I. Institutional Racism and the Misuse of the LSAT
Racism is any conscious or unconscious action or attitude that subordinates an individual or group based on skin color or race. It can be enacted individually or institutionally. Most civil rights activities over the past thirty years have focused on individual racism. Institutions, however, are just as capable of being racist. Institutions can behave in ways that are overtly racist (i.e., specifically denying services to Blacks) or inherently racist (i.e., adopting policies that result in the exclusion of Blacks). This racism causes institutions to respond differently to Blacks and Whites. When Blacks are injured as a result, this behavior is racist in outcome, whether or not racist in intent.
When individuals disagree on elementary justice, their most insoluble conflict is between institutions. . . . The more severe the conflict, the more useful to understand the institutions that are doing most of the thinking. Exhortation will not help. Passing laws against discrimination will not help. . . . Only changing institutions can help. We should address them, not individuals, and address them continuously, not only in crises.
Racism is both overt and covert, and takes three closely related forms: individual, institutional, and cultural or systemic. Individual racism consists of overt human actions that cause death, injury, and the destruction of property or the denial of opportunity. Institutional racism is more subtle, but no less destructive. Institutional racism involves policies, procedures, or patterns of behavior that have a disproportionately negative effect on racial minorities' access to and quality of goods, services, and opportunities; the intent is irrelevant.
Stokely Carmichael and Charles Hamilton coined the term institutional racism in 1967. The MacPherson Report, a British government report, defines it as:
[T]he collective failure of an organisation to provide an appropriate and professional service to people because of their colour, culture or ethnic origin. It can be seen or detected in processes, attitudes and behaviour which amount to discrimination through unwitting prejudice, ignorance, thoughtlessness, and racist stereotyping which disadvantage minority ethnic people.
Over the last thirty-eight years, institutional racism has assumed new meanings. The definition used in this paper is consistent with Professor Haney Lpez's new institutionalism, which consists of the background scripts and paths that mark social and organizational life and that impose harmful effects on minority communities, irrespective of the actions or attitudes of individual decision makers. Cultural or systemic racism is the basis of individual and institutional racism, as it is the underlying value system that supports and allows discrimination based on perceptions of superiority and inferiority. To understand institutional racism, it is important to understand the interaction between discrimination and prejudice. Prejudice is an attitude that is based on limited information or stereotypes. While prejudice is usually negative, it can also be positive. Both positive and negative prejudices are damaging because they deny the individuality of the person. No one is completely free of prejudices although they may not have any prejudice against a particular group. Discrimination is behavior, intentional or not, which treats a person or a group of people disrespectfully on the basis of their racial origins. In the context of institutional discrimination, power is a necessary element, for it depends on the ability to withhold social benefits, facilities, services, or opportunities from someone who should be entitled to them. Intent is irrelevant; the focus is on the result of the behavior.
Given the interaction of prejudice and discrimination, an institution can be non-racist, timidly or reformed racist, reluctantly racist, and overtly racist. Using Blacks as the focal group, a non-racist institution has no biases or prejudices against Blacks and no discriminatory behaviors. Such an institution is very rare. Where an institution describes itself as non-racist, it might be because it operates in an arena that has very little contact with Blacks; it is more likely, however, that the institution is in denial.
A timid or reformed racist institution has definite bias or prejudice against Blacks (for example, a law firm that believes that Blacks are less capable of being good attorneys), but does not engage in discriminatory behaviors (admission or hiring policies and practices that do not discriminate against Blacks). This form of racism involves institutions that harbor biases or prejudices but are either too timid to act upon those prejudices or that are actively working to be less discriminatory. The prejudices or biases are still present, but these institutions no longer act on them.
An overtly racist institution has definite bias or prejudice, such as a belief that Blacks are less capable of being good attorneys, and definite discriminatory behaviors, such as deliberately refusing to admit or hire Blacks who meet its other criteria. Most people are familiar with this form of racism. It involves actively and intentionally expressing bias or prejudice and actively discriminating against others in public and private ways. A reluctantly racist institution purports to have no negative biases or prejudices against Blacks but engages in clearly discriminatory behaviors, such as setting presumptive denial cut-offs at a very high level which have a disproportionately negative impact on Black applicants. This type of racism is perhaps the most pervasive form of racism and also the hardest to challenge. Because the discriminatory behavior is motivated by reasons that are separate from race, such as economics, it is difficult for the institutions that are discriminating to believe that they are being racist and even more difficult for them to abandon the behavior. For example, law schools that adopt admission policies that are not related to performance but that threaten the inclusion or admission of Blacks would fit into this category. Once an institution becomes aware of the discriminatory impact of its policies and practices and fails to change the policies and practices, however, then the institution is no longer reluctantly racist but overtly racist.
Given the above understanding, the essential elements to establishing institutional racism are:
(A) Is the problem based on policies, practices, procedures, or patterns of behavior of an institution?
(B) Do the policies, practices, procedures, or patterns of behavior have a disproportionately negative impact on racial minorities?
(C) Do the policies, practices, procedures, or patterns of behavior serve a legitimate educational goal or purpose?
(D) Are the policies, practices, procedures, or patterns of behavior necessary? In other words, are there alternative policies, practices, or procedures that would serve the educational goal or purpose and have a less discriminatory impact?
A. Is the Problem Based on Policies, Practices, Procedures, or Patterns of Behaviors of an Institution?
The first step in establishing the presence of institutional racism is determining whether the problem is based on the policies, practices, procedures, or patterns of behavior of an institution. The primary institutions in this case are law schools, and the policy or practice in question is the misuse of the LSAT in the admission process by using it as the sole or primary factor in the admissions process. At least ninety percent of law schools have admission practices that presumptively deny applicants based on where they fall on a grid formulated around LSAT and UGPA. Little or any consideration is given to other factors. This misuse of the LSAT has a discriminatory impact.
In the upcoming discussion, I will use my school, the University of Dayton, as an example of a systemic problem in legal education. The University of Dayton is a private, Catholic, Marianist law school with a stated mission of social justice and commitment to diversity. The city of Dayton, Ohio is almost fifty percent African American. Three colleges in the area have a significant black student body: Wright State University, Central State University, and Wilberforce University, with the latter two being Historical Black Colleges and Universities (HBCUs).
The University of Dayton School of Law's admission practice is to establish an LSAT/UGPA Grid and to admit most students on the basis of that grid. In 2003, I was on the admission committee for the third time. This discussion is based on my experience. Other than the grid, there was no written standard or criteria against which a candidate's file is reviewed. When applications arrive in the admission office, the admission director reviewed the complete file. The admission director ultimately assigned an admission status based primarily on the applicant's LSAT/UGPA. That status was either presumptive admit, presumptive deny, or committee review.
Presumptive admit meant that an application was admitted primarily on the basis of the student's combined LSAT and UGPA. Presumptive deny meant that the application was denied primarily on the basis of the student's combined LSAT and UGPA. Files that were sent for committee review were evaluated on the basis of a full file review and voted on by the individual committee members. There were no written criteria for committee review. Each committee member applied his or her own unarticulated criteria. Thus, in practice, some committee members continued to use the LSAT/UGPA as the exclusive factor in making admission decisions.
The admission director could ignore the grid and send some exceptional files to the admission committee. This occurrence was rare, but when it did occur, the candidate usually was a presumptive admit with a criminal record. Less frequently, the candidate was a presumptive deny with an extraordinary record, such as a high UGPA and a graduate degree.
For the 2003-2004 admissions committee, the presumptive deny level was set at any LSAT score below 145. This meant that if two similar files were received by the admission office, the only significant difference between the files was that one had an LSAT of 144 and the other had an LSAT of 145 and both had a UGPA of 3.49, the file of the applicant with the 145 LSAT would have been sent to committee review. The applicant with a 144 LSAT would have been presumptively denied.
One aspect of the University of Dayton's 2003 admission grid was that it gave preference to test-taking ability over demonstrated classroom ability. For instance, an applicant who scored between 165 and 180 on the LSAT was presumptively admitted even though that applicant only performed in undergraduate school at a C+ (2.5) level. In contrast, an applicant who scored between 120 and 139 was presumptively denied even though the applicant performed in undergraduate school at an A+ (3.75 or above) level.
The 2003 admission committee consisted of six members: four faculty members, the admission director, and a student. The faculty members were chosen by the dean. At committee meetings, members were given three lists: presumptive deny, presumptive admit, and committee review. Each of these lists included: the name of each applicant; the LSAT score and UGPA; the major and name of the educational institution from which the applicant graduated; and the applicant's age and racial/ethnic background, if available. Based on this limited information, the committee members could ask questions about any applicant and request the committee to review the file. Otherwise, committee members voted on the admission of every candidate based on the information provided by the Admission Director and the presumptive category assigned, not on a full file review. Proponents of the cut-off system defend it on the grounds that the presumptive deny is refutable and that any faculty member could pull any file and put it on the list for full committee review. In practice, however, very few files were pulled out of the presumptive deny category and reviewed by the committee, and only very few of these candidates were admitted. In 2003, 95.8% of all presumptive admits were admitted and 99.55% of all presumptive denies were denied. That is, of the 449 applications that were presumptively denied, only two were subsequently admitted. Thus, the presumptive deny category for 2003 was virtually absolute. This is not uncommon among law schools.
In any law school, there may be some disagreement over what process is being used, but those disputes are irrelevant to the essential nature of the problem. Whatever the process, the key is whether it includes the practice of drawing a line below which students are denied and above which students' files are either reviewed or the candidates are admitted. The practice of having a virtually absolute presumptive deny is objectionable. The higher the presumptive deny level, the more this practice will work against racial minorities, and especially Black Americans.
B. Do the Policies, Practices, Procedures, or Patterns of Behavior Have a Disproportionately Negative Impact on Racial Minorities?
A facially neutral practice is discriminatory when it has a disproportionately negative impact on a particular racial group. In 1988, the Supreme Court ruled that practices, adopted without a deliberately discriminatory motive, may in operation be functionally equivalent to intentional discrimination. In fact, when confronted with discrimination in the context of standardized tests, the Supreme Court has repeatedly held that the facially neutral practices may violate civil rights law even in the absence of a demonstrated discriminatory intent.
In the case of law schools, a prima facie case of institutional racism is established by showing a disproportionately negative impact on minority applicants. For instance, disparate impact is demonstrated by statistical evidence that shows that the admission practice has disproportionately excluded Blacks from the law school in question.
Thus, when looking at a law school's admission practices, if no significant portion of the presumptive deny category is actually admitted, then the category is absolute. For instance, at the University of Dayton, during the 2003 admission cycle there were 449 files in the presumptive deny category. Of those files, the admission director sent approximately forty--or ten percent-- to the committee for review. Only .45%--or two out of 449--of presumptive denies were admitted, however, making presumptive deny essentially absolute. It is this admission practice that has a disparate impact on the admission of qualified Blacks. The statistical disparities in this case are sufficiently substantial that courts will allow an inference of causation. In fact, the admission practice of having a presumptive deny level has a disparate impact in that (1) the majority of Blacks are given a different review from the majority of Whites; and (2) many otherwise qualified Blacks are denied admission based solely on the LSAT.
1. The Majority of Blacks Are Given a Different Review from the Majority of Whites
Schools argue that all files are reviewed and are given similar consideration under the same factors. Typically, applicants who fall into the presumptive deny category are reviewed differently from those who fall into the committee review. For instance, at the University of Dayton in 2003, applicants who were in the presumptive deny category had a full file review only by one person, the Admission Director; applicants in the committee review category had a full file review by six persons (four faculty, the admission director, and a student). Admission or denial by the committee review required a vote of four persons. Although it is a color-blind policy, it has a known disproportionate negative effect on Blacks. In short, the majority of black applicants do not receive a complete review. In 2003, at the University of Dayton, over 64.3% of Black applicants were presumptively denied, compared to only 20.4% of White applicants. In other words, 80% of the White applicants had a chance of being admitted because their files were either presumptively admitted or underwent full file review. Only 35.7% of the Black applicants had that same process. In short, only 35.7% of the Black applicants had a realistic chance of admission. This difference is generally regarded by courts as evidence of adverse impact.
2. The Presumptive Deny Admission Practice Results in Many Qualified Blacks Being Denied Admission
An applicant is qualified for law school if his or her application demonstrates that, with appropriate educational interventions, he or she is capable of successfully completing law school, passing the bar, and becoming a competent lawyer. The law schools' changes in admission practice had a significant impact on Black enrollment, denying admission to many qualified Blacks based solely on LSAT scores. In the incoming 2003 class at the University of Dayton only 3.8% were Blacks. In previous years, Black enrollment reached highs of eight to ten percent. The decrease was due entirely to the law school failing to admit Blacks whose LSAT scores were in the 140-144 range. From 1991 through 1997, Blacks with LSATs below 145 accounted for approximately 42% of those who matriculated to the University of Dayton School of Law. This is based on information collected about Blacks who participated in the Academic Excellence Program (AEP). This program provided academic support to racial minorities and other non-traditional students. While not all Blacks participated, the vast majority of them did. Thus, a change in policy which placed primary emphasis on the LSAT and presumptively denies anyone with an LSAT below 145 necessarily resulted in a disproportionately high number of otherwise qualified Blacks being denied.
Admission policies and practices that set a presumptive deny level unfairly disadvantaged Blacks by not taking into account the full range of indicators of merit, such as UGPA, graduate degrees, unique work and life experiences, and ability to overcome hardships. Accordingly, policies or practices that have the effect of placing a disproportionate number of Blacks in the presumptive deny category, causing a disproportionate percentage not to receive the benefit of full file review by a committee and causing a substantial drop in enrollment of Black students establishes a disparate negative impact on Blacks. C. Do the Policies, Practices, Procedures, or Patterns of Behavior Serve a Legitimate Educational Goal or Purpose?
Once a discriminatory impact has been established, the practice can still be justified by showing that it serves a legitimate educational goal or purpose. In the case of educational institutions, the courts have held that a policy, practice, procedure, or pattern of behavior can be justified through a legitimate claim of educational necessity. Educational necessity exists when the challenged practice serves a legitimate educational goal. In an educational context, the challenged action must bear a manifest demonstrable relationship to classroom education. The doctrine of necessity is very narrow, and once disparate impact has been established, the educational institution engaged in the challenged action has the burden of establishing educational necessity.
The justifications for implementing the presumptive admit/presumptive deny grid include: (1) assuring admission to students who have the requisite ability to complete the program successfully; (2) improving first-time bar passage rate; (3) improving the overall quality of the class and improving the U.S. News & World Report ranking; and (4) reducing work load. The question is whether these reasons satisfy the legal requirement for educational necessity, or in other words whether the justifications are based on a manifest demonstrable relationship to classroom education.
1. Justification Based on Assuring Admission of Students Who Have the Requisite Ability to Successfully Complete the Program Is an Educational Necessity, But the Use of Cut-offs Is Generally Not Supported by Academic Performance Evidence
Law schools have a responsibility to admit students who can be academically successful in law school and in the practice of law. The immediate concern of law schools is to admit students who, at a minimum, have sufficient ability to maintain a cumulative grade-point average of 2.00 or higher, which is necessary to graduate. The LSAT is used as an admission tool to help predict ability to perform successfully in the first year of law school. According to the Law School Admission Council (LSAC),
[t]he LSAT should be used as only one of several criteria for evaluation and should not be given undue weight solely because its use is convenient. Those who set admission policies and criteria should always keep in mind the fact that the LSAT does not measure every discipline-related skill necessary for academic work, nor does it measure other factors important to academic success.
Furthermore, as LSAC acknowledges, the LSAT is a skills test and not an abilities test. This is an important distinction, as skills can be taught, but abilities cannot. Other factors need to be considered to determine whether the individual has the requisite abilities to succeed in law school and as a lawyer. Second, the accuracy of the LSAT as a measure of skill is, at best, moderate. According to LSAC, the probability of a single score--a score of 150, for example--representing the true ability of a student is sixty-five percent with a seven point spread (147 to 153). For a statistically significant probability of ninety-five percent, there is a fourteen point spread (143 to 164), and a statistically significant probability of ninety-nine percent probability requires a twenty-one point spread (140 to 160). In plain English, an applicant who receives a score of 144 could have skills somewhere in the range of 137 to 151.
Thus, according to LSAC President Philip D. Shelton, the LSAT is good--but not that good as a predictor of future performance based on existing skills. In fact, according to LSAC, if students with an LSAT of 145 and 144 took the test a dozen more times, [LSAC would] have no idea which student would end up with the higher average score. Disturbingly, law schools know that the LSAT is not capable of making fine distinctions among candidates. LSAC has taken action to educate law schools on the proper use of the LSAT. LSAC has published articles such as Cautionary Policies Concerning LSAT Scores and Related Services, and The LSAT: Good--But Not That Good It sponsors a program that conducts conferences and trains law school faculty and staff on the appropriate use of the LSAT. Nevertheless, the misuse of the LSAT continues, pushing institutional racism into overt racism.
Third, the use of a cut-off score should be related to an applicant's ability to compete successfully in the particular school. While using a cut-off scores is not inherently invalid, courts have held that there must be a statistical, independent basis for the use of one minimum score rather than another. No such basis exists in most law schools.
a. Correlation Between LSAT and First Year Law School Grade Point Average Is Only Moderate at Best
Certainly, there is a correlation between LSAT and first year grade point average (FYGPA). Correlation, however, is not causation. The strength of the LSAT as a predictor of FYGPA can be measured through correlation coefficients that measure the linear relationship between two variables. It ranges from 1 for a perfect positive relationship to -1 for a perfect negative relationship. A positive relationship is one in which the increase in the value of one variable increases the value of the other. A perfect negative relationship is one in which the decrease in the value of one variable decreases the value of the other. The 0 means that there is no correlation between the two variables.
LSAC provides each school with the correlation coefficients based only upon the grade point averages of the school's students who have completed the first year of the program. It is also based on incomplete but useful information reported to LSAC.
At the University of Dayton, the average correlation coefficient for 1999-2002 between LSAT and first year school of law grade point average was .4265; between UGPA alone and first year school of law grade point average is .2725; taken together the coefficient is .50. In short, the LSAT alone accounted for fewer than half of the factors that were related to first-year performance. Other factors account for at least as much as LSAT/UGPA for achieving success in the first year of law school. When law schools deny admission primarily on the basis of LSAT/UGPA, they ignore the other factors that contribute to performance of excellent students and lawyers.
Beyond the correlation coefficients, LSAC conducts correlation studies for each school. Schools have kept these correlation studies a secret even from their faculties. Each school has its own correlation study, which are based upon the information provided by that school. Correlation studies calculate a predictive scale for performance using LSAT and UGPA. The 2002 correlation studies for the University of Dayton showed that even students with an LSAT as low as 135 and UGPA of 4.0 were predicted to perform at a 2.0 or higher level. Applicants whose LSAT and UGPA were even lower might have other qualities that were conducive to success in law school and should therefore have been given the opportunity to show these qualities in their application. Furthermore, while we do not want students on academic probation--for the University of Dayton that would be a FYGPA between 1.8 and 1.9--most students on probation will remove themselves from probation. So based on the 2002 correlation studies, applicants with an LSAT score as low as 135 and a GPA as low as 3.2 would still be in the range of successful academic performance. Thus, there is no relationship between the scores selected as the cut-offs and the level of acceptable academic performance at the University of Dayton School of Law. This same relationship can be demonstrated in correlation studies done at other law schools.
The most significant caveat is that the correlation studies are based on the performance of students who attend law school. If a school does not admit students with a certain LSAT/UGPA, then the correlation studies will reflect that. If a school has deliberately changed its admission standards, it is important to look at the correlation studies that preceded the change to determine whether or not the change was supported by the correlation studies. For example, the year before the University of Dayton raised its presumptive deny score from 139 to 145, the correlation studies indicated that applicants with an LSAT as low as 140 and a UGPA as low as 2.4 would still have performed satisfactorily.
In a letter written November 20, 2003, LSAC President Philip D. Shelton criticized the presumptive deny policy, writing: The [University of Dayton's] selection of a cutoff at 145 is particularly problematic because of its disparate impact on minority students. Looking at data for the Fall 2002 application cycle, 25% of all African-American applicants fell within the grid cell with LSAT scores between 140 and 144; 39% were above 145 and 36% were below 140. By making the cut at 145, an enormous number of African-Americans are eliminated from consideration by use of an admission tool that tells you so very little about the difference between those above and below that line. Shelton explained the arbitrariness of a cut-off for LSAT scores by writing:
Consider what the statisticians tell us about what happens with students whose scores are separated by ten points, or one standard deviation on the LSAT. If your school had 200 students, 100 with an LSAT of 155 and 100 with an LSAT of 145, we would expect 39 of the students with 145 . . . and 61 students with 155 scores [to be] in the top half. This represents, roughly, a 3-2 advantage for students with scores 10 points higher. Another way to visualize the issue is to consider a law school, such as the University of Dayton, which has two hundred students, forty of whom have LSAT scores of 140, 145, 150, 155, and 160. Assume that the UGPA is not a variable and the correlation coefficient is .50. Four of the students with LSATs of 140 will perform better than at least twenty-two of the students with LSAT scores of 160. Thus, working backward, a single point (145 over 144) is a nominal difference that does not justify a difference in admission decisions.
b. The Difference in Admission Process--Use of a Cut-Off and Failure to Consider Other Factors--Is Not Justified by Actual Performance of Students
Clearly, if students below the cut-off had a disproportionately high rate of failure, then actual performance would justify the cut-off. In most law schools, however, no such evidence exists. For instance, in 2002, the University of Dayton sent data for 156 students to LSAC. Of those 156 students, twenty-two would have been presumptively denied based on the 2003 admission policy of presumptively denying anyone with an LSAT of 145 or lower and a UGPA of 3.4 or lower. Of the 156 students, eleven were on probation or had been dismissed. However, only two of the eleven students in academic difficulty would have been in the presumptive deny category. On the other hand, sixteen of the twenty-two students had a C+ or better first year GPA. Furthermore, four of the twenty-two had a B or better average. When looking at performance, only eight of the twenty-two students' FYGPAs placed them in the lowest twenty-five percent of the class. Thus, in terms of actual performance, there was no justification for the practice of presumptive denial. Similar evidence should be available at other schools.
Upon examining a four-year period at the University of Dayton, specifically the years 1999 through 2002, there is clear evidence of competent academic performance. From 1999 through 2002, seventy-two students had LSAT scores below 145. Of those students, only five, or 6.9%, were dismissed and 12, or 16.7%, had a FYGPA of B- or better, on a C+ curve. Finally, at the University of Dayton, many students with LSATs as low as 138 graduated from law school, passed the bar exam, and became accomplished representatives of the legal profession.
c. The Difference in Admission Process--Use of a Cut-Off and Failure to Consider Other Factors--Is Not Justified by Actual Performance of Black Students
Studies have shown that Black students who participate in an appropriately structured academic support program perform successfully. For instance, the graduation rate of Black participants in AEP is seventy-six percent to eight-two percent for all LSAT groupings except for those whose scores fell into the 120-139 range. Other than this grouping, the second highest dismissal rate, which was nineteen percent, was not among the 140-144 LSAT grouping but the 145-149 LSAT grouping. These numbers are particularly significant because they show that only a small number of Black students failed, between two and four per LSAT category over a seven year period. In fact, only eleven Black students were dismissed during this period, an average of 1.5 per year.
TABULAR OR GRAPHIC MATERIAL SET FORTH AT THIS POINT IS NOT DISPLAYABLE
Not only was the graduation rate high, but the actual performance of Black students with LSAT scores below 145 was comparable to the class as a whole. The mean FYGPA was 2.326 for Blacks with LSATs below 145 and 2.380 for Black students with LSATs 145 or above. In other words, there was no appreciable difference in mean performance among Black students in the two respective categories. Furthermore, the averages earned were consistent with the University of Dayton's C+ curve--the standard against which all students are measured.
The median third year GPA was higher for the students with LSATs between 120 and 139. Furthermore, all categories of LSAT scores had median GPAs that were well above the minimum (2.0). A cut-off score is educationally justified when it yields an appropriate and meaningful inference about the applicant's successful performance in law school. In light of the gathered data, however, it appears that the mandated cut-off score of 145 could not have been based on denying persons who were incapable of graduating from law school. The lack of correlation between the designated cut-off score and its relationship to academic performance is the crux of the problem with law school admissions. Law schools have adopted an admission practice that is unrelated to academic performance and that has a disparate impact on the admission of Blacks.
Thus, where there is evidence of: (1) an admission practice that presumptively denies admission to a disproportionate number of Blacks; (2) LSAC correlation studies predicting that the students who are presumptively denied are capable of performing successfully; and (3) a historical record that establishes that students with LSAT scores that are below the presumptive deny can perform successfully, a policy and practice that denies admission to a disproportionate number of Black applicants based primarily on the LSAT without serious consideration of other relevant factors cannot be justified by claims of academic performance. While academic performance is a goal of educational necessity in that it serves a legitimate educational goal, this offers no support for the use of a presumptive deny category because the cut-off score is often times unrelated to that goal in most law schools. Furthermore, in most law schools, the cut-off score shows no demonstrable relationship to a student's ability to learn--a fundamental consideration of the courts when examining this issue.
2. Justification Based on Improving the Rate of Bar Passage Is an Educational Necessity, But the Use of a Cut-off Is Not Supported by Evidence. There Are Educational Interventions That Have a Less Disproportionate Impact
Under ABA accreditation rules, a law school has a responsibility to admit students who, with educational intervention, can successfully pass the bar. Thus, as an issue of accreditation, law schools have to be concerned about the number of graduates who pass the bar exam. Furthermore, it is unlikely that a school would want to admit students who are unable to practice law because they cannot pass the bar.
Most studies have shown a correlation between LSAT and bar passage. Educational necessity exists here because the challenged practice-- presumptive deny--serves a legitimate educational goal--assuring that students can pass the bar. Furthermore, the challenged action--presumptive deny--bear[s] a manifest demonstrable relationship to classroom education in that students who are admitted to law school affect the educational process.
Much of the concern about bar passage rate, however, is about first time bar passage rates. Studies show that, out of those graduates who do not pass the bar on their first attempt, over seventy percent pass on their second attempt and over twenty percent pass on their third attempt. Among those examinees of color who eventually passed, between 94 and 97 percent passed after one or two attempts and 99 percent passed by the third attempt.
The data show that among minority ethnic groups, some of whose members entered law school with academic credentials [UGPAs and LSAT scores] substantially below the majority of the admitted students, eventual bar passage rates ranged between 78 and 92 percent. These data provide positive support both for admission practices that look beyond LSAT scores and UGPA to define merit, and for a legal education system that adequately services students whose needs and preparations vary.
Again, correlation is not causation. Even after combining LSAT and law school GPA, the two most significant factors correlated with bar passage, studies have found a considerable amount of unexplained variance. Finally, while bar passage is a legitimate concern, there are alternatives to raising the rate of bar passage without resorting to the draconian measure of not admitting students based solely on LSAT and UGPA. 3. Justification Based on Desire to Improve Overall Quality of Class and Ranking of Law School Is Not an Educational Necessity Justifying Discrimination
In the early 1990s, U.S. News & World Report started ranking the top one hundred law schools. Schools have become obsessed with their rankings. According to some law school deans, pressure to raise their ranking in the U.S. News & World Report forces them to raise the median LSAT score. The median LSAT score, however, only makes up 12.5% of the law school rankings, and therefore this reaction does not have the desired effect.
Nevertheless, while improving the overall ranking might have other important purposes, it is not a legitimate educational necessity. The law school will not suffer any competitive disadvantages in this regard as a result of a higher or lower ranking. Law schools that have similar cut-off policies will face the same charges of discrimination and will be required to drop the practice. Thus, all law schools will be on equal footing.
Moreover, the Supreme Court has rejected use of cut-offs in an attempt to improve quality or ranking. For instance, in Griggs v. Duke Power Co., the Supreme Court rejected the position that, because a policy requiring a high school education serves the business purpose of upgrading the quality of the work force, such a policy is justified despite its discriminatory impact upon Blacks.
According to many deans, [t]hese ranking systems are inherently flawed because none of them can take [the student's] special needs and circumstances into account when comparing law schools. Law schools' policies and practices that use cut-offs that are virtually absolute are unacceptable because they, at best, assess a person in the abstract and ignore that person's ability to be successful. Consequently, improving or maintaining a ranking is not a compelling educational necessity.
4. Justification Based on Reducing Workload Is Not an Educational Necessity
Some faculty members complain that reading so many files is onerous and time consuming. It should go without saying that a discriminatory policy or practice should not be adopted for the sake of convenience. Faculty should not limit access and opportunity for racial and ethnic minorities because we are unwilling to do our job. Furthermore, as long as we invite applications, they should all receive the same careful consideration. It is a markedly different process to have a file reviewed and rejected by a faculty committee as opposed to a file reviewed and rejected by one person. Educational necessity does not exist here because the challenged practice--presumptive deny--does not serve a legitimate educational goal, since convenience is not a legitimate educational goal. Furthermore, the challenged action--presumptive deny-- does not bear a manifest demonstrable relationship to classroom education.
D. Given a Legitimate Educational Necessity, Are There Alternative Polices, Practices, or Procedures Which Would Have a Less Discriminatory Impact?
This question is relevant only when there is a legitimate educational necessity. If there is no educational necessity, then policies, practices, or procedures must be modified or abandoned. That is, it is not a remedy to continue the policies, practices, or procedures that discriminate by the adoption of mitigating practices. Of the cases offered above--assuring academic success, improving bar passage, raising rankings, and making the lives of faculty easier--only improving bar passage would be justified on the basis of educational necessity and evidence. Even for bar passage, however, there are alternative policies, practices, or procedures which would have a less discriminatory impact than the use of a cut-off. In this case, there is an educational solution rather than an admission solution to the issue of bar passage.
Sociologist Timothy Clydesdale examined rates of bar passage in 1991 and reports that differences in bar passage based on race
cannot be reduced to (1) academic preparation, effort, or distractions; (2) instructional or law-school-type characteristics; (3) social class; or (4) acceptance of an elitist legal ethos. Rather, results suggest that . . . minorities . . . confront stigmatization throughout legal education . . . [and] this stigmatization is continuous with prior socialization, making resistance difficult and consequent impact sizable.
In fact, using the Bar Passage study, Professor Clydesdale shows that even though bar passage is correlated to final grades, minority law students have lower grades than their white counterparts, even when controlled for LSAT, GPA, planned study hours, and other critical distractions. Although Black law students had the highest levels of self confidence at matriculation, they reported nearly twice as many experiences of race discrimination during law school as any other minority group. If entering LSAT and GPA do not account for the gap in final grades, perhaps stigmatization and discrimination do. Law schools could change their programs so as to reduce stigmatization. The most effective way of doing so would be to increase the number of each racial minority group to a critical mass. In addition, effective academic support programs that provide students access to the information stream necessary for success are essential. Another explanation for the disparate performance of minorities on the bar exam is stereotype threat. According to Claude M. Steele's theory of stereotype threat, members of stereotyped groups are especially wary of situations in which their behavior can confirm that stereotype. The extra pressure caused by the fear of reinforcing the negative stereotype interferes with performance, resulting in lower scores. Reducing stereotype threats should improve rates of bar passage.
Even so, there are other ways of improving the likelihood of bar passage without limiting opportunity to attend law school, such as teaching law students how to pass the bar. For instance, I conducted a supplemental bar passage course for sixteen high-risk graduates. Thirteen of these students passed the bar the first time; of the three who failed, one developed a chronic illness during the bar review, one had family commitments that interfered with his participation in the course, and the other student admitted that she froze during the test.
Impressed by these results, the University of Dayton created the Road to Bar Passage program, which raised the bar passage rate of our students with no change in admission standards. Our graduates placed third in the state of Ohio, with an eighty-six percent pass rate. In any case, a manifest relationship with no degree of certainty is not acceptable for establishing cutoff scores. In this case, the correlation between LSAT and first-time bar passage is not only weak, but also proves that educational remedies are more appropriate.
Schools often defend the discriminatory impact of their admission policies and practices by arguing that they will undertake other actions--such as increasing the pool of Black applicants and the yield from applicants with LSATs above 144--to raise the Black matriculation rate. Such actions or results will not ameliorate the discrimination that occurs due to a practice that presumptively denies admission to a disproportionate percentage of Black applicants.
[I]rrespective of the form taken by the discriminatory practice, an [institution's] treatment [of others] can be of little comfort to the victims of . . . discrimination. [The law] does not permit the victim of a facially discriminatory policy to be told that he has not been wronged because other persons of his or her race or sex [benefited]. That answer is no more satisfactory when it is given to victims of a policy that is facially neutral but practically discriminatory.
Schools that use an admission policy or practice that includes a presumptive deny process cannot rebut the presumption of discrimination where the school has no legitimate justification of the cut-off. Cutoff scores arbitrarily deny African-American applicants a fair opportunity to attend law school. Even if a law school could show an educational purpose, it must prove that the cut-off point selected was superior to any other. Most schools cannot provide such proof. What's particularly ironic is the assertion of some law school deans that [t]he idea that all law schools can be measured by the same yardstick ignores the qualities that make you and law schools unique, and is unworthy of being an important influence on the choice you are about to make. Surely the idea that all applicants can be measured by the same yardstick--namely LSAT/UGPA--ignores the qualities that make each applicant unique.