The Implementation of the Karabel Report on Freshman Admissions at Berkeley: 1990-1993

A Report of the Committee on Admissions and Enrollment of the Berkeley Division of the Academic Senate
In the spring of 1989 the Berkeley Division of the Academic Senate adopted the recommendations contained in what has come to be known as the Karabel Report. This report of the Senate Committee on Admissions and Enrollment was formally titled "Freshman Admissions at Berkeley: A Policy for the 1990s and Beyond" and has provided the framework for admissions decisions on the campus since it took effect in 1991. The general principles and specific recommendations contained in the report and adopted by the Senate were sweeping in their scope. (See Tables 1 and 2.)
Inasmuch as some of the policies adopted represented innovations in Berkeley admissions practice, the Berkeley Division of the Senate requested that the Committee on Admissions and Enrollment report back to the Division on their implementation when sufficient experience had accumulated to suggest how they worked. The complexities of admissions at Berkeley are so great and the implications of the Karabel Report are so extensive that one could easily say that the Committee on Admissions and Enrollment in the ensuing years has done nothing except prepare for the implementation of the Report's principles and recommendations and then refine them. Thus this report construes the mandate of the Division broadly and covers all aspects of the work the Committee on Admissions and Enrollment has done over the years 1990-1993.
David Leonard, Chair
David Auslander, Mechanical Engineering
Ken Simmons, Architecture
John Polt, Spanish and Portuguese
Eleanor Swift, Law
Chi-yuen Wang, Geology and Geophysics
Andre Bell, Ex-officio
Patrick S. Hayashi, Ex-officio
Larry Natividad, Student
I. The Karabel Report Principles
II. General Evaluation
III. The Matrix System
IV. Affirmative Action for Socio-Economically Disadvantaged Students
V. Affirmative Action Concerns: Filipino and Latino Applicants
VI. Special Action Admissions of Students at Risk
VII. Extension of the Karabel Principles to the Professional Schools and Colleges
VIII. Reform of the Fall Extension and Defer-to-Spring Program
IX. Junior Transfers
X. Special Challenges to Excellence through Diversity
I. The Karabel Report Principles
  The policy of the University of California at Berkeley is to maintain high standards of academic accomplishment and broad diversity in its undergraduate student body. As a teaching and research university of international renown, Berkeley gives priority in admission to students with exceptional academic accomplishments (Principle 1). At the same time, Berkeley must strive to serve all of California's people by training the future leadership of a remarkably diverse state (Principle 2) and should seek to create a stimulating educational environment by recruiting a student body that represents a broad diversity of backgrounds, values and viewpoints (Principle 3).
  The experience of the Admissions and Enrollment Committee in administering these three principles of the Karabel Report has been that together they serve the goal of excellence in undergraduate education. Over the last decade Berkeley has been able to recruit an ever more accomplished and talented group of students while simultaneously becoming more diverse. This is an achievement of which the State, the University and the campus can be proud.
  The more specific principles set by the Berkeley Senate to guide the admissions process firmly establish that the provision of educational excellence for undergraduates is not achieved by a single criterion. For example, outstanding accomplishment is defined broadly to include achievements in the arts and in athletics in addition to academics (Principle 5). Diversity in social background is defined broadly as well, to include those who contribute the special perspectives of socio-economic disadvantage; of cultural, ethnic and racial heritage; of geographically different upbringing; and of age and disability (Principles 3, 5 and 6). And most important, the admissions process is grounded on the belief that the best assessments of contribution to the educational environment and the best predictions of future achievement include human, qualitative judgments (Principle 8). After 50% of our admission target is admitted on the basis of high school grades and test scores, remaining admissions are based on the careful reading and assessment of thousands of individual application files. Working closely with the Office of Undergraduate Admissions, the Admissions and Enrollment Committee has promoted policies that have resulted in the reading of more than 6,000 files for the Class of 1992, a significant increase from the past.
  Such attention to the individual applicant could be avoided were Berkeley admissions to be determined solely by "the numbers" -- by the grades and test scores that have served to make the applicant eligible for the University of California in the first place. But the Berkeley Senate has, in adopting the Karabel Report, eschewed such simplicity in its admissions process. The principles adopted by the Senate acknowledge the complexity of the task of admitting as our students those who will contribute to and benefit from the ideal of excellence in Berkeley's undergraduate education. The Karabel Report recognizes that the individual qualities of the applicant matter and that, therefore, applicants with certain qualities, talents, and backgrounds should receive special attention in the admissions process. The Matrix System, described in Section III below, has been developed over the past three years to take account of all these special factors, in addition to "the numbers," that will create a student body that best serves, and is best served by, educational excellence at Berkeley.
  Table 1
1. As an institution of international renown and as one of the nation's leading research
universities, Berkeley has an obligation to admit students with exceptionally distinguished
academic records.
2. As a taxpayer-supported public university, Berkeley must strive to serve all of
California's people.
3. Berkeley should actively seek diversity -- socioeconomic, cultural, ethnic, racial, and
geographic - in its student body.
4. Berkeley will absolutely not tolerate quotas or ceilings on admissions or enrollment of
any racial, ethnic, religious, or gender groups.
5. In its admissions criteria, Berkeley will recognize outstanding accomplishment in a
variety of spheres, including (but not limited to) art, athletics, debating, drama, and music.
6. While continuing to grant preference to California residents, Berkeley will continue to
admit out-of-state students.
7. Berkeley should accept only those students who have a reasonable chance of persisting
to graduation.
8. The admissions process should include a human element and must not be based on
grades and test scores alone.
9. In constructing and altering Berkeley's admissions practices, the faculty should insist
upon at least a co-equal role with the administration.
10. The admissions criteria and practices of the College of Letters and Science as well as
those of the Professional Schools should continue to be described in detail and to be made
fully available to the public.
  Table 2
1. The proportion of the fall freshman admits selected by academic criteria alone should be
increased from 40 to 50 percent.
2. Eliminate the second tier of the current admissions policy, which admits students on the
basis of academic index scores and supplementary points.
3. A new Tier 2 should be established consisting of the old "complemental" categories
from Tier 3 as well as three new categories; the 45 percent of the fall freshman class
admitted into Tier 2 will be selected through a process of "secondary review."
3A. A new secondary review category of students who come from socio-economically
disadvantaged backgrounds should be created; this category should be open to all
disadvantaged students, regardless of race or ethnicity.
3B. A new secondary review category of mature or "re-entry" students will be created.
3C. A new secondary review category of students whose academic index scores narrowly
missed gaining them admission into Tier I should be created.
4. The number of Special Action admits should not exceed 5 percent of all fall freshman admits,
and the number of Special Action registrants should not exceed 6 percent of all fall registrants.
5. Berkeley should continue to offer qualified UC-eligible students who are not admitted into the
fall freshman class the option of applying for spring enrollment; these applicants should be
selected primarily on the basis of academic criteria, though Berkeley's commitment to a
diverse student body should also be taken into consideration in determining whom to admit.
6. As part of its effort to extend the process of diversification, Berkeley will need better data on
the socio-economic composition of its applicant pool as well as of the state's graduating seniors;
in cooperation with other UC campuses, Berkeley should, therefore, formally request that the
appropriate state agencies provide it with the data that it will need to carry out its policy of
admitting a socio-economically diverse student body.
II. General Evaluation
  The campus can take great pride in the overall admissions picture at Berkeley. Both the accomplishments and the diversity of new freshmen at Berkeley have increased substantially over the last eight years.
  The academic accomplishments that Berkeley freshmen bring to the campus from high school have improved dramatically and steadily since 1986. Table 3 presents the mean high school grade point averages of new freshmen over the last dozen years. Between 1980 and 1985 these high school grades improved from 3.60 to 3.66; by 1992 they had jumped to 3.83. This significant improvement was reflected in high school performance of every ethnic group represented at Berkeley.
  Scholastic Aptitude Test scores of Berkeley freshmen also have improved significantly, both in absolute numbers and relative to California and national averages. Since 1978 the mean SAT totals of Berkeley freshmen have risen 100 points, and the margin by which they exceed state and national means has increased by 100 points as well (Table 4). As is demonstrated in Table 5, every ethnic group has participated in the improvement in test scores since 1980. Most have increased dramatically since 1985, when admissions became significantly more competitive at Berkeley; only Chicanos have been static in their test performance over this most recent period.
  A further indicator of academic accomplishment is how the students do once they arrive at Berkeley. The National Collegiate Athletic Association now evaluates campus success by examining the percentages that have graduated within six years. Of those who entered Berkeley as freshmen in 1980, 68% persisted to graduation. The 1985 class, the most recent one for which we have this long time-series data, already shows improvement on this measure to 75% (Table 6).
  Furthermore, most of those who drop out of Berkeley do so in their first two years of study, and for this measure we have more recent data. Twenty percent of the members of the freshman classes of 1984 and 1985 had dropped out of Berkeley by the start of their junior years. The average drop-out rate for those entering in 1988 and 1989 had decreased to 14.5%, a decline of 28%. These improvements are seen in all ethnic groups. (See also Table 7.)
  As pleasing as the improvements are in persistence on the campus, the Committee on Admissions and Enrollment believes that these figures underplay Berkeley's real success with its undergraduates. The image of an undergraduate as one who starts and finishes her education at the same institution is substantially less apt now than it was 25 years ago. Today many students leave Berkeley, not because of academic failure, but simply because they want to continue their studies elsewhere - for a variety of sound personal and academic reasons. In addition, some students leave in good academic standing because of family or financial pressures. The Committee believes that a more appropriate measure of Berkeley's success is the percentage of a freshman class who have a passing grade point average (2.0 or above) when they leave the campus, whether by graduation or to go elsewhere. Students who leave the University prior to graduation but with a GPA of 2.0 plus have kept their financial aid eligibility intact and can readily transfer to other institutions to complete their education. The freshman classes of 1984 and 1985 averaged 85.8% on this measure of success. The campus has witnessed a steady improvement in this type of success in the ensuing years, so that the average for the classes which entered in 1988 and 1989 stood at 90.4%. All of Berkeley's ethnic groups have achieved a decline of at least 30% in the rates of this kind of failure over this time period. This is an impressive accomplishment.
  The Committee has conducted a special study to confirm that all those admitted to Berkeley do have a reasonable chance of success, as mandated by the Karabel Report. Figures 1 and 2 plot students' UCB cumulative grade point averages against their Academic Index Scores (high school grades plus test scores). It can be seen, first, that all registrants, no matter how weak their formal qualifications, have been selected in such a way as to average better than passing grades at UCB. Second, at the point at which UC eligibility ends there ceases to be a relationship between the Academic Index Score and UCB grades. (The cut-off for UC eligibility generally falls in the range of 4,300 to 5,000 points.) In the lower reaches of formal qualifications students are admitted only if a qualitative evaluation of their applications indicates that they nonetheless have a good chance of success at Berkeley. Figures 1 and 2 demonstrate that these qualitative judgments are being well made. They also indicate that any tightening in the formal criteria for admission would yield little or no improvement in student success at Berkeley. Berkeley is fortunate enough to have such a well-qualified group of applicants that only intensive, qualitative admissions reviews can produce a better student body.
  Concurrent with Berkeley's increased academic achievement has been its heightened diversity. The largest amount of this change has occurred by itself. The part of the admissions process that is completely blind with regard to race and ethnicity has produced increasing numbers of Asians on the Berkeley campus. Thus, whereas Asians (excluding Filipinos) were only 19.6% of the freshman class in 1984, they were 37.1% of the 1992 class (Table 8). The reasons for this shift have been several. The first is demographic. While whites have been decreasing as a proportion of California's high school students and graduates, Asians have been increasing. Thus the California Postsecondary Education Commission (CPEC) indicates that whites fell from 68% of the 1981 graduating public high school seniors in California to 55% in 1990, while Asians rose from 6% to 14%. The second factor causing heightened diversity concerns preparation. The CPEC indicates that all groups improved their eligibility for UC by grades alone from 1983 to 1990. But whereas white eligibility increased from 7.7% to 12.7%, Asian eligibility rose from 17.3% to 32.2%. The combined effect of the demographic and preparation factors is that Asians represent twice as large a proportion of the UC-eligible pool today as they did a decade ago. The third factor in heightened diversity is economic. The incomes of most white families in the state are higher than those of Asians. Thus white high school graduates are more likely to apply to and be able to attend private universities and colleges than are their Asian counterparts. (We should note in passing that the "Asian" category is an exceedingly broad one and includes within it quite different ethnic, demographic, economic and preparation dynamics for the different Asian sub-groups.) Thus, the largest part of Berkeley's diversity is attributable exclusively to the strong academic records of Asian American applicants and is a major strength of the University.
  Berkeley has also been able to maintain the approximate proportions of underrepresented minorities in its freshman classes while simultaneously improving its academic standards. This achievement has been the result of diligent recruiting work on the part of the Office of Undergraduate Admissions and Relations with Schools (OUARS). It has not been the result of quotas, which the policy of the University expressly forbids. (See Table 1, Principle 4.) The numbers of any given racial or ethnic group admitted vary from year to year depending on the variation in the quality and quantity of their applicant pool. Thus, although Berkeley admitted 916 African American freshmen in 1987, it admitted 671 in 1991. This drop partly reflected a decline in the percentage of African Americans in California's high schools, similar to that of whites. Similarly, although the numbers of Chicanos at Berkeley are increasing, they are not growing as rapidly as their percentage in the state's schools. Eligibility for the University of California is lowest in this ethnic group, and -- with the largest proportion of poor in the state -- it has been most severely affected by the rise in the University's fees. Thus, although the picture generally has been improving, it is not all that we would like it to be. Nonetheless, the several parts of the admissions process operate to assure both that the campus remain diverse and that those admitted to Berkeley have a strong probability of being successful here (as is evidenced by the above statistics on UCB grades).
  It is misleading to cite only statistics about freshmen in talking about diversity at Berkeley. Undergraduates come onto the campus at other times and in other ways than through admission to the freshman class in the fall of any given year. Many are admitted as transfer students, usually having studied first at a community college. Others arrive in the midst of their lower division studies, having been granted a deferred admission to Berkeley when they first applied for the freshman class. Those taking these other routes onto the Berkeley campus also are quite diverse, but underrepresented minorities are less likely to be among them. Tables and 10 indicate the diversity of the Berkeley undergraduate population as a whole, compared with the diversity of the fall freshman class in 1992 shown in Table 8. Thus in 1991, for example, while African Americans and Chicanos represented, respectively, 7.8% and 13.1% of the freshman class, they constituted 6.8% and 9.1 all undergraduates. Similarly, in 1992 Asians (including Filipinos) at 40.5% were the largest group in the freshman class, followed by whites at 30.8%. But the order is reversed among all undergraduates, with whites representing 36.5% and Asians 33.6%.
  Table 11 presents the percentages of the various transfer (advanced standing) students admitted by ethnicity. Note that the proportions of non-whites entering the campus by this route is much smaller than those coming in through freshman admission. It is a concern of the Committee on Admissions and Enrollment that those community colleges in California that serve predominantly African American and Chicano student bodies do substantially less well in preparing them for admission to the University of California than do wealthier, suburban colleges. The solutions to this problem, however, seem largely beyond Berkeley's control.
  We turn now from this general evaluation of admissions at Berkeley to a more detailed treatment of the ways in which the principles of the Karabel Report have been implemented.
III. The Matrix System
  Starting with the Fall 1992 admission process, the Office of Undergraduate Admissions (OUA - now OUARS) used a matrix format for considering freshman applicants to the Colleges of Letters and Science and Natural Resources. (The Colleges of Chemistry and Environmental Design use a similar process; Engineering includes a high degree of faculty review in its decision-making.)
  The Committee's decision to adopt the matrix was based on a variety of factors. The matrix clarifies and facilitates the application of Berkeley's selection criteria, enables OUARS to identify the academically strongest students more effectively, and thus produces more refined admissions decisions. The matrix process also is easier to explain to prospective students and their parents -- overcoming a difficulty with the Fall 1991 process often noted by admissions and outreach staff -- and thus strengthens relationships with secondary schools, campus communities and the public. Furthermore, the matrix approach corrects the appearance (never the reality) that "Karabel category" applicants were being reviewed in exclusive groups without competing with other applicants. The matrix format does not change the substantive principles established in the Karabel Report. Instead, use of the matrix simply alters internal OUARS procedures and the way the campus describes them.
  To build the matrix, OUARS assigns all freshman applicants an academic score (the horizontal axis) and places them on a social diversity scale (the vertical axis). OUARS then arrays the applicants on the matrix shown as Figure 3. For Fall 1992 and 1993, the academic score was based on bands of the Academic Index Score (AIS), which has been used by OUARS since 1984. The AIS multiplies an applicant's high school grade point average (capped for all students at 4.0) by 1,000 and adds to it the student's scores on the SAT and the three required achievement tests, for a total of 8,000 possible points. The social diversity scale reflects the approximate weight assigned to various attributes by the Karabel Report, including ethnic underrepresentation, low socio-economic status, geographic origin (California residency, rural area), athletic recruitment, age (over 24), special talent and disability. No single attribute can earn an applicant an "A" rating.
  The first step in the admissions process is the development of the admissions target by the Budget and Planning Office in consultation with the Provosts, the Committee on Admissions and Enrollment and the OUARS. In recent years this has been about 7,200.
  All freshman applicants are arrayed against each other on the matrix. Then, in keeping with the recommendations of the Karabel Report:
1) Fifty percent of the admissions spaces are filled on the basis of the AIS alone. (These are the cells labeled "Admit.")
2) Next considered are "Special Promise" applicants, those with Academic Index Scores immediately below those granted admission via the AIS alone. Approximately 5% of the admission target is admitted from this group. Two or three times as many applicants as will be admitted in this group are selected for detailed reading of their files at this stage. Due to the similarity of the academic records of these applicants, final selection is based on qualitative reading and scoring of their essays, activities, honors, awards and employment. (These cells are labeled "Read" in the matrix.)
3) Admission is granted to the top students in the social diversity rankings. (These cells also are labeled "Admit.")
4) Several thousand additional applicant files are reviewed in great detail, including having their essays read and their activities, honors and work experience evaluated. (These cells also are labeled "Read" in the matrix.)
  Approximately 6,000 applicants for Fall 1992 had their files read in great detail. The essay reading team of 40-50 people is composed of staff from the admissions office, faculty, selected high school counselors, and volunteers from other administrative offices and academic support services.
  Depending on the outcome of the "Read" reviews, applicants are admitted or denied. Those who are denied are offered other admissions options, including consideration for admission to the following spring semester and guaranteed admission in the junior year if a minimum set of grades are maintained in junior college. Those in the matrix cells labeled "Options" are denied admissions for the fall without a full reading of their files, but are also offered the alternative admissions options (such as consideration for admission in the spring). Applicants in the "Deny" cells are denied admission and are not offered other options.
  The matrix procedure enables OUARS and the Admissions and Enrollment Committee to focus more clearly on the choices required to build a freshman class that strengthens Berkeley's goal of providing excellence in undergraduate education. The matrix approach improves our ability to compare applicants to one another across a wide range of academic and social-diversity profiles. In addition, the matrix approach clearly compares applicants to each other along a continuum rather than in distinct groups.
  During the actual selection process, a series of comparative judgments are made by assessing the number and quality of those in a matrix cell against the enrollment target range set for that group by the Karabel diversity goals. Based on these assessments and a series of comparisons at the margins between groups, the final selections are made. The numbers admitted in any given diversity group will vary from year to year. In some years the numbers admitted will be greater or less than the Karabel target ranges for particular categories. During the matrix review process the OUARS staff regularly confers with the Admissions and Enrollment Committee to fine-tune the Karabel target ranges as well as the relative priority to be given to the applicants in the various cells of the matrix.
  The admissions policies that underlie this procedure balance a wide range of competing and equally valid interests and goals in a complex public and legal environment. We believe that the result is a process that meets our educational goals, is as understandable to our public as is possible, and is in the interests of the greatest number of Californians.
IV. Affirmative Action for Socio-Economically Disadvantaged Students
  The Karabel Report also mandated attention to diversity with regard to those who come from socio-economically disadvantaged backgrounds, regardless of race or ethnicity. (Table 2, Recommendation 3A.) However, it did not specify how this category should be defined and operationalized. The Committee on Admissions and Enrollment devoted considerable attention to this question in both 1989-90 and 1990-91. It decided that:
(a) disadvantage with respect to higher education was a function not only of income but also of parental education and occupation.
  At the urging of Berkeley administrative and faculty representatives, the systemwide application was revised, adding a question on parental occupation and placing the questions on occupation, income and parental education in a place where they were more likely to be answered.
  The next step was to identify what constitutes disadvantage on these matters. The Committee decided that:
    (b) a parent with a managerial or professional occupation is a sign of relative advantage, as is a parent with a college degree. A family with an income at or above those of comparably situated families (as defined below) also is advantaged.
  The Committee deliberated at length as to what is the relevant base of comparison for family income for Berkeley applicants. Family incomes vary considerably over the life time of their senior members, being low when wage earners are young adults and when they retire. Most families approach their peak incomes at about the time when their children are old enough to go to college. Thus even if the poorest children had the same chance for a college education as the richest, the median income of the parents of college students would be higher than the median of all family incomes.
    (c) The Committee undertook a study of the income of California families with children between 17 and 19 years of age and found that the median in 1988 was $39,635. 1 Disadvantage with regard to income was then defined as parental income at or below this mark.
  (As a matter of interest, approximately 28% of Berkeley's undergraduates come from families at or below median income by this measure. 2)
    (d) An applicant was defined as socio-economically disadvantaged for admissions purposes if he or she was not advantaged on at least two of the three indicators used. A secondary category of disadvantage was established for those whose family incomes were below $50,000 (just shy of the Berkeley freshman median) and were disadvantaged by one of the other two measures as well.
  The Committee decided to discontinue this secondary category beginning with the Fall 1993 class because the more stringent definition is producing more than enough highly qualified applicants. 3
  As is evident from the foregoing, affirmative action for applicants who are socio-economically disadvantaged was not implemented all at once. Data from the systemwide application were available to make the decision as to who was disadvantaged beginning only with the 1991 class. As the category was a new one, the Office of Undergraduate Admissions and Relations with Schools used it conservatively that year, and unintentionally undershot the Karabel target range. Also many socio-economically disadvantaged applicants who were members of under-represented minorities were admitted instead on the basis of their ethnicity that year. For Fall 1991, 347 students were admitted out of consideration for their socio-economic disadvantage. 4 Another 557 socio-economically disadvantaged students were admitted through regular admission procedures, so that the group represented 11% of all admits. (See Table 12.)
  For Fall 1992 the Committee had more confidence about the utility of the category and mandated that the Karabel target range be reached if sufficient qualified applicants were available. It also decided as a matter of principle that applicants should be considered for admission under this race-blind criterion before being evaluated under any other affirmative action criterion. As the original Karabel targets were based on the assumption that socio-economic disadvantage would be the final, rather than the first, category for affirmative action, the targets were adjusted upward accordingly. Thus for the Fall of 1992, 1,330 freshmen were admitted who were socio-economically disadvantaged, 464 because of special consideration for this fact (Table 12).
Table 12
  Admission of Freshmen from Socio-Economically Disadvantaged Backgrounds
  Fall 1991 Fall 1992 Fall 1993
Low SES5 by Attribute      
Total Low SES Applicants 1640 2607 2965
Total Low SES Admitted 904 1330 1367
Low SES as % of All Admits 11.0% 15.3% 16.5%
Total Low SES with SIR6 371 603  
Low SES by Category      
Considered Under Category 576 722 N.A.
Admitted Under Category 347 464 N.A.
SIR 165 247 N.A.
  For Fall 1992, as noted earlier, freshman admissions were based on a matrix, rather than categories, sorting applicants by both academic and social/diversity criteria and accounting for applicants who satisfied multiple social/diversity criteria. The matrix approach carefully followed the principles of the Karabel Report. Nevertheless, the OUARS retained a count of admissions by category to monitor unanticipated differences between the two approaches. OUARS also relied upon an attribute count, whereby applicants could be and often were assigned more than one attribute. The sum of applicants by attribute, consequently, exceeds the total number of individuals in the applicant pool. Category counts were not maintained in 1993.
  The number of applicants identified as socio-economically disadvantaged increased from 1,640 to 2,965 between Fall 1991 and Fall 1993 and the number admitted with the attribute jumped from 904 to 1,367, representing 16.5% of all admits in 1993. The primary reason for the increase in numbers stems from limitations in data collection, rather than actual changes in applicant volume or admissions patterns. For Fall 1991, UC Systemwide was unwilling to key enter information on parental occupation. Berkeley staff could identify parental occupation only by individually reviewing files and then entering a separate code on the database. Given time limitations, OUARS staff did not review files of applicants who might have been disadvantaged, but who would have been admitted in any event (e.g., students admitted by Academic Index Score alone, and student athletes, rural students or under-represented minorities who were admitted by AIS within their category).
  Some observers of the admissions process have been understandably confused by the difference between counting by attribute and counting by category. Since Berkeley now uses a matrix approach rather than a category one and because category counts are confusing, an important task for OUARS and the next Committee on Admissions and Enrollment is to revise Karabel target ranges to report attributes rather than categories.
  The Committee is pleased with its experience with this form of diversity. It enriches the experience of all students both to have a range of socio-economic backgrounds represented on the campus and to have these backgrounds scattered across all ethnic groups. As a matter of principle, the Committee is convinced that all forms of socioeconomic disadvantage should be considered in the admissions process, not just those associated with race and ethnicity. The Committee also believes that nonracial measures of disadvantage are ethically more desirable and it would be preferable if Berkeley could eventually be race-blind in its admissions. Unfortunately, race adds to other forms of social and economic disadvantage in America today and is not simply a function of them.
  Finally, we have found that the number of highly competitive applicants who are socio-economically disadvantaged still exceeds Berkeley's capacity to admit them. The Committee wishes to continue to admit freshmen under this category in approximately the same numbers as it did for 1993. The Committee did devote one meeting to the testimony of a faculty member who would like to see affirmative action for the socio-economically disadvantaged decreased or eliminated. We are grateful for his thoughtful input, but were unpersuaded by his arguments. If any change in numbers were to be made in the future, we would want to see them increased, not decreased.
V. Affirmative Action Concerns: Filipino and Latino Applicants
  Crucial to our concept of excellence in undergraduate education at Berkeley is diversity and participation by disadvantaged groups. Following the Karabel Report, this Committee holds that, to be considered for affirmative action at Berkeley, a group should be under-represented, relative to its proportion in the population of University-eligible California high school graduates, both at Berkeley and in the University of California system as a whole and should be subject to widely acknowledged disadvantage in the society at large. Disadvantage is indicated by a group's rate of eligibility for admission to the University of California. The University is committed by the Master Plan to provide access to the top 12.5% of California high school graduates. According to the Office of the President, "The primary barrier to access to the University ... is the low rate at which students from some ethnic and racial groups attain eligibility for University admissions." 7
  Although whites are proportionately under-represented at Berkeley, they are not eligible for affirmative action on the basis of race because they are neither underrepresented in the rest of the UC system nor subject as a group to disadvantage. The latter is indicated by the fact that in 1990 over 20% of white high school graduates in California had the potential to achieve UC eligibility compared to a state average of 18.8%.8 Of course, many whites have a unique contribution to make or are disadvantaged for other reasons, and we give them affirmative action consideration through our socio-economically disadvantaged, rural, athlete, special talent and disabled categories.
  After lengthy deliberation, in January 1988 this Committee voted to end special consideration for Filipinos. To mitigate unanticipated effects on potential applicants, the Committee instructed the Admissions Office to implement its decision over a five-year period. Effective with the Fall 1993 cycle, Filipinos no longer received special consideration (based on ethnicity) in the freshman admissions process.
  Filipinos have attained a rate of UC eligibility above that of the statewide average. The California Postsecondary Education Commission reported for 1986 that 19.4% of Filipino high school graduates met UC-eligibility requirements or would meet them by taking the required standardized tests. The UC-eligibility rate for Filipinos exceeded that of all public high school graduates (14.1%) and that of all other groups except for Asians (32.8%). Additional indicators support the Committee's action. Filipinos comprised 3.7% of Berkeley's undergraduate student body in Fall 1992 while comprising 2.9% of 1991-92 California public high school graduates. 9 Moreover, Filipinos at Berkeley are not significantly disadvantaged by income or parental education. The median income of Fall 1992 Filipino applicants of $58,789 exceeded the Berkeley campus median of $56,000 and was higher than the median of all ethnic groups other than whites. 10 Furthermore, 84% of Filipino registrants reported that at least one parent had earned a four-year college degree, compared to 72% of all Fall 1992 new freshman registrants. 11 Seventy-one percent of Asian registrants and 89% of white registrants reported that at least one parent had a four-year degree. Of course Filipino applicants who are socio-economically disadvantaged or who demonstrate other social-diversity attributes reviewed in the matrix approach continue to receive additional consideration.
  The Committee and OUARS also have been reviewing special consideration for Latinos, following the release of the University's "Undergraduate Student Affirmative Action Five Year Plan" in 1990. That document estimated that Latinos had comprised 7.6% of Berkeley's new freshmen from 1987 to 1989 compared to 3.4% of 1987 California public high school graduates, bringing into question the continued necessity of special consideration for this group.
  Reviewing the available data on Latinos, we encountered some of the difficulties involved in considering race in the admissions process. Data are limited. Neither CPEC nor the Department of Finance distinguish Chicanos from Latinos in their basic reports on high school graduates. The University's affirmative action plan had estimated Latino high school graduates by projecting from the proportion of Chicanos and Latinos in the general population -- an approach leaving a significantly high level of uncertainty.
  Latinos as a group at Berkeley report socio-economic characteristics which differ on the one hand from groups with historically low UC-eligibility rates (i.e., African Americans and Chicanos) and on the other hand from groups meeting or exceeding UC-eligibility rates (whites and Asians). The median parental income for Fall 1992 Latino freshman registrants of $50,000 was well above the $31,000 median for Chicanos but below the $56,000 median for all freshmen. For Fall 1992, 63% of the Latino freshman registrants responding reported a parent with a four-year college degree, again comparing favorably to Chicanos (28%), but unfavorably to the freshman class as a whole (72%).12 Perhaps because of these socio-economic conditions, Latino applicants have not demonstrated achievement levels, measured by grades and standardized test scores, equivalent to whites or Asians.
  In response to this information, the Committee decided that generalized affirmative action for Latinos should end. But it further decided that:
UC-eligible California Latinos who are socio-economically disadvantaged will be placed in the Disadvantaged read pool regardless of their AIS score and will be admitted on the basis of that evaluation.
  In this way Latinos will retain a small amount of affirmative consideration relative to Asians and whites, but this consideration will be focused on those who are genuinely disadvantaged.
  The issues related to affirmative action for Latinos and Chicanos and the distinctions between the two groups are being reviewed by the University-wide Task Force on Latino Admission Eligibility. The Committee will continue to review its decision with regard to Latinos as new, more reliable information becomes available.
VI. Special Action Admissions of Students at Risk
  The Karabel Report established the principle that "Berkeley should accept only those students who have a reasonable chance of persisting to graduation" (Tables 1, 7). We noted above that the Committee on Admissions and Enrollment has interpreted this principle marginally to mean that a student should be carrying a GPA of 2.0 when he or she departs the campus, whether through graduation or transfer. The Committee has asked that those making admissions decisions satisfy themselves that there is a high probability of achieving this minimum standard of success whenever a student is granted admission to Berkeley. Although this criterion affects all parts of the admissions process, operationally it has been most important to Special Action admissions.
  Under Regental policy guidelines each campus of the University of California is permitted to enroll up to 6% of its new freshmen from those who do not meet the minimally established combination of test scores and high school grades necessary to be eligible for admission to the University. Applicants considered for admission under this policy are subjected to a particularly intensive qualitative evaluation by the Special Action Committee, which is appointed by the Chancellor. In recent years the Special Action Committee has shared the concern of the Senate Committee on Admissions and Enrollment that young people not be set up for failure by being admitted to Berkeley when their chances of surviving here are not good. Thus the Special Action Committee has increased the intensity of its reviews, with the positive effect of continuing improvement in the performance of Berkeley undergraduates noted above.
  Another way of assuring the success of undergraduates is the provision of support services for students who are at risk. The best endowed support program is that run by Intercollegiate Athletics (in part out of its own funds) for student athletes. This program is a good example of what it is possible to achieve. Despite the substantial commitment of time that participation in varsity sports requires, student athletes at Berkeley do better than non-athletes with the same test scores and high school grades (Figure 4). As a consequence, Intercollegiate Athletics and Berkeley's NCAA academic representative have been allowed a good deal of discretion in recruiting student athletes and recommending them for Special Action admissions, a discretion they have used responsibly. The Committee on Admissions and Enrollment wishes that the same quality support services were available to all undergraduates. But the general support program also is a good one, and in the current budgetary climate it is unlikely that additional funding will be available for it. This unfortunate reality reinforces the importance of being clear-headed about the chances of success that various applicants have.
  The foregoing speaks to the minimum standards used in the Special Action process. But in a period in which admissions to Berkeley have become (and will continue to become) ever more competitive, minimum standards are hardly enough. Historically the Special Action Committee has tended to treat the Regentally established maximum numbers for this category as if they also were a minimum. A serious anomaly had thereby arisen whereby an applicant with special mitigating circumstances was more likely to be admitted if he or she failed to meet University eligibility standards and was put into the Special Action process. Beginning with the class admitted in Fall 1992, the Committee on Admissions and Enrollment mandated that the Special Action Committee, in its review of applicants, also had to consider comparable UC-eligible students who would otherwise be denied admission. Special Action is still free to make the admissions decisions according to whatever qualitative standards it has established, but in doing so it must satisfy itself through reading actual applicant files that there are no UC-eligible applicants who are more attractive than the non-eligibles who are at the top of the Special Action list.
  This change in Special Action policy has created an increase in workload for that committee, which does pose a potential problem. More faculty are needed to volunteer to help with the reading of files on the Special Action Committee. But the Committee on Admissions and Enrollment is convinced that the change in admissions principles here is appropriate. The long-term consequence is likely to be a reduction in the proportion of freshmen who were not UC eligible. Already in the fall of 1992 those who were not UC eligible fell to 3.8% from 4.9% among 1991 admits (and to 5.2% from 6.5% among those who registered). Nonetheless, Special Action admissions will continue to be necessary, especially for athletes, if Berkeley is to remain competitive in NCAA Division I competition. It is worth noting here that, contrary to popular opinion, special consideration for athletes does not particularly favor under-represented minorities. White freshmen constitute a far larger percentage of recruited athletes than any other ethnic group.
  The blurring of the distinction between eligible and ineligible applicants marks two important developments: (1) a recognition that applicants on either side of a line developed by quantitative measures alone are more similar than dissimilar and (2) the placement of all applicants in the same review process via the "matrix."
VII. Extension of the Karabel Principles to the
Professional Schools and Colleges
  The Office of Undergraduate Admissions and Relations with Schools makes only a portion of the admissions decisions at Berkeley. This is overwhelmingly the largest portion, but Ethnic Studies, the College of Engineering and the College of Environmental Design do conduct their own admissions processes. This Committee has asked these schools to conform to the Karabel principles. Ethnic Studies and Environmental Design have been notable for their contributions to diversity at Berkeley (although the numbers admitted to the former are exceedingly modest). Engineering has the highest admissions standards on the Berkeley campus. All three units have intensive faculty involvement in their admissions decisions. Their processes therefore already produce qualitative judgments of high worth about likely student success. All three units assure the Committee that they are doing their utmost to promote the diversity that is consistent with their standards. As these are faculty committees making these decisions, this Committee defers to their judgment.
VIII. Reform of the Fall Extension and Defer-to-Spring Program
  The Fall Extension Program began in 1983. It was offered to Fall freshman applicants who had been denied admission to the College of Letters and Science for Fall but who had been offered deferred admission to the Spring term. The campus proposed the Fall Extension option as a way to offer Fall course work to a portion of these students so that they would not lose time-to-degree. That first program in 1983 enrolled 34 students. In 1984 enrollment jumped to 178 students and then to 305 in 1985 and to 500 in 1987, a level where it has since remained.
  All UC-eligible freshman applicants who are denied admission to the College of Letters and Science for the Fall semester (typically about 8,000) may request consideration for deferred admission to the Spring semester. Each year about 2,000 of these students ask for such consideration, and between 1,200 and 1,400 of them are admitted, based on their Academic Index Scores. These admitted students may also request consideration for the Fall Extension Program (and each year about 600 students do so). Because of the high degree of selectivity for the Spring semester, the admitted students rank in the top 60% of the Berkeley overall freshman applicant pool.
  The Defer-to-Spring option, including the UC Fall Extension Program, has enabled the campus to maintain a steady number of students throughout the year, to stabilize its Spring intake of new freshmen, and to strengthen the academic quality of the Spring freshman class. In addition, the Defer-to-Spring option has reduced political pressure -- especially from alumni-caused by the intense competition for Fall admission. (Each fall Berkeley turns away about 3,000 applicants with nominal 4.0 grade point averages.) A good-but-not-great student who really wants to come to Berkeley has a good chance to do so through the Defer-to-Spring option, and the UC Fall Extension Program makes the deferral relatively easy to accept.
  The Fall Extension Program is robust. The Fall 1992 program enrolled 502 students, and demand for the Extension option has remained strong. In addition, course offerings have been expanded over the last few years, and a growing number of students have been accommodated in University residence halls.
  Most observers agree that the program is the academic equivalent of the educational experience that most new freshmen enrolling in the regular Fall semester are likely to have. Classes are generally smaller, the curriculum is carefully structured, and the academic environment may be warmer and less intimidating than that of the larger campus. Student evaluations consistently reflect a high degree of satisfaction with the program, and, in general, students do well both academically and personally once they enroll as regular Berkeley students.
  There are, however, some criticisms. First, students in the program have not been eligible for federal and state financial aid programs. The program has been able to provide some limited scholarship funding, but in general these monies have not amounted to significant financial aid awards. Second, until recently, students have not been eligible to live in campus residence halls during the Fall semester. Although that has changed somewhat, Extension students are only accommodated in residence halls after demand from regularly-enrolled students has been met. Third, students in the program have had restricted access to student services and activities. At present, participants are eligible for some services and activities (e.g., library access), pay additional user fees for others (e.g., the student health service), and are excluded from still others (e.g., intercollegiate athletics, fraternities and sororities).
  Extension participants also cite a fourth criticism: the lack of ethnic diversity in the program. Currently the program in almost entirely Asian American and white. That is largely because the Fall freshman applicant pool contains relatively few African American, Chicano, Latino, and Native American applicants and because most of those who do apply are admitted to the regular fall semester. From the Fall 1992 applicant pool, for example, only six of the 1,531 students admitted to Spring were from these ethnic groups, and none of them chose to enroll. The lack of financial aid eligibility may well have contributed to the non-enrollment of such students: the median annual family income of Chicano students enrolled in Fall 1992 as freshmen at Berkeley is $31,000 compared to $77,000 for whites.
  The Admissions and Enrollment Committee believes that there are strong academic and social reasons for increasing the diversity of the Fall Extension Program. Extension staff report that many instructors and students believe that their educational experience would be improved by exposure in the classroom to a broader range of experiences and viewpoints. Many of these students have chosen Berkeley specifically because of its reputation for diversity and are disappointed that their initial experience here does not provide that richness.
  At the request of the Committee, the OUARS and Associate Vice Chancellor for Admissions and Enrollment Patrick Hayashi have developed a marketing and recruitment strategy for Fall 1993 freshman applicants who will be offered the Defer-to-Spring option for Spring 1994. This experimental approach will identify and target those students for whom the nature and relative cost of the Extension Program would be particularly attractive. The targeted group will include students from all ethnic groups, but a particularly strong effort will be made to enroll African American, Chicano, and Native American students in the Fall Extension Program.
  Toward this goal, Richard Black, Director of Financial Aid at Berkeley, reports that his office has recently convinced the U.S. Department of Education that the Fall Extension Program meets the federal guidelines for eligibility for federal student aid funds, beginning with Fall 1993. Building on that approval, Director Black hopes to convince the California Student Aid Commission to allow use of Cal Grants as well. If this effort is successful, students in the Fall Extension Program will have the federal and California student aid sources available to them. University extension also is considering expanding its own grant program. Should this be done, students in the program would have the same aid resources available to them as regularly-enrolled undergraduates at Berkeley.
  The Committee will continue to monitor the program closely and will consider other program or policy changes as may be required to achieve the goal of making the Fall Extension Program reflect the rich diversity of the regular Fall semester freshman class.
IX. Junior Transfers
  The Karabel Report governed only the admission of freshmen to Berkeley. However, a third of those who enroll at Berkeley come in as transfer students, generally at the junior level. One of the major tasks of the Committee on Admissions and Enrollment over the last two years has been to devise a comparable set of principles to govern the transfer admissions process. The recommendations of the Committee for action by the Senate are being presented in a separate report authored by Professor Norton Grubb, chair of the Committee for 1991-92. Since the full report is available to the Senate, we need mention here only some of the highlights in the proposed changes. The Grubb Report will result in: a tightening of admission standards for junior transfers, an opening up of the possibilities for transfer by students at four-year colleges, and a reduction in the restrictions on transfer applicants' choice of prospective majors.
  The Committee is convinced that the changes proposed in the Grubb Report will reduce the complications of the transfer process for potential applicants and will lead to an increase in the quality of transfers (most of whom are already quite good). The Committee hopes that there may be some improvement in diversity as well, but this is less certain. By and large, disadvantaged minorities are much less well represented in the transfer applicant pool than they are among freshman applicants. Community colleges in the state are not serving well this part of their constituency. The Committee hopes that by having a more open and less complicated transfer process, Berkeley may be able to improve its recruitment of under-represented minority transfers somewhat.
  Nonetheless, given the financial uncertainties facing community colleges at the moment, only marginal improvements in diversity among junior transfers are likely. It appears that under-represented minorities are so heavily recruited at the freshman level that those who remain in the community college system need exceptionally good services if they are to become good prospects for Berkeley. Apparently they are not now getting these services and are not likely to get them in the near future.
  The fact that the ethnic and racial mix of transfer students is so different from that of freshmen makes the general public's fixation with the fall freshman admit and enrollment figures quite misleading. It would be much better to focus attention on the extent of diversity in the overall undergraduate student body at Berkeley and to urge the public to see fall freshman admissions as part of the whole.
  A further matter affecting the community colleges is the Cooperative Admissions Program. One of the options extended to UC-eligible applicants who are denied admission to Berkeley is to attend an approved community college, with a guarantee of junior admission to Berkeley if the community college GPA is at least 2.4. This minimal GPA is now well below our other admissions standards and the community colleges have complained that it does not motivate those affected very well and seems unfair to the other students.
  The Committee now has decided that when the option is offered to UC-eligible applicants this spring of 1993, the required minimum level of performance in an approved community college will be raised to a GPA of 3.0. In the spring of 1995 this minimum will be further raised to 3.3.
X. Special Challenges to Excellence through Diversity
  In general, Berkeley is doing a commendable job of promoting excellence through diversity in its undergraduate education. In terms of academic achievement, no public university in the United States admits a better qualified freshman class. Indeed, we believe that no other university in the country, public or private, rejects as many applicants each year as does Berkeley. Similarly, in terms of diversity, other than UCLA, no university of remotely comparable quality has as ethnically mixed a student body.
  We believe that the present blending of past attainment and diversity accomplished through the matrix process is near to being the best we are likely to achieve. This is not to say that there is no room for improvement in our admissions policies. We ourselves have been engaged in an intensive reexamination of many of them and have made a number of changes where the evidence suggested they were warranted. We would expect our successors on this Committee to be no less committed to careful, empirically-grounded evaluation and change.
  But the major factors governing the shape of the undergraduate student body at Berkeley are beyond our control. The broad parameters of our mandate to pursue excellence through diversity have been set by the state legislature and the Board of Regents, governed by the current interpretation of the U.S. Constitution. The mix of upper and lower division students on the campus also is fixed by the State Master Plan for Higher Education as it has been reaffirmed by the legislature and The Regents. The constraints on the resources for recruitment and support services are largely external as well. Even these constraints are the minor ones, however. Society itself sets still more stringent parameters on the possibilities of our action.
  There is a serious crisis in the high school education of African Americans in California at present. The proportions of this group eligible for University of California admission is low, particularly among males. Although those fully eligible for University admission have improved to 5% among African Americans, this is well below the average of 12.5% eligible. 13 The fading ideal of integrated schools, the inadequate resources for urban schools, the severe unemployment rates for African American young adults, and the effects of the drug trade have all conspired to create ever-diminishing motivation and opportunity for African American high schoolers. We at Berkeley can and should try to lean against these elemental forces, but in large measure they are beyond our direct control and explain the decline in African American numbers at Berkeley.
  The eligibility proportions of Chicanos have improved somewhat as well. But at 3.9% these proportions remain low and are declining relative to other groups. The number of Chicanos in the state's high schools is increasing rapidly. Thus the numbers of this group who are qualified to attend Berkeley will probably increase. But as the proportion of Chicanos increases in the high schools, it may be difficult for Berkeley's admissions to keep up, and the proportion of Chicanos at Berkeley relative to their numbers in the secondary school system could decline further. Because of its low income and Spanish-language background, this group needs strong support services in high schools, the financial resources for which are not keeping pace with their increasing numbers.
  Finally, the numbers of Asian and white students at Berkeley are driven by the dynamics of demographics and the cultures of immigration. Whites are a declining proportion of high schoolers in California, and because they are relatively advantaged, they have many opportunities to go elsewhere -- and do. The numbers of Asians in the system are increasing, and this is reflected in their presence at Berkeley. But not all Asian groups are represented equally on the campus. The category of "Asian" itself is impossibly wide, representing over two-thirds of the world's peoples. Different Asian groups are over-represented or represented in equal portions to whites at Berkeley. The vagaries in this part of our ethnic mix are produced by of a combination of merit, alternative opportunities, and desire for the Berkeley experience. This is as it should be as long as no serious disadvantage is involved. The ethnic mix of our campus will shift back and forth over time. The appropriate response for us as Berkeley faculty is to enjoy the intellectual stimulus that this change and diversity bring to us. Certainly it has brought to us the best student body we have had in the last two decades.
  Berkeley admissions processes are not perfect; they are in need of constant attention and improvement. But they have achieved the goal of excellence mandated by the Karabel Principles. And they are close to being as good as we are likely to be able to get, given the society in which we live.
1 Data are from the March 1989 Annual Demographic File of the Current Population Survey. Analysis was provided by the staff of the Survey Research Center Data Archive.
2 Tom Cesa, "Parental Income Data for New Freshmen at UC Berkeley," Berkeley: Office of Student Research, October 22,1992.
3 We have provided preference for applicants with family incomes between $39,635 and $50,000, but only when they were not advantaged with respect to parental education or parental occupation, thereby meeting the test of being disadvantaged on two out of three criteria. Some applicants were included who reported parental education equivalent to an AA degree at a community college.
4 The figure of 347 is obtained by using a category count, whereby applicants are assigned to a single category whether or not they satisfy the criteria for more than one category. The assignment was determined by a hierarchy consistent with the listing of categories in Table 7 on page 48 of the Karabel Report.
5 Low SES refers to those of low socio-economic status, i.e., those who are socio-economically disadvantaged .
6 SIR means Statement of Intent to Register.
7 University of California, "Undergraduate Student Affirmative Action Five Year Plan: 1990-1995" (Oakland, 1990).
8 California Postsecondary Education Commission, "Eligibility of California's 1990 High School Graduates For Admission to the State's Public Universities. A Report of the 1990 High School Eligibility Study" (Sacramento: March 30, 1992).
9 California Department of Finance, "K-12 Public High School Graduates by Ethnicity, History and Projection: 1992 Series."
10 8.5% of Filipinos did not report income, while 22% of freshman applicants in general did not report it. It is believed that applicants not reporting income have family incomes well above the mean.
11 Calculations exclude those not reporting data.
12 Comparable rates for whites are 89%, Asians 71%, and Filipinos 84%.
13 California Post-Secondary Education Commission.

Please send any comments or questions regarding this web site to our Web Site Administrator (


Committee Ref 
Academic Planning and Resource Allocation