STATEMENT OF UNDISPUTED FACTS
The undisputed evidence in the record demonstrates that the University conscientiously applies what Bakke explicitly permits — the individualized, competitive consideration of race and ethnic origin as one of many factors used in selecting a class of students from among thousands of qualified applicants.
A. Diversity Is Essential to the University’s Educational Mission.
The University of Michigan is a preeminent public institution of higher education. It seeks to compose a class of students of different racial, ethnic, cultural, socioeconomic, and geographical backgrounds, who bring with them different interests, achievements, experiences, and beliefs. The University has made an educational judgment that this “diversity is extremely important to opening the minds and challenging the thoughts of students and faculty alike,” (Cantor Dep. at 45), and that all students derive educational benefits from diversity of all sorts. The University strives to offer an education “that emphasizes the joys of variety and the benefits of engaging those with different experiences and perspectives.” President Lee Bollinger and Provost Nancy Cantor, The Educational Importance of Race, Wash. Post, Apr. 28, 1998, at A17. Creating a diverse student body, therefore, is central to the University’s educational mission: “Diversity is in our best intellectual interest because diversity will increase the intellectual vitality of our education, scholarship, service, and communal life.”6/ (Ex. D, The Michigan Mandate, March 1990, at 4.)
6 / This commitment to diversity as a way of achieving the University’s educational objectives has been echoed by other parts of the University, including the faculty and the alumni association. (See Ex. A, Senate Assembly Statement, The Value of Diversity, Jan. 26, 1998; Ex. B, Statement of National Board of Directors, University of Michigan Alumni Association, Apr. 24, 1998.) Indeed, diversity of all kinds has always been essential to the University’s mission. In 1879, James B. Angell, the University’s third president, delivered a commencement address entitled “The Higher Education: A Plea for Making it Accessible to All.” (See Ex. C, Nancy Cantor, A Michigan Legacy: Ensuring Diversity and Democracy on Campus, Michigan Alumnus, Summer 1998.) In that speech, Angell explained that higher education is “essentially democratic in the best sense of that term.” (Id.) Angell sought to admit students from broader segments of society so that they could transcend the “artificial distinctions of conventional society” through shared interactions. (Id.)
Racial and ethnic diversity is an integral part of the University’s conception of education:
A first-class education is one that creates the opportunity for students, expecting differences, to learn instead of similarities. Likewise, encountering differences rather than one’s mirror image, is an essential part of a good education. Race is educationally important for all students, because understanding race in America is a powerful metaphor for crossing sensibilities of all kinds.Bollinger & Cantor, supra, at A17. Accordingly, the University takes race and ethnic origin into account as one of many factors in making admissions decisions. (See Cantor Dep. at 51.)7/ This fact is undisputed in this litigation. At the same time, racial and ethnic diversity is only one aspect of the overall diversity that the University seeks to achieve on its campus. (See Spencer Dep. at 301-02.) Admissions Director Spencer testified:We look at a number of things when we look at all students. The university is very clear that one of its fundamental educational values is diversity and that we feel as though if we’re going to prepare students for the 21st century or prepare them for living in a very pluralistic society, we need to create an environment on our campus where all types of diversity [are] available.(Id. at 301.) For that reason, the admissions guidelines are structured to allow admissions personnel “to admit not only individuals, but also an entire entering class that can collectively add to the diversity and academic vitality of the University.” (Ex. F, Memo from Theodore Spencer, Sept. 25, 1996.)
7 / See also Gerald R. Ford, Inclusive America: Under Attack, N.Y. Times Op-Ed, Aug. 8, 1999 (extolling the benefits of a diverse campus that emerge “only if students are exposed to America in all her variety,” and explaining that LS&A -- President Ford’s alma mater -- seeks to obtain those benefits by taking race and ethnicity into account in admissions in an “eminently reasonable . . thoughtful . . . fair” manner); William G. Milliken (former Governor of Michigan), Ban on Affirmative Action Can Have Insidious Effects, The Ann Arbor News, Dec. 31, 1999 (noting that “our universities play a vital role in bridging the racial divide by preparing our young people to work and live with members of different racial and ethnic groups” and that the University of Michigan, in particular, has implemented a “thoughtful, multidimensional admissions process” that takes race and ethnicity into account to accomplish this goal).
The Michigan Mandate reflected this conviction. The Mandate was a university-wide initiative launched by then-President James Duderstadt in 1988 to focus attention on the importance of all types of diversity, but with particular emphasis on racial and ethnic diversity. (See Ex. D; Bollinger Dep. at 228; Spencer Dep. at 293.) According to President Duderstadt, the Michigan Mandate “was stimulated by a growing realization of the importance of the diversity of our campus, our students, our faculty and our staff, to the quality of our academic programs and to the educational experience of our students.” (Duderstadt Dep. at 20.) The Mandate recognized that a “multicultural community” would draw “its intellectual strength from the rich diversity of peoples of different races, cultures, religions, nationalities, and beliefs.” (Ex. D at i.) Indeed, the Mandate challenged the University to learn “how to weave together these dual objectives of diversity and unity in a way that strengthens our fundamental goal of academic excellence and serves our academic mission and our society.” (Id. at iii.)
B. The Educational Benefits of Diversity.
Nearly all major colleges and universities in this country join the University of Michigan in recognizing the educational value of admitting and enrolling a diverse group of students. The Association of American Universities, which comprises 62 leading research universities, has recognized the important mission of institutions of higher education, and the indispensable contribution that diversity, including racial and ethnic diversity, makes to that mission:
We speak first and foremost as educators. We believe that our students benefit significantly from education that takes place within a diverse setting. In the course of their university education, our students encounter and learn from others who have backgrounds and characteristics very different from their own. As we seek to prepare students for life in the twenty-first century, the educational value of such encounters will become more important, not less, than in the past.(Ex. E, Association of American Universities, On the Importance of Diversity in University Admissions, Apr. 14, 1998.)The work of Defendants’ nationally recognized experts in the fields of history, sociology, education and psychology confirms the view of the academic community by demonstrating that there is a direct and provable relationship between the significance of race in our society and the quality of education in a racially diverse setting. Furthermore, these experts prove that racial and ethnic diversity on campus enhances the academic skills and civic preparedness of all students — minorities and non-minorities alike — because students educated in a diverse student body (1) are able to think and analyze in more active and complex ways and (2) are better prepared to participate as citizens in our pluralistic democracy. The unrefuted findings of Defendants’ experts confirm that the University’s commitment to the racial and ethnic diversity that contributes to its educational mission reflects a compelling government interest. In fact, the view that diversity in an educational setting provides a broad range of concrete, measurable educational benefits is long-established and widely held. (See Brief for the United States as Amicus Curiae (identifying the virtues of education in a diverse setting, for intellectual, civic, and social activities both on and off campus); Brief of Amici Curiae American Council on Education, et al., (same); Brief of Amici Curiae Association of American Law Schools, et al. (same).)
The expert testimony submitted by the University provides empirical proof for what educators have long believed based on their experience working with students: that “students benefit significantly from education that takes place within a diverse setting.” (Ex. E.) Using national and Michigan student databases, Patricia Y. Gurin, a Professor of Psychology at the University of Michigan with over 35 years of experience in the field, conducted an extensive empirical analysis of how and why diversity in higher education affects students. She demonstrates that students, who come to universities at a critical stage of their development, learn better and think more actively and critically when they are educated with racially and ethnically diverse peers. Learning in a diverse environment hones critical thinking skills by exposing students to situations, perspectives, and experiences that are discrepant from those they are accustomed to encountering and that therefore require them to use deeper modes of thought to understand and analyze. (See Gurin Rep. at 9-18.) Professor Gurin’s analyses show that students who experienced the most racial and ethnic diversity in classroom settings and in informal interactions with other students showed the greatest engagement in active thinking processes, increase in intellectual engagement and motivation, and growth in intellectual and academic skills. (See Id. at 35-38.)
Professor Gurin also demonstrates empirically that academic and social interactions among students in a racially and ethnically diverse student body are instrumental in preparing students to participate more fully and productively in our pluralistic democracy. (See Id. at 39-41.) Education in a diverse setting better equips students to understand and consider multiple perspectives, to deal with the conflicts that different perspectives may create, to appreciate how differences can be harnessed in pursuit of the common good, and to perceive commonalities amidst differences. (See Id. at 18-21.)
It should come as no surprise that education that takes place in a racially and ethnically diverse atmosphere has educational benefits for all students, minorities and non-minorities alike. The simple point Defendants’ experts make is this: race is a defining characteristic of American life. It is a crucible in which Americans experience their lives. This is not to say that race somehow predetermines an individual’s perspectives, or that members of racial or ethnic groups share monolithic views. However, the relevance of race is pervasive.
Our country is a remarkably diverse place, populated by individuals of virtually every race and ethnicity in the world. Indeed, the proportion of non-white individuals in the United States is increasing. See U.S. Census Bureau, Projections of the Resident Population by Race, Hispanic Origin, and Nativity: Middle Series, 1999-2020, January 13, 2000. Yet, Americans of different races and ethnicities tend to live in separate communities and to lead separate lives. Indeed, in some areas, such as housing and elementary and secondary education, our society is as racially separate today as it was before Brown v. Board of Education, before the Civil Rights Act of 1964 and the Voting Rights Act of 1965, and before the Bakke decision. See Kevin Sack & Janet Elder, Poll Finds Optimistic Outlook But Enduring Racial Division, N.Y. Times, July 11, 2000 (reporting that 85% of white respondents said they live in areas with no or few black neighbors; 83% said they worked with only a few blacks or none at all). As a result, Americans tend to enter college without having had meaningful and sustained contact with people from races and ethnicities other than their own. Rather, their views are formed by media images and other stereotypes. Thomas J. Sugrue, Associate Professor of History and Sociology at the University of Pennsylvania and the author of a prize-winning book on postwar urban racial history, submitted an expert report detailing this racial separation and its consequences. (See Sugrue Rep. at 19-44.)8/
8 / Professors Foner and Camarillo, in their respective expert reports, describe the historical events and circumstances that have given rise to the conditions that Professor Sugrue describes. Eric Foner is the DeWitt Clinton Professor of History at Columbia University and the President of the American Historical Association. He points out that race has been a crucial line of division in American society since the 17th century, and that our nation’s experience with its African -American minority has shaped the way Americans view questions of race and issues of similarity and difference. (See generally Foner Rep.) Albert M. Camarillo is Professor of History and Director of the Center for Comparative Studies in Race and Ethnicity at Stanford University. His report outlines the historical patterns and legacies of racial isolation and separation of Hispanics in American society. (See generally Camarillo Rep.)
Michigan provides a stark example of these patterns. Three of the ten most segregated metropolitan areas in the United States are in Michigan. (See Id. at 22, Table 4.) In fact, Detroit is the second most segregated metropolitan area in the country, and the rates of residential segregation in that metropolitan area were higher in 1990 than they were in 1960. (See Id. at 22-23.) Similarly, Michigan ranks in the top four states in the country in the degree of black/white school segregation; far more students attend racially integrated elementary and secondary schools in the South than in Michigan. (See Id. at 38-39.)
The consequences of this persistent racial separation are enormous, creating a profound impact on students’ experiences and perspectives. Most entering students have had few opportunities to benefit from exposure to the ideas and perspectives of people from racial and ethnic groups other than their own. Indeed, according to a survey of University of Michigan students, 92 percent of all white students grew up in neighborhoods that are predominantly white, and 83 percent went to predominantly white high schools. (See Gurin Rep., App. E, at 3.) This lack of contact among individuals of different races and ethnicities fosters misconceptions and mistrust on all sides and affords little or no opportunity either to disrupt the perpetuation of racial stereotypes or to experience the richness of different racial and ethnic communities. (See Sugrue Rep. at 67.) Thus, despite -- or perhaps because of -- the salient role that race and ethnicity play in where Americans live, where they are educated, and whom they See and with whom they interact on a daily basis, many Americans who are not members of minority groups are entirely unaware of the pervasive influence that race still has. Strikingly, the perceptions of Americans who are members of minority groups are vastly different. (See Sugrue Rep. at 60-67;Sack & Elder, supra.) Airing and addressing these divergent perceptions in order to reverse the retreat toward segregation and to prepare students to function in a diverse society is an educational challenge -- one which the University has embraced. This challenge cannot be met without a racially and ethnically diverse student body.
The University necessarily makes its admissions decisions against this backdrop of our diverse, but largely segregated, society and in recognition of the reality of the consequences of this separation. Professor Gurin’s empirical work shows that “patterns of racial segregation and separation historically rooted in our national life can be broken by diversity experiences in higher education.” (Gurin Rep. at 3.) Bringing together on a college campus students from different races and ethnicities cannot help but challenge preconceived notions. It is precisely these encounters -- both in and out of the classroom -- that create an educational opportunity for more complex and active thinking. (See Id. at 15.) Educational experiences in a diverse environment stimulate conscious learning and critical thinking, allowing students to analyze and to move beyond the assumptions and misconceptions created by racial separation. (See Id. at 9-21.) They also prepare individuals to be productive citizens who can participate fully in the political, economic, and social spheres of our multiracial, multiethnic society. (See Id. at 39-41.)9/ For an institution of higher education, creating a diverse student body in which all students can interact with members of different groups is, therefore, central to fulfilling its educational mission.
9 / Moreover, as William Bowen attests, the call for diversity in higher education is echoed outside the academy by businesses that recognize the advantages of hiring university graduates who have learned to work productively in a diverse environment. (See Bowen Rep. at 14-15.) Bowen, one of Defendants’ experts, is President of the Mellon Foundation and the former President of Princeton University. He is the co-author, along with Derek Bok (a professor at the John F. Kennedy School of Government, former President of Harvard University, and former Dean of the Harvard Law School) of The Shape of the River: Long-Term Consequences of Considering Race in College and University Admissions (1998).
C. The LS&A Admissions Process.
Having reached the educational judgment that a racially and ethnically diverse student body is essential to its mission, the University of Michigan takes race and ethnic origin into account as a factor in making admissions decisions. (See Cantor Dep. at 51.) The basic facts and key characteristics of the admissions system are not in dispute: race and ethnic origin is only one of many factors considered in making admissions decisions, all applications receive individualized review by an admissions counselor and are judged against a uniform set of criteria, there is no separate program, tier, or review for minority applications, and there are no quotas, numerical targets, or goals for the admission of minority students.
1. Selective Admissions. Thousands of students from across the country and, indeed, around the world apply to the University each year. The University only admits applicants whom it believes are fully qualified and will succeed on campus. (See Ex. K, University of Michigan Viewbook, 1994-95 at 19 (“[A]dmission is dependent upon a high probability of success in the chosen school or college.”).) Admission to the University is selective. Many more students apply each year than can be admitted. In 1997, for example, the University enrolled 3,958 freshmen in LS&A from over 13,500 applications. (See Raudenbush Supp. Rep., March 3, 1999, at Table 2c.) In addition, a very high percentage of these applicants are fully qualified for admission, providing a talented and rich pool from which to compose a class. (See Cantor Dep. at 54-55; Spencer Dep. at 84-86.)
2. Admissions Procedures. The Office of Undergraduate Admissions (“OUA”) reviews all applications. The sheer volume of applications received each year requires the establishment of procedures and routines to promote consistency in applying admissions standards. (See Ex. Z at 1.) At the same time, OUA values the individualized review professional counselors accord each application, the myriad factors that combine to make a candidate attractive, and the expertise of the counselors in discovering and assessing these factors based on a student’s application. (See Spencer Dep. at 56; McKinney Dep. at 25; Ex. Z.) In response to these competing considerations, OUA has created written guidelines for evaluating applications and assisting in making admissions decisions. (See Ex. Z at 1.) With these guidelines, OUA aims to “blend the consistency of a formula with the flexibility of a review that is ultimately a matter of human judgment …” (Id.) In fact, OUA cautions its personnel that “[a]dmissions is more art than science, and these guidelines should not be read otherwise.” (Id.)
Over the years, the guidelines used by the admissions counselors have taken different forms. (See, e.g., Exs. N, O, P, M, BB, Y (College of LS&A Guidelines for 1995-2000).) The OUA reviews its admissions process every year and considers ways to make it simpler, clearer, and easier to implement. (See Spencer Dep. at 96-97, 116, 143-44; Seltzer Dep. at 113.) But, while the mechanics of the admissions process have varied over time, the policies underlying the guidelines have not changed -- including the policy of considering the race of applicants as a single but important factor in admissions, as permitted by Bakke, in order to achieve racial and ethnic diversity on campus. Clerks at OUA handle the initial processing of all applications to LS&A. One set of clerks processes the application fee and enters biographical data into the OUA admissions database. (See Gauss Dep. at 12-13.) In addition, another set of clerks, using a procedure book, computes an academic grade point average (“GPA”) using a student’s academic courses from 10th and 11th grades. (See Id. at 16-17.) Clerks do not make any decisions to admit or reject any applicant. (See Ex. EE, Defs.’ Supp. Resp. to Int. 1, June 9, 2000, at 4.)10/
10 / Under a prior incarnation of the admissions program, students -- of any race -- with exceptionally high standardized test scores and high school GPAs typically were admitted based solely on these rankings. (See Spencer Dep. at 39-40.) LS&A expedited admission of these top scholars, whether by clerk or counselor (See Ex. I, Instruction for Automatic Decisions on LS&A Applicants; Ex. J, 1998 Freshman Application Sort) as a recruitment device designed to maximize the chance that these highly sought-after applicants would choose to enroll at the University. (See Spencer Dep. at 58-59.) Although these "automatic admissions" have been eliminated, (See Ex. EE at 4), LS&A has tried to preserve its ability to compete with other institutions for top scholars by expediting review of applicants with particularly high grades and test scores. (See Ex. AA at 5.) In addition, the guidelines previously permitted clerks to cause rejection letters to be sent to applicants with very low grades and test scores without counselor review. However, despite the wording of the guidelines, in practice the clerks were uncomfortable making decisions to reject students without counselor input and consequently generally forwarded all applications to the counselors for review. (See Gauss Dep. at 61-62.)
Rather, all applications are turned over to admissions counselors for individualized review. Each of the approximately 20 counselors is assigned a geographic territory and reviews all applications from that region, thereby developing territorial expertise, including detailed knowledge of school strength, curricula, and any special geography-based factors. (See Spencer Dep. at 28; Gauss Dep. at 30; Vanhecke Dep. at 99.) Applications from minority students are assigned to counselors responsible for the relevant geographic area, and are reviewed along with the applications of everyone else. This fact bears emphasis: unlike the admissions system that was rejected in Bakke, there is no separate assignment or review of minority applications. (See Spencer Dep. at 273, 27; Gauss Dep. at 31; Vanhecke Dep. at 99.)
Counselors review applications as they are received, and admissions decisions are made on a staggered basis at certain intervals throughout the admissions cycle. (See Knepp Dep. at 21; Spencer Dep. at 177.) This kind of phased admissions system requires careful management to avoid over-enrollment and to enroll an entering class with the spectrum of attributes the University values. (See Knepp Dep. at 31.) As part of this process each year, the University sets an enrollment target for the total number of students in the entering freshman class in order to allow appropriate planning for academic programs (course development and scheduling, faculty workload and hiring, class size, etc.), the budget, financial aid, housing, orientation, and other services. (See Cantor Dep. at 73.) The University also sets targets for in-state residents and for enrollment in each of the academic units. (See Spencer Dep. at 12, 183; Knepp Dep. at 14.) There are no other enrollment targets. In particular, there are no targets for the admission of underrepresented minority students. The consistent testimony in the record unequivocally establishes that the University does not have any numerical targets or goals for the admission or enrollment of underrepresented minority students.11/ (See Cantor Dep. at 83; Spencer Dep. at 12, 290; Vanhecke Dep. at 67; Seltzer Dep. at 36, 118; McKinney Dep. at 111, 121.)
11 / The University has discontinued its prior practice of using so-called “protected seats” as an enrollment management technique (See Ex. EE at 4) to ensure that the University did not exceed or fall short of its overall enrollment target, and that it achieved the desired distribution of students among the academic units, the appropriate balance of in-state and out-of-state enrollment, and the desired composition of the entering class (See Spencer Dep. at 52; Seltzer Dep. at 27.). “Protected seats” were estimates of the number of applications that should be expected in the last stages of the admissions process from groups who tend to apply later in the cycle -- including athletes, foreign students, ROTC candidates, and underrepresented minorities. (See Spencer Dep. at 177-78; Knepp Dep. at 17.) They did not reflect specific slots in the class. These predictions were used to pace admissions decisions to permit full consideration of these later applications without over-enrolling the class. (See Knepp Dep. at 23, 28; Spencer Dep. at 52, 232; Seltzer Dep. at 27, 151.) Minority students who applied during the later stages of the admissions cycle were evaluated competitively against the same criteria that had been applied to all other applicants, regardless of race. (See Spencer Dep. at 107-08.)
3. The Selection Index. The counselors evaluate applications to LS&A using a “selection index” worksheet. The worksheet prompts the counselor to enter a numerical value for each of a number of academic and other factors, up to a possible total of 150 points. (See Ex. Z at 1; Spencer Dep. at 133-35.) Although some of the categories provide for fixed point values, several permit the counselor to exercise discretion in determining how many points to award for a given factor. A student’s high school academic GPA receives the most points -- a possible 80 out of 150. (See Ex. Y; Ex. Z at 1.) Students also receive a range of points depending on their standardized test scores.12/ (See Id. at 4.) The “School” factor accounts for the strength of a student’s high school. The guidelines call for counselors to exercise judgment in determining the School factor for each school in their territories, in light of their familiarity with the quality of education offered at that school; each student from that school then receives the same number of points. (See Id. at 2; Spencer Dep. at 200.) The rigor of each applicant’s particular high school course of study is reflected in the “Curriculum” factor. As the guidelines state, “[g]iven the wide disparity in high school course selection and offerings, it is imperative that the choice of strong courses … be considered in the review process.” (Ex. Z at 3.) Counselors therefore subtract points if an applicant’s record shows that he or she chose a weaker curriculum when a stronger one was available, and then add points for an applicant who chose a more challenging course load. (See Id. at 3-4.) Again, in assigning the appropriate point value, counselors are encouraged to draw upon their knowledge of the meaning of course labels (e.g., “honors”) used by the schools in their territory. (See Id.) Applicants can be awarded a maximum of 110 points for all of these academic factors taken together. (See Id. at 1.)
12 / Applicants receive far fewer points for standardized test scores than for GPA because test scores are highly correlated with GPA and therefore do not add much predictive value to an applicant’s future academic performance. Standardized test scores only increase the prediction of an applicant’s college freshman grades by about 3 or 4 percent over predictions using high school grades alone. (See Steele Rep. at 5.)
Applicants can also receive a maximum of 40 points from a variety of other factors evaluated by the counselor. (See Id. at 4.) The “Geography” factor assigns points to Michigan residents (ten points), students from underrepresented Michigan counties (six points), and students from underrepresented states, such as many western and southern states (two points). (See Id. at 4-5.) The “Alumni” factor awards either one or four points to applicants who are close relatives of alumni, depending on the specific relationship. (See Id. at 5.) Counselors may also award up to three points for the required personal essay, taking into account the originality, organization, subject matter, and writing quality. (See Id. at 6.) In addition, based on the information applicants supply as to activities, work experience, and awards, counselors may, in their discretion, assign up to five points for leadership and service, as well as an additional five points for personal achievement -- as evidenced by persistence, character, commitment to high ideals, and level of awards. (See Id. at 6-7.) Finally, applicants may receive 20 points for being socioeconomically disadvantaged, a member of an underrepresented minority group, 13/ from a predominantly minority high schools (regardless of race), or a recruited athlete. Twenty points may also be provided by the Provost’s discretion. (See Id. at 7-8.) Only 20 points, in total, may be awarded for these factors. (See Spencer Dep. at 156-58.)
13 / The University of Michigan considers underrepresented minorities to be African-Americans, Hispanics, and Native Americans. (See Spencer Dep. at 73.) Because the University receives a substantial pool of applications from Asian-American students, the University is able to achieve diversity with respect to such students without considering their race in the admissions process. (See Cantor Dep. at 72.)
The guidelines themselves make clear that admissions decisions are not reduced solely to grades and test scores.14/ (See Ex. Z; Ex. AA.) As a result, it often turns out that there are applicants in every racial and ethnic group who have been accepted with lower grades and test scores than other applicants with higher grades and test scores who have been denied admission. Nevertheless, only qualified applicants who can be expected to succeed at the University are admitted. (See Ex. K at 19.)
14 / For classes admitted before 1998, the guidelines were implemented by means of grids instead of a selection index. Each grid had a vertical axis with adjusted GPA ranges (that included school, curriculum, and other factors) and a horizontal axis with standardized test score ranges; each cell on the grid contained various admission action options available to counselors. Michigan residency, alumni status, and underrepresented racial and ethnic status were accounted for in different grids, as a “plus” factor rather than within one grid or along one axis of a grId. The University devised the selection index in 1997 to simplify its admissions process. The selection index “was designed to be more straightforward, more understandable.” (Seltzer Dep. at 93.) While the selection index is different in appearance, the changes to the process were mechanical, not substantive. (See Id.) David Hunter, the statistician who created the selection index in consultation with the OUA, testified that the selection index was designed to compose a student body with a range of attributes similar to the classes selected under the prior system. (See Hunter Dep. at 76-77.) Under the grids, as with the selection index, an applicant’s race was considered simply as one of many factors in the admissions process.
Once a counselor has arrived at a selection index score for an applicant, the counselor then decides whether or not to flag the application for consideration by an Admissions Review Committee (“ARC”), which is comprised of members of OUA and the Office of the Provost. (See Ex. EE at 2-3.) The ARC provides an additional level of consideration beyond the initial counselor review. It is designed to allow discussion for the more difficult decisions for those applicants whose selection index score may not accurately reflect the contribution they would make to the University. The ARC makes admissions decisions while -- quite literally -- sitting around a table reviewing and discussing the flagged applications. This kind of committee format for admissions decisionmaking is used widely by smaller colleges around the nation. See, e.g., Jacques Steinberg, College Gatekeepers Read Between the Lines, N.Y. Times, Apr. 3, 2000 (discussing the “communal” discussions the admissions committee at Wesleyan University engages in for some applications after rounds of individual application review). The University has adopted a modified version of a committee review (as the volume of applications permits) in order to capture the virtues of extended discussion and debate that it offers for the more difficult and complex admissions decisions. The ARC differs only in form, not substance, from the individualized counselor review that each application always reviews: “[a]s with the other components of [the University’s] admissions process, the overall goal is to admit a class of qualified students who best serve our educational mission both individually and as a group.” (Ex. AA at 2.)
A counselor may, in his or her discretion, choose to flag an application for ARC discussion that meets all of the following three criteria: (1) in the estimation of the counselor, using his or her professional expertise, the applicant is academically prepared to do the level of work required at the University; (2) the applicant’s selection index score is greater than or equal to 80 for Michigan residents or 75 for non-residents; and (3) the applicant possesses a quality or characteristic important to the University’s composition of its freshman class. These attributes include high school class rank; unique life experiences, challenges, circumstances, interests or talents; socioeconomic disadvantage; underrepresented race, ethnicity, or geography; connections to the University community; athlete; or application through the “on-the-spot” program.15/ (See Id. at 2-5; Ex. EE at 2-3.) In addition, an admissions counselor may flag an application if something in the applicant’s file, such as a pattern of declining grades or disciplinary problems, indicates that the applicant may not be suitable for admission despite a high selection index score. (See Ex. AA at 5.)
15 / The “on-the-spot” program is an admissions recruiting program designed to attract students from underrepresented minority groups from Michigan high schools with relatively high concentrations of such students. (See Spencer Dep. at 188-95.)
After the counselors complete their individualized review of each application, they enter the applicant’s selection index score and flag (if any) into the OUA database. Then, admissions decisions are generally executed in one of two ways. First, the Enrollment Working Group (“EWG”)16/ sets parameters, by selection index score, that determine what admissions action -- admit, defer, or deny -- will be executed with respect to all applications awaiting action in the OUA database. (See Ex. EE at 3.)17/ These parameters apply across-the-board, regardless of the race of the applicant, to all applications in the OUA database. Decisions are made at certain intervals during the admissions season. Periodically, EWG examines and, if necessary, adjusts these parameters to prevent over-enrollment and to pace the admissions process appropriately. (See Id.) Second, the ARC reviews and discusses applications that were flagged by the admissions counselor but turn out not to be admitted based on EWG parameters. After discussion, the committee decides whether to admit, deny, or defer the applicant. (See Id.)
16 / The EWG, chaired by the assistant provost for University budget and planning, Marilyn Knepp, (See Knepp Dep. at 8), comprises representatives from various University offices, including the Office of Budget and Planning, the Office of Academic Affairs, OUA, the Office of Financial Aid, the Office of Academic Multicultural Initiatives, LS&A, and other colleges of the University. (See Ex. R, Defs. Resp. to Int. No. 1, April 7, 1998, at 7.)
17 / The University no longer undertakes to make immediate decisions to admit or deny underrepresented minority applicants. (See Ex. EE at 4.)
D. The Data Illustrate the Admissions Process.
The LS&A admissions data in the record paint a picture of how the admissions process works and the role that an applicant’s race may play. Several points, all undisputed, can be made from a review of the basic, descriptive numbers. First, the pool of minority applicants to LS&A is consistently small. Applications from underrepresented minority groups constituted only 11.4 percent of applicants to LS&A in 1995, 11.2 percent in 1996, and 9.5 percent in 1997. (See Raudenbush Supp. Rep., March 3, 1999, at Tables 2a-2c.)
Second, the number of students from each racial or ethnic group admitted to LS&A varies from year to year. Since 1995, underrepresented minority students have represented between 13 and 16 percent of the entering classes. (See Id. at Tables 3a-3c.) The figures on admitted students do show that a somewhat greater proportion of minority applicants are admitted than non-minority applicants. In 1997, for example, 87.0 percent of underrepresented minority applicants were accepted, compared with 74.2 percent of non-minority applicants. (See Id. at Table 3c.) According to Stephen W. Raudenbush, a leading educational statistician and one of Defendants’ experts, “[c]omparison of average probabilities of admission for sub-groups of students defined by membership in an under-represented racial or ethnic minority group provides a simple, first check on the importance of [race] in admissions.” (Id. at 4.) These figures indicate that race does not play the overwhelming role in the process that Plaintiffs claim it does. If race played an overwhelming role, Raudenbush pointed out, one would expect the differences in the average probabilities of admission to be greater. (See Id. at 4-5.)
This analysis can be misleading if not understood in the context of the absolute numbers behind the percentages, which are strikingly different for minority and non-minority students. For example, in 1995, for students who had a GPA between 3.80 and 3.99, and SAT scores between 1400 and 1490, the tables show that 262 out of 268 non-minority students were admitted, as compared to five out of five minority students. (See Ex. S at UMA 030393; Ex. T, at UMA 030396.) The percentages indicate that 100 percent of the minority students in this range of grades and test scores were admitted, while only 97.76 percent of the non-minority students were admitted. (See Id.) Knowing that 100 percent of admitted minority students actually refers to five students is quite different than analyzing the percentage alone. It is highly likely that of any group of five non-minority applicants with this combination of grades and scores, all five were admitted as well.
It is also clear that admissions rates vary significantly depending on the grades and test scores of the applicants. The statistical analysis done by Professor Raudenbush confirms that students of all races with high GPA and test scores typically are admitted. Students of all races with low grades and test scores typically are denied admission. For the many students in the middle range, the decision to admit or deny is more complex, and other factors in the process have greater significance. (See Raudenbush Supp. Rep., Mar. 3, 1999, at 6.)
Third, while there is some difference between the average grades and test scores of admitted majority and admitted minority students, the average grades and test scores for all students admitted to LS&A are quite high. For example, in 1996, the average SAT score of majority students enrolled in LS&A was 1272, and the average high school GPA was 3.66, (See Ex. U, Profiles for Non-minority Students, Spring-Fall 1996, at UMA 022187-88), while the average SAT score for minority students was 1113, and the average GPA was 3.29, (See Ex. V, Profiles for Underrepresented Minorities, Spring-Fall 1996, at UMA 0022107-08.). 18/ These facts reflect the implementation of the University’s policy only to admit applicants who can succeed in its academic program.
18 / These small differences in average SAT scores are not significant. Professor Claude Steele, who is Chair of the Department of Psychology at Stanford University and one of Defendants’ experts, pointed out that a difference in SAT scores as large as 300 points "actually represents a very small difference in skills critical to grade performance." (Steele Rep. at 5.)
Finally, one cannot adequately understand the complex admissions process employed by the University without considering the proportion of admitted students who choose to enroll in LS&A -- the so-called "yield." The stiff competition among selective colleges and universities for qualified students means that well under half of the admitted students actually choose to enroll.
The small pool of minority applicants and the low yield for admitted minority students combine to produce a predictable result: in order to achieve even the limited diversity it does, the University ends up admitting virtually all qualified underrepresented minorities. (See Spencer Dep. at 83-84; Vanhecke Dep. at 36, 38; Seltzer Dep. at 121-22; McKinney Dep. at 56-57.) This practical result does not mean that minority applications have been insulated from “competitive consideration” with non-minority applications. Nor does it mean that a different “policy” (i.e., admitting all members of a given group) is applied to minority applicants, as opposed to non-minority applicants. The applications of all students, regardless of race, are considered individually against the same selection index criteria. (See Spencer Dep. at 107-08.) Those students who are not qualified are rejected. (See, e.g., Ex. W, Memo from Theodore Spencer, Oct. 4, 1995.) Among those students who are qualified, the University must make choices and enroll a class that reflects its educational mission and goals. (See Spencer Dep. at 306.) Not all qualified students can be admitted to the University because of overall enrollment targets that the University cannot exceed. Because the number of minority applicants is so small, as a practical matter, the University ends up admitting nearly all who are qualified in order to enroll meaningful numbers of minority students. (See Spencer Dep. at 83-84.)
E. The University Vigorously Recruits Minority Applicants.
The pool of qualified underrepresented minorities that do apply to the University is the result of aggressive efforts to recruit as many qualified applicants as possible. As Provost Cantor said, “[as] a consequence of our desire to admit and compose a diverse class, given the small application pool and the selective nature of our admissions, we work to recruit as many highly qualified minority students to the University of Michigan as we can.” (Cantor Dep. at 68.) Accordingly, the University has launched numerous recruiting programs designed to attract minority students, and it is constantly trying to find new ways to recruit underrepresented minorities. The University’s efforts have ranged from personal contact with minority applicants by counselors to minority student symposia, attendance at recruiting fairs in major urban areas, direct mailings, and campus visits for minority students. (See Vanhecke Dep. at 11-12; Ex. D at vi, 19.) OUA also maintains an office in Detroit to recruit local high school students, most of whom are African-American. (See Spencer Dep. at 196-97, 29.) The University’s efforts to “do just about all aspects of recruitment that [are] possible,” (Vanhecke Dep. at 9), nevertheless produce only a small pool of qualified minority applicants, most of whom are highly recruited by other selective colleges and universities as well.
|
|