Pursuing Campus Diversity After Affirmative Action: An Assessment of Class Rank Plans for College Admissions
Colleges across the country have been searching for alternatives to traditional affirmative action programs to preserve diversity in the wake of hostile legislative initiatives and court decisions. Class rank plans, which guarantee admission to a fixed percent of high school students, have been adopted by public colleges across three states and debated by other public higher education systems despite the paucity of data on such plans.
In response to the Hopwood decision, Texas adopted a class rank plan beginning in 1998 that guarantees admission to any student who graduates in the top 10 percent of his or her high school class. California has instituted a number of steps to regain diversity in higher education, including a plan to admit into the University of California system the top 4 percent of students from each high school. And most recently, the "One Florida" plan calls for guaranteed admission for members of the top 20 percent of each high school class.
Under such plans, students are still required to complete a defined college prep curriculum and submit scores from admissions tests. However, differences in the academic rigor of high school courses, high school grades, test scores, and extra-curricular activities of students in the top percentile band for each state are not factored into the admission decision. The California and Florida plans do not guarantee that students qualifying under a class rank plan will be admitted to their first-choice college, but rather that they will be admitted to one of the state colleges or universities. The Texas plan ensures students will be admitted to their choice of state university or college, as well as their choice of major (with exceptions for Architecture and Fine Arts).
Class rank plans have primarily been adopted in an attempt to increase minority enrollment in colleges after affirmative action plans have been abolished. Minority enrollment at the University of Texas at Austin (UT, Austin) decreased from 38 percent in 1996 to 32 percent in 1997 after the state's affirmative action program was disbanded. A recent study by this university shows that two years after increased recruiting and outreach efforts were combined with the state's new top 10 percent plan, minority enrollment has generally rebounded from the 1997 low. The chart below illustrates that the fluctuations in minority enrollment between 1996 and today have been relatively small (Chart A on p. 26).
There is an absence of data examining the actual or potential impact of such class rank plans across geographic regions within a state and across states. Specifically, such plans are favored by many because they attempt to equalize admissions across schools within a state. A student ranked 10th from a graduating class of 50 would be guaranteed admission to a college in Florida under the proposed plan, just as a student ranked 50th from a graduating class of 250 would be, irrespective of other differences.
Measuring the Impact of Class Rank Plans
Clearly, class rank plans involve a number of complex policy considerations. Any major change in an entire state system's admissions criteria will affect: the numbers of students admitted, the diversity of incoming classes, the academic quality of entering students, college attendance across geographic regions within a state, out-of-state enrollment patterns, student performance in college (grades, persistence), and yield (the number of students who matriculate to a campus vs. the number of students accepted). Unfortunately, research to inform such changes in admissions has generally not been available to guide many of the discussions that have led to the implementation of these policies.
Data is beginning to emerge, however, on the impact these plans could have. Class rank plans can allocate a substantial number of available openings at a given college, leaving fewer openings for highly competitive students whose class rank does not meet the threshold. Data from UT, Austin illustrate that about 98 to 99 percent of students in the top 10 percent of their class for each ethnic group have been admitted in 1998 and 1999. That compares to an admissions rate of about 93 percent for White students, 90 percent for African American students, 91 percent for Asian American students and 92 percent for Hispanic students who ranked in the top 10 percent of their class between 1992 and 1996. The number of students in the top 10 percent of their class admitted to UT, Austin has increased by 77 percent from 3,796 in 1997 to 4,930 in 1999, while the overall enrollment remained virtually unchanged in these years.
The lower the class rank is set, the greater proportion of seats within an entering class will be filled based solely on class rank. That is, California's top 4 percent plan is likely to account for no more than 20 percent of freshmen openings at most campuses of the University of California. This is far less than the impact at Texas which has a 10 percent threshold. The impact should be greatest in Florida where a top 20 percent plan could account for an overwhelming majority of slots at some campuses.
Percentage plans are designed as an alternative to standard admissions models which combine high school grades, the rigor of courses completed, admissions tests and other personal experiences and extra-curricular activities. However, when such plans begin to account for nearly half of all admissions decisions as they are approaching in Texas and are quite likely to do at several Florida colleges, these approaches become the primary basis for college access within a state. It is essential that we measure carefully the impact these plans are having.
In the debates over traditional admissions practices, differences on test scores have attracted the greatest attention. Substantial differences also exist, however, between ethnic groups in class rank and grades. For example, nearly 28 percent of Asian American students and 23 percent of White students taking the SAT are in the top 10 percent of their class compared to 12 percent for African American students and 17 percent for Hispanic students. Females also comprise over 58 percent of students taking the SAT who are in the top 10 percent of their class.
Using any single measure such as class rank for a high stakes decision such as college admissions may result in outcomes different from those anticipated. First, these plans will limit access to college for students within a state who may not be placed in the top ranks of their school, but otherwise have stellar academic credentials. The Provost of the University of Florida has expressed his concern that middle-class minority students who may attend more rigorous and competitive high schools will be disadvantaged under these plans.
It may be possible to maintain the same levels of diversity with a class rank plan in states with some level of segregation across communities and high schools while increasing the odds of admissions for minorities attending rigorous schools. Again using Florida as an example, the media have reported that while it took a 4.32 GPA to be in the top 20 percent of one high school, a GPA of less than 3.0 was required at another high school.
Class rank plans also do not look at other measures of academic or non-academic success, and it is quite plausible that a student from one high school taking many honors courses and having higher grades and admissions test scores would not be admitted, while a student from a neighboring high school with less rigorous courses, lower grades, and test scores would be admitted.
Class rank plans are certain to result in a somewhat different pool of admitted students than current admissions models do. However, it is not certain that the students admitted under such plans will be more diverse nor adequately prepared for college. Such plans may not have the desired effect on diversity if they are not accompanied by aggressive recruitment and retention plans. Demographic profiles of the states will also have an impact on the results of such plans.
A study by UT, Austin shows that the SAT scores for students admitted as part of the top 10 percent plan decreased 26 points from 1997 to 1999 while the scores for students admitted through other means increased 20 points. In 1997, only 14 percent of students admitted in the top 10 percent of their class had SAT scores below 1100 compared to over 21 percent in 1999, and students in communications and education admitted under the top 10 percent plan actually had lower SAT scores than other students admitted to these majors. The UT, Austin study does illustrate that students admitted under the 10 percent plan in 1998 had higher freshmen grades than other students (3.26 vs 2.73). However, because these students are on average better prepared than students not in the top 10 percent of their class, this is not a meaningful comparison.
Generally, most students admitted under the current class rank plan likely would have been admitted under a holistic admissions model--most students who rank in the top of their class have taken more rigorous courses, have high grades and admissions test scores, and are likely to have substantial extracurricular activities. However, there is some proportion of this group who may be less prepared and would not have been admitted under a holistic admissions policy.
These policies clearly raise many questions and policy considerations. Additional research will surely be needed in order to craft fair and effective admissions policies in the future. For instance, research is needed that can differentiate groups of students with different levels of preparation. One method would be to compare the performance of students admitted today based solely on class rank to students previously admitted to the same institution and who also were in the top 10 percent of their class.
Similarly, one could differentiate students within the top 10 percent who would likely have been admitted under a more holistic approach from those who would not likely have been admitted under such an approach to examine the efficacy of the class rank policy. Such research would also be useful in determining if other types of institutional support may be needed for students entering an institution today who may not have gained admission in the past.
Finally, when high stakes decisions are made on the basis of only one indicator, whether it is a test score or a class rank, one must be attentive to efforts to influence the measure. With time, will such policies put added pressure on high school faculty to give even higher grades? Will families begin to consider opting for a less competitive or rigorous school district so that their children have better odds at acceptance in flagship state institutions? Will students be less inclined to take more advanced and honors courses if lower grades may be risked? Will class rank plans increase diversity at flagship public institutions or only equalize opportunities at less competitive state institutions? The systemic effects of any complex change in admissions policy is likely to require several years of data and research. However, data that are currently available can be used to aid policy makers in a number of these issues today and help them make informed inferences about the future.
back to top