I am a stereotype-threat skeptic, for three reasons.
63. First, most of research showing stereotype-threat research has been done in laboratories; tests in real-world settings have not produced comparable results.
64. Michael J. Cullen, Chaitra M. Hardison & Paul R. Sackett, Using SAT-Grade and Ability-Job Performance Relationships to Test Predictions Derived from Sterotype Threat Theory, 89 J. OF APPLIED PSYCH. 220 (2004).
65. Second, in law school at least, entering credentials accurately predict first-year GPA performance for minorities.
66. Research I have completed since Reply to Critics is even more compelling, demonstrating that blacks and whites have indistinguishable first-semester grades when we control for entering credentials and undergraduate college.
67. Third, again in law school, the black-white performance gap is the same or larger in legal writing classes as in timed exam classes, even though the former should not evoke the stereotype threat effect.
68. Some research has at least examined how self-perceptions about affirmative action affect stereotype threat, finding that students who believe they have received large preferences are more vulnerable to the threat effect.
69. Colette van Laar, Shana Levin & Stacey Sinclair, Social Identity and Personal Identity Stereotype Threat: The Case of Affirmative Action, 30 BASIC AND APPLIED SOC. PSYCH. 295 (2008).
70. Gary Orfield & Dean Whitla, Diversity and Legal Education: Student Experiences in Leading Law Schools, in GARY ORFIELD & MICHAL KURLAENDER, DIVERSITY CHALLENGED: EVIDENCE ON THE IMPACT OF AFFIRMATIVE ACTION 143 (2001).
71. Further evidence that the Orfield/Whitla survey was merely summoning up PC responses comes from an analysis done by the eminent sociologist Thomas Espenshade, who asked college students (in a survey that had no visible "diversity" agenda) to list their five closest friends; the survey administrators than coded the race of each identified friend. These results showed
72. No logical inconsistency in the number of interracial friendships
73. Far lower levels of interracial friendship than those produced by the Orfield-Whitla survey. Personal communication from Dr. Espenshade (March 2007).
74. See generally Kevin Deasy, Enabling Black Students to Realize Their Potential in Law School: A Psycho-Social Assessment of an Academic Support Program, 16 T. MARSHALL L. REV. 547 (1991)
75. Cathaleen A. Roach, A River Runs Through It: Tapping into the Informational Stream to Move Students From Isolation to Autonomy, 26 ARIZ. L. REV. 667 (1994).
76. Matt Moore, Ailing Method, Essential Motive: An Examination of Two Strategies to Improve Core Legal Learning Among Underrepresented Minority Law Students (2005)
77. Unpublished manuscript) (on file with author).
78. Grutter v. Bollinger, 539 U.S. 306, 327-33 (2003).
79. The General Social Survey website, found at http://www.norc.org/ GSS+Website/, has a wealth of information about the survey and its methodology.
80. Note, for example, that since I use only a single year of GSS data, the sample sizes for the black cells are measured only in the dozens.
81. Some of the literature cited in this section provides examples of more sophisticated methods that sociologists routinely use to reach more definitive conclusions about attitudinal patterns.
82. Charles R. Chandler & Yung-mei Tsai, Social Factors Influencing Immigration Attitudes: An Analysis of Data from the General Social Survey, 38 SOC. SCI. J. 177 (2001).
83. Lawrence Bobo & Frederick C. Licari, Education and Political Tolerance: Testing the Effects of cognitive Sophistication and Target Group Affect, 53 PUB OPINION Q. 285 (1989).
84. Even recreational activities vary more across working-class groups of different races than among middle-class groups of different races.
85. Floyd et al., Race, Class, and Leisure Activity Preferences: Marginality and Ethnicity Rrevisited, 26 J. LEISURE RES.158 (1994).
86. Sean-Shong Hwang, Kevin M. Fitzpatrick & David Helms, Class Differences in Racial Attitudes: A Divided Black America?, 41 SOC. PERSP. 367 (1998).
87. Jason E. Shelton & George Wilson, Race, Class, and the Basis of Group Alignment: An Analysis Of Support for Redistributive Policy Among Privileged Blacks, 52 SOC. PERSP. 385 (2009)
88. Zoltan Hajnal, Black Class Exceptionalism: Insights from Direct Democracy on the Race Versus Class Debate, 71 PUB OPINION Q. 560 (2007).
89. DUNCAN KENNEDY, LEGAL EDUCATION AND THE REPRODUCTION OF HIERARCHY: A POLEMIC AGAINST THE SYSTEM (2004)
90. RICHARD ABEL, AMERICAN LAWYERS 48-73 (1989).
91. Sander, Systemic Analysis 40.
92. On eschewing abolitionist positions
93. My comments in a January 2005 debate: "I want to keep reminding participants [in the mismatch debate] that one can make enormous progress in addressing this problem short of abolishing preferences." Legal Affairs Debate Club, January 14, 2005, available at http://www.legalaffairs.org/ webexclusive/debateclub-sander0105.msp.
94. On my advocacy of race-conscious strategies in addressing housing segregation, see generally Richard H. Sander, Individual Rights and Demographic Realities: The Problem of Fair Housing, 82 NW. U. L. REV. 874 (1988).
95. Regents of the Univ. of Cal. v. Bakke, 438 U.S. 265 (1978)
96. Grutter v. Bollinger, 539 U.S. 306 (2003).
97. Richard Sander, Why Strict Scrutiny Requires Transparency: The Practical Effects of Bakke, Gratz, and Grutter, in NEW DIRECTIONS IN JUDICIAL POLITICS (Kevin McGuire, ed., forthcoming 2012).
98. Sander, Systemic Analysis 40, at 390-410
99. Ian Ayres & Sydney Foster, Don't Tell, Don't Ask: Narrow Tailoring After Grutter and Gratz, 85 TEX. L. REV. 517 (2007)
100. Fisher v. Univ. of Texas, 631 F.3d 213, 249-51 (5th Cir. 2011)
101. Grutter, 539 U.S. at 326-28.
102. Vill. of Arlington Heights v. Metro. Hous. Dev. Corp., 429 U.S. 252 (1977).
103. David Segal, Law School Economics: Ka-Ching!, N.Y. TIMES, July 16, 2011, available at www.nytimes.com/2011/07/17/business/law-school-economics-job- market-weakens-tuitionrises. html
104. Elie Mystal, A Resigning Law Dean Spills the Beans on the Fleecing of Law Students, ABOVE THE LAW, July 29, 2011, available at http://abovethelaw.com/ 2011/07/a-law-dean-resignsand- spills-the-beans-on-how-his-university-has-been- taking-advantage-of-law-students/.
105. In the wake of Proposition 209, the University of California adopted innovative, largescale outreach efforts to improve the rate at which disadvantaged students could qualify for university admission.
106. KARL S. PISTER, UC OFFICE OF THE PRESIDENT, UC OUTREACH: SYSTEMWIDE PERSPECTIVE AND STRATEGIC Plan (1998) , available at http://www.ucop.edu/ ucophome/commserv/outreach/outpdf/outreach.pdf.
107. Within a few years of implementation, the number of blacks and Hispanics admitted to the UC undergraduate system exceeded the numbers achieved under the old, racial preference system.
108. It is plausible that improved outreach played a significant role
109. http://nces.ed.gov/surveys/nels88/.
110. Richard Lempert, Reflections on Class in American Legal Education, 88 DENV. U.L. REV. 683, 702 (2011).
111. High school students who take the SAT complete a questionnaire in their initial application to the College Board which asks them, among many other things, about their highest degree aspiration.
112. These answers are anonymized and combined with the results of their SAT I and SAT II examinations into large databases; researchers can request extracts of these data. My research team obtained a large extract originally created for the economists David Card and Alan Krueger, which Professor Card generously (and with College Board's permission) shared with us.
113. Pell Grants are federal scholarships available, roughly speaking, to students in the bottom half of the income distribution.
114. They are discussed in more detail at CALE 1, at 641 n.29.
115. CALE 2, at notes 28 & 30.
116. The Bar Passage Study ("BPS") was a longitudinal study of law students conducted by the Law School Admissions Council in the 1990s.
117. It tracked over twenty-seven thousand students from matriculation to law school in 1991 through their efforts to pass the bar in the mid-1990s.
118. All participants completed a detailed "entering student questionnaire", from which this socioeconomic data is drawn.
119. The survey covered roughly eighty percent of all law students at roughly ninety percent of all law schools, providing an excellent sample for the purposes discussed here.
120. It is certainly not subject to any of the selection bias problems identified by Dr. Lempert.
121. And the small gap is plausibly due to the slightly older cohort.
122. Lempert, at 690-91.
123. Angela Onwuachi-Willig & Amber Fricke, Class, Classes, and Classic Race-Baiting: What's in a Definition?, 88 DENV. U. L. REV. 807, 808-09 (2011).
124. Note that there are several occupational tiers of nurses; vocational nurses are assigned lower scores for occupational prestige than registered nurses, while nurse practitioners are assigned somewhat higher scores.
125. Author's analysis of data contained in: U.S. BUREAU OF THE CENSUS, AMERICAN COMMUNITY SURVEY (2009).
126. CALE, at 670 app.1.
127. Calculations by the author from the CALE data; original data available from the author.
128. As I have written many times, and repeat again in CALE (see page 666), the mismatch effect is not caused by racial preferences but by large preferences.
129. If law schools instituted SES preferences on the scale currently used for blacks and American Indians, this would have the same counterproductive effects on low-SES students that are currently experienced by those racial minorities.
130. But I think it quite unlikely that any school would ever extend widespread SES preferences on the scale of many current racial preferences.
131. This is, in large part, because one can achieve substantial SES diversity with small preferences (see the analysis in Part I, as well as the UCLA law school experiment, where the school achieved very high SES diversity with preferences that were, on average, one-fifth the size of its earlier racial preferences for blacks and American Indians.
132. The smaller preferences, as I tried to make clear in CALE, are an important contributor to the lower visibility of SES preference recipients.
133. For example, the General Social Survey asks respondents questions (e.g., about education, occupation and income) that allow us to assign them an SES score. It also asks them a couple of questions about how they would describe their own "class" (e.g., "working," "middle," "uppermiddle").
134. If we rank-order these self-assigned classes, they correlate poorly with the objective measure of SES (correlations are generally under .25). For the source data, see supra note 59.
135. The UCLA class of 2000 had a bar passage rate of 90% on the July 2000 California bar exam.
136. Its bar passage rate on the prior three July exams-the last three years of conventional racebased affirmative action-had averaged under 82%.
137. Sander & Yakowitz 45.
138. Richard H. Sander, Experimenting with Class-Based Affirmative Action, 47 J. LEGAL EDUC. 472, 496-97 (1997).
139. Average enrollments of minorities during the six years before Prop 209 went into effect were 9.7% black, 14.1% Hispanic, and 16.7% Asian.
140. This point is elaborated in Sander, at 492.
141. The enrollment figures can be independently verified in a number of sources, such as the admissions statistics compiled by the University of California Office of the President.
142. Under the objective system, UCLA measured four "household" and three "neighborhood" characteristics of each applicant, and used an algorithm to assess overall socioeconomic disadvantage.
143. Under the subjective system (which, in my judgment, was less effective), admissions officers read essays and looked at each individual's reported statistics, and developed an intuitive judgment about socioeconomic disadvantage.
144. Deirdre M. Bowen, Meeting Across the River: Why Affirmative Action Needs Race & Class Diversity, 88 DENV. U. L. REV. 751, 768 (2011).
145. Systemic Analysis 40, at 416. Using the Bar Passage Study data, I calculated the black-white gap as measured by a standard academic index; the gap was 170 points among the most elite schools; in the next four tiers, the gap was, respectively, 174 points, 202 points, 165 points, and 172 points.
146. Only in the sixth tier, of historically minority schools, was the gap significantly smaller.
147. Statement from Stephen T. Schreiber, Exec, Vice President, Law School Admissions Council (Jan. 14, 2010) (on file with author) available at http://www.saltlaw.org/userfiles.
148. Unfortunately, Johnson's inaccurate findings got wide play in a January issue of the New York Times.
149. See Tamar Levin, Law School Admissions Lag Among Minorities, N.Y. TIMES, Jan. 6, 2010, available at http://www.nytimes.com/2010/01/07/education/07law. html.
150. When I wrote to Tamar Lewin, the author of the Times article, pointing out the inaccuracies, she expressed "hope" that the Times "will do a better job" next time. No published correction was forthcoming.
151. Email from Tamar Lewin, New York Times Correspondent, to the author (February 2, 2010)
152. Systemic Analysis 40, at 409; Jane Yakowitz & Richard H. Sander, Lifting the Veil on Law School Admissions, (unpublished manuscript) (on file with author).
153. Much of Bowen's article seems to either miss the point of my arguments in CALE, or insinuates, without providing evidence, that I'm misleading readers. For example, she writes that my discussion of admissions at the University of Missouri at Columbia ("UMC") "indicts students of color
154. Sander goes into great detail offering the gradation of scores and the odds that a white student was admitted within a certain range, but does not offer the same data regarding students of color.
155. There are two misrepresentations here. The point of my discussion of UMC's admissions is not to "indict" black students, but to give the reader insight into both the mechanical use of race by law schools admissions officers and the focus of these officers on admitting those blacks with the highest scores, not the blacks who would most enhance "diversity" at the school.
156. I use fewer categories in discussing black admissions at UMC than white admissions, because there were only two relevant categories for blacks in the admissions cycle I discuss: 93% of blacks with an academic index above 44 (UMC's scale) were admitted, and 100% of blacks with an academic index below 44 were rejected (as noted in CALE, whites were rarely admitted with index scores below 58).
157. Bowen goes on, "The reader is left to wonder, how many black students applied? What were their scores? Were they all lower than the white students? Without providing this information, Sander gives an impression that all black students were mismatched or robbed more qualified white students of seats at UMC."
158. Bowen, at 781-82.
159. I hope most readers recognize that none of my work on affirmative action is concerned with whether white students are "robbed" admissions places, but rather with how well a school's diversity objectives, and the preferenced students themselves, are served by highly mechanical processes that focus on a single factor (race) and ignore the academic disparities that result.
160. Bowen (like any reader) is welcome to examine the data from UMC (or dozens of other datasets our research group has collected). The data about UMC offered in CALE is sufficient to make the point developed there.
161. As to Bowen's other concerns, here is some additional data: in the admissions cycle I examined, UMC had forty-one black applicants for UMC's class of about one hundred fifty.
162. Fourteen were admitted, and seven enrolled. The median black applicant had an index of 42.
163. Many white applicants had even lower scores, but these were all rejected; the median white applicant had an index of 60.
164. The blacks who ended up enrolling from this group had a median index of 54, about eight points lower than the class median and enough, I believe, to put them at serious risk of mismatch and bar failure.
165. Arin N. Reeves, Race as a Red Herring? The Logical Irrelevance of the Race vs. Class Debate, 88 DENV. U. L. REV. 835, 842 (2011).
166. This issue is thoroughly discussed and documented in JONATHAN ALTER, THE DEFINING MOMENT: FDR'S HUNDRED DAYS AND THE TRIUMPH OF HOPE (2007).
167. Helene Cooper, Black Voters' Support for Obama Is Steady and Strong, N.Y. TIMES, Oct. 26, 2011, available at http://www.nytimes.com/2011/10/27/us/ politics/obamas-support-amongblacks- remains-strong.html?pagewanted=all
168. Beyond issues, many African-Americans feel an emotional connection to Mr. Obama that seems unshakable, saying that nothing can compare with seeing someone who looks like them in the White House."
169. Tim Groseclose discusses in detail research that suggests that to the extent race factored into the 2008 presidential election, it was a net benefit for Obama. TIM GROSECLOSE, LEFT TURN: HOW LIBERAL MEDIA BIAS DISTORTS THE AMERICAN MIND 85-86 (2011).
170. Esther Duflo, Pascaline Dupas & Michael Kremer, Peer Effects, Teacher Incentives, and the Impact of Tracking: Evidence from a Randomized Evaluation in Kenya, 101 AM. ECON. REV. 1739 (2011).
171. Duflo won the 2010 John Bates Clark medal, now given annually to the American economist under the age of forty judged to have made the most significant contribution to economic thought and knowledge.
172. Peter Arcidiacono, Esteban M. Aucejoz, Hanming Fang, Kenneth I. Spenner, Does Affirmative Action Lead to Mismatch? A New Test and Evidence, QUANTITATIVE ECON. (forthcoming 2011)
173. Arcidiacono et al. 47.
174. Barnes's originally advanced her critique in Katherine Y. Barnes, Is Affirmative Action Responsible for the Achievement Gap Between Black and White Law Students?, 101 NW. U. L. REV. 1759 (2007).
175. Katherine Y. Barnes, Is Affirmative Action Responsible for the Achievement Gap Between Black and White Law Students? A Correction, A Lesson, and an Update, 105 NW. U. L. REV. 791 (2011).
176. Doug Williams, Richard Sander, Marc Luppino & Roger Bolus, Revisiting Law School Mismatch: A Comment on Barnes, 105 NW. U. L. REV. 813 (2011).
177. Doug Williams, Does Affirmative Action Create Educational Mismatches in Law Schools?, (Working Paper 2009), available at http://econ.duke.edu/~hf14/ ERID/Williams.pdf.
178. This paper was presented at the American Law and Economics annual meeting in 2010, and revised version in 2011.
179. Deborah C. Malamud, Class Privilege in Legal Education: A Response to Sander, 88 DENV. U. L. REV. 729, 730 n.2 (2011).
180. Her full language on this point follows: "In my own work, I have acknowledged that the diversity rationale pushes institutions towards a focus on their own institutional goals, rather than on the consequences of affirmative action for the lives of its intended beneficiaries [citation omitted].
181. I believe Sander has made it impossible (and rightfully so) to ignore those consequences-although I am more persuaded by his critics on the merits of the question of what those consequences are in fact."
182. An editor at Stanford Law Review told me of this decision; proposed essays by James Lindgren and William Henderson were among the more balanced responses that the journal bypassed to focus only on very critical pieces.
183. Bar Exam Passage Conference, held in Chicago in October 2008, and sponsored by the ABA and the Law School Admissions Council. It is described here: http://apps.americanbar.org/legaled/calendar/conferences/Bar%20Passage/ barpassageagenda.html.
184. U.S. COMMISSION ON CIVIL RIGHTS, AFFIRMATIVE ACTION IN AMERICAN LAW SCHOOLS: A BRIEFING BEFORE THE UNITED STATES COMMISSION ON CIVIL RIGHTS 143 (2006), available at http://www.usccr.gov/pubs/AALSreport.pdf.
185. It is striking, for example, that law students (often minority students) at Harvard, Stanford, New York University, Michigan, Northwestern, and Duke have organized forums or debates on law school mismatch issues.
186. So have nearly all of the major organizations of black lawyers.
187. But no law faculty at these schools (or any other "top 10" law school) has organized any kind of debate or forum exploring the evidence for and against the mismatch effect.
188. This paragraph summarizes an argument laid out in Systemic Analysis 40, at 425-54
189. Which is revisited and elaborated upon in Sander 48, 57 STAN. L. REV. 1963, 1966-78 (2005).
190. The only part of this logic disputed by the critics, to my knowledge, is the first statement- that black grades in law school show little evidence of underperformance.
191. But my claim is backed not only by my own analyses, but by a series of authoritative LSAC studies (which again, to my knowledge, no one has disputed), showing very small levels of black underperformance in law school.
192. My own research suggests that when such studies control for undergraduate college attended by students, even this slight underperformance goes away.
193. Ian Ayres & Richard Brooks, Does Affirmative Action Reduce the Number of Black Lawyers? 57 STAN. L. REV. 1807, 1838-40 (2005).
194. Ayres and Brooks cite no actual evidence of stereotype threat in law schools and concede that "we do not have a compelling theory as to what is causing" lower rates of black achievement on the bar.
195. In their piece, they simply sidestep the logic of the argument summarized here. With no sense of self-irony, Ayres and Brooks suggest that the historically black law schools are good places to see where things are going right for black students.
196. Of course, it is at these schools that mismatch is minimized, and it is these schools that Ayres and Brooks inexplicably omit from one of their central mismatch analyses!
197. Ayres and Brooks suggest that perhaps stereotype threat affects every intellectual task undertaken by law students; but this proves too much, since it implies that black law students will go on to become black lawyers whose performance is hindered throughout their careers by stereotype threat.
198. Moreover, this entire explanation seems to hinge on substantial black underperformance in law school grades, for which there is no evidence.
199. Christian DuBois, "Too Old for Law School?" (2005 working paper on file with author).
200. The UCLA data for this class comes from a spreadsheet prepared by the UCLA Records Office, and available from the author; data from Southwestern comes from the ABA-LSAC Official Guide to ABA-Approved Law Schoools, 2004 edition (2003) at 663 (for Southwestern student credentials), and State Bar of California, General Statistics Report, July 2005 California Bar Examination (2006).
201. Systemic Analysis 40, at 444.
202. Sander 48, at 1972
203. discussing the unobserved credentials problem, which will make bar passage performance at elite schools appear stronger).
204. As required under the state's public record laws). This analysis is based on the George Mason Disclosed Database for 1995-2007 and is available from the author.
205. AMER. BAR ASS'N, ABA-LSAC OFFICIAL GUIDE TO ABA-APPROVED LAW SCHOOLS (2002-2005 editions).
206. I averaged Howard data on attrition and New York bar passage over four years.
207. For the reasons discussed supra, text accompanying note 138, this is almost surely an underestimate of the success rate of Howard students with credentials at the class "median".
208. 539 U.S. 306 (2003).
209. Lempert, Chambers & Adams 50, at 395 (suggesting minority bar passage rate varies from 95% to 98%).
210. Grutter v. Bollinger, 137 F. Supp. 2d 821, 862-63 (E.D. Mich. 2001) ("almost all" minority graduates pass a bar exam).
211. U.S. COMMISSION ON CIVIL RIGHTS 127, at 9, 52
212. The BPS has, on the one hand, wonderfully detailed data on individual students, including detailed surveys completed at various points during the student's progression through law school.
213. On the other hand, as discussed further below, the BPS obscures vital data for comparative analyses: schools are grouped into six broad 'clusters'-it is not even possible to know for sure which schools are put in which cluster-and state bars are grouped into broad geographic areas.
214. The UMLS Bar data was supplied by Tim Raubinger, Assistant Secretary, Michigan Board of Law Examiners, in the fall of 2005 (on file with the author).
215. Richard Sander, Do Elite Schools Avoid the Mismatch Effect?, EMPIRICAL LEGAL STUDIES (Sept. 22, 2006), http://www.elsblog.org/the-empirical-legal- studi/2006/09/do-elite-school.html.
216. Coal. to Defend Affirmative Action v. Regents of the Univ. of Mich., 652 F.3d 607 (6th Cir. 2011).
217. One might criticize this method for relying on facebooks to classify students by race.
218. We had two different graduate students classify the students, and they produced essentially identical results.
219. Moreover, to the extent our classification of blacks might contain errors (say 10% of those we classified as black are of another race), that would tend to raise, rather than lower, our estimate of the group's bar passage rate. Of course, I would also welcome the school to release its own reports on bar outcomes by race.
220. These results imply that Lempert, Chambers & Adams' study 50, overlooked or omitted virtually all the black students-and a very large number of them-who never became lawyers.
221. Since the whole point of the study was to evaluate the post-graduate outcomes of UMLS's minority graduates, it is hard to see how this problem does not invalidate all of their results. Note, however
222. Richard Lempert, after reading a draft of this article, vigorously disputed my estimate. Readers may find additional debate and discussion of this issue at http://www.seaphe.org/.
223. AMER. BAR ASS'N, ABA-LSAC OFFICIAL GUIDE TO ABA-APPROVED LAW SCHOOLS (2002-05 editions). As with the analysis of Howard, supra, I analyzed four years of attrition and bar passage statistics for Wayne State. Because of the small sample size in the disclosed data
224. Michigan estimate applies to blacks taking the bar in all the states disclosed by UMLS. 150.
225. Ayres & Brooks 131; David L. Chambers, Timothy T. Clydesdale, William C. Kidder & Richard O. Lempert, The Real Impact of Eliminating Affirmative Action in American Law Schools: An Empirical Critique of Richard Sander's Study, 57 STAN. L. REV. 1855 (2005)
226. Jesse Rothstein & Albert H. Yoon, Affirmative Action in Law School Admissions: What Do Racial Preferences Do?, 75 U. CHI. L. REV. 649 (2008)
227. Jesse Rothstein & Albert H. Yoon, Mismatch in Law School (May 2009) (unpublished manuscript), available at http://gsppi.berkeley.edu/faculty/ jrothstein/workingpapers/rothstein-yoon-may2009.pdf
228. Daniel Ho, Why Affirmative Action Does Not Cause Black Students to Fail the Bar, 114 YALE L.J. 1997 (2005)
229. Barnes, 121
230. Barnes, 122
231. Williams 124; Timothy Clydesdale, A Forked River Runs Through Law School: Toward Understanding Age, Gender, Race, and Related Gaps in Law School Performance and Bar Passage, 29 LAW & SOC. INQUIRY 711 (2004).
232. Williams 123.
233. Sander, 47
234. Richard H. Sander, Mismeasuring the Mismatch: A Reply to Ho
235. YALE L.J. 2005 (2005).
236. Williams, Sander, Luppino & Bolus 123.
237. This problem is often referred to as "attenuation" or "regression dilution."
238. Rothstein & Yoon 150, at, provides a good discussion of this problem.
239. Ayres and Brooks acknowledge this problem in motivating their second-choice method, but then discount the results that flow from their second-choice model.
240. This is the primary outcome focused upon by Ayres & Brooks, Barnes, Rothstein & Yoon, and, apparently, Ho.
241. Williams refers to this as the "smooth passage" outcome variable; I believe he is the only scholar in this literature to use this very logical measure.
242. A notorious example of this problem is in Ayres & Brooks, 131.
243. When the authors used their "second-choice" model to evaluate mismatch, they found that second-choice students were significantly more likely to pass the bar on their first attempt, and more likely to take fewer attempts to eventually pass the bar.
244. But because they found merely a positive (and not "statistically significant") effect of "second choice" on the probability of ever passing a bar exam, they essentially dismissed the importance of all three results.
245. In fact, as Williams shows in some detail in his 2011 article, and as I suggested in Reply to Critics, all three outcomes are part of a consistent and logical pattern showing mismatch.
246. Ayres & Brooks 131, at 1824. Ayres & Brooks did two principal tests of the mismatch effect in their paper, a "relative tier" test and a "second choice" test.
247. The relative tier test omitted historically black schools, and the "second choice" test generally produced results confirming or consistent with mismatch (tests for mismatch outcomes either showed statistically significant mismatch effects, or showed results consistent with mismatch but not statistically significant).
248. In summing up, the authors discounted the second-choice results and emphasized the deeply flawed "relative tier" results. Id. at 1838.
249. Personal communication from James Lindgren, Northwestern University.
250. Clydesdale 150. Clydesdale's error is that he predicts student law school performance without standardizing his two most important predictors (LSAT scores and undergraduate grades) for the school attended.
251. This is like using today's temperature to predict what month it is without controlling for one's latitude (or even whether one is in the southern or northern hemisphere).
252. Had the editors of Law and Social Inquiry been familiar with the data Clydesdale used, and understood this problem in his method, it is inconceivable that the article would have been published in anything like its current form.
253. Rothstein & Yoon 150 (unpublished manuscript).
254. Doug Williams, Do Racial Preferences Reduce Minority Learning in Law Schools? (Working Paper 2011), available at http://www.seaphe.org/pdf/ williamsseptember.pdf.
255. Extensive materials on the California Bar lawsuit are available at http://www.seaphe.org/topic-pages/california-bar-lawsuit.php.
256. Sander v. State Bar of California, 126 Cal. Rptr. 3d 330, 340 (Cal. Ct. App. 2011).
257. Sander v. State Bar of California, 2011 Cal. LEXIS 9272 (Cal. Aug. 25, 2011).
258. This was one of the central messages of the classic study by Christopher Jencks and David Reisman, The Academic Revolution (2d edition, 1969).