ANNUAL
REPORT OF THE COMMITTEE ON
ADMISSIONS
AND ENROLLMENT
2002
- 2003
To: The Representative Assembly, Davis
Division of the Academic Senate
This year the UCD Senate Committee on
Admissions and Enrollment focused on issues related to the campus selection
process for the freshman class. It met five times. The Committee also used
email exchanges to address certain critical questions it was asked to comment
on. As the Davis campus representative
on the Board of Admissions and Relations with Schools (BOARS), the chair
attended the nine regular BOARS meetings.
Additionally, the chair attended two meetings of the Undergraduate
Enrollment Workgroup and three meetings of the Enrollment Policy Advisory
Committee. The chair also served as an
ex officio member of the Davis Undergraduate Council, which met four times, and
the Executive Council that met nine times.
The principal results of the Committee's
work are:
During February, many of the Committee members participated in the read of selected applications from Fall 2002 admission as part of a study of the Comprehensive Review (CR) process initiated at the request of BOARS. Our report was presented at the February 19, 2003 BOARS meeting, and a draft of the BOARS report to the Academic Council was distributed at the March 18, 2003 BOARS meeting. The report does not mention any campus, nor does it present any data. It just outlines the procedures used and concludes that “the processes developed to implement the [comprehensive review] policy are characterized by a high degree of integrity and consistency, and that the admission decisions resulting from these processes are reasoned, consistent, and defensible.”
Members of the Committee attended an orientation and training session given by staff from Undergraduate Admissions and Outreach Services. The committee members did not attend norming sessions or additional training sessions usually required for readers.
Seventy admission files were selected by staff for Committee members to read. Twenty of the files were from applicants with high academic scores, 20 were from applicants with low academic scores, and 30 were from applicants on the border between admits and non-admits. The scores by the Committee members were the same as the scores from the actual comprehensive review 93%, 82%, and 18% of the time for applicants with low, border, and high academic scores, respectively. Two experienced staff read the files when there was a discrepancy between the committee read and the actual read. In reads of the low files, the committee read and the second read were always the same (no points awarded), while the actual read had awarded hits on 7% of the applications. The outcome of the committee read would not have affected the admission status of the applicant. In reads of the border files, the actual read awarded points in excess of the committee read 86% of the times, while the committee awarded more points than the actual read only 5% of the time. The re-read sided with the committee read 6% of the times and with the actual 94% of the time. In reads of the high files, the discrepancy did not arise from the lack of awarding hits by the committee, but from the areas in which the hits were attributed. In comparison to the actual read, the committee awarded fewer hits 41% of the time, the same number of hits 35% of the time, and more hits 24% of the time. Awarding these points would not have affected the admission status of the applicant.
The experienced staff readers attributed most of the discrepancies to the lack of training and the inexperience of the committee readers. Multiple readings of the low and high files did not affect the admission outcomes of the applicants. The UCD Admission and Enrollment Committee concluded that more time should be spent reading the border files where the awarded points could actually affect the admission decision.
The new ‘core’ SAT I examination is being rewritten and expanded to test an applicant’s competency in the areas of reading, writing and mathematics. The addition of the writing component to SAT I is new and would duplicate the writing part of SAT II as currently used. Previously, applicants were required to submit scores from the SAT II subject matter examinations in three areas: writing, mathematics or science, and social science or foreign language. Since the new SAT I will include a writing portion, BOARS has suggested that the three SAT II subject matter tests be reduced to two by the elimination of the writing component. The two SAT II subject matter tests would be chosen from the six curricular areas specified by the University’s ‘a-g’ course requirements. To minimize the impact of this change, BOARS has also proposed that the two SAT II subject matter tests be given equal weight to the previously required three SAT II subject matter tests.
On the face of it this is a very logical change. However, a more through analysis reveals some disquieting ramifications of the proposed changes. It is tacitly assumed that knowledge of a subject - irrespective of how that knowledge was acquired - will equate with the students performance at the University. Since all but one of the ‘a-g’ areas that are included in the SAT II subject matter are academic in nature, the test in reality not only measures the students knowledge of the subject, but also their ability to acquire that knowledge in an academic setting. All the ‘a-g’ subject matter tests are good indicators of the student performance at the University, except the ‘Language Other than English’ test taken by non-native English speaking students.
It is unlikely that a student has acquired their knowledge of mathematics, laboratory science, or history in a non-academic setting. Since teaching at the University is in an academic setting, prior learning in an academic setting would be a good indicator of the student’s ability to learn at the University. Learning a language at home or through trips to a foreign country may be an effective way to learn a language, but the knowledge so obtained has little relevance to the ability of the student to learn in an academic setting.
An inquiry was sent to the Chair of the Academic Senate from the Committee on March 14, 2003 asking for help in determining the faculty’s opinion about the value of the criteria we use for admission. The Committee has labored under the assumption that the faculty would not tolerate a significant reduction in academic indicators (i.e., HSGPA or SAT scores) to increase diversity. We know that diversity of the admits could be significantly increased by increasing the value of the non-academic indicators, but that this change would result in a lower average HSGPA and SAT scores for the class. The Committee needs guidance before suggesting changes that could significantly alter the statistics of the admitted class.
A talk presented by the Chair to the Executive Committee centered on two disturbing problems: increasing numbers of freshmen students taking remedial courses, and the conflict between maintaining high academic standards and maintaining diversity (i.e., percent of under represented minorities, URM) in the admitted students. The presentation to the Executive Council was to elicit comments on these two problems. While some of the Executive Committee were adamant that the academic index should not be allowed to decrease at all, most were willing to trade a small decrease (~1%) in average SAT scores for a comparable increase in percent URM. Results from Fall 2003 admission indicated that the percent of URM actually increased in the admitted class. Modification of the point distribution between academic and non-academic criteria may not be necessary to maintain the percentage of URM in the entering class.
The admission process at UC Davis involves the summation of points from a number of machine and reader scores. Applications are ranked by those scores, and a minimum score is selected such that enough applicants are admitted to insure that an enrollment goal is reached. It does not seem logical to do a read of the personal statement in the complete comprehensive review for those applications that will either be admitted or rejected irrespective of the outcome of the read.
The Committee considered changing the procedure so that applicants would be admitted if their score were above the minimum score without addition of the score from the personal read. An applicant would not be admitted if their score plus the maximum read score were below the minimum required. Applicants in these two groups would not benefit from a read of their personal statement, and the effort expended in reading the personal statement of these applications could be better spent to augment outreach programs or to do additional reads of marginal applicants.
An inquiry was sent to BOARS on March 11, 2003 about the possibility of eliminating some applications from a personal read during the comprehensive review of all applications. Applicants with more than enough points to be admitted without the addition of points from the read can be admitted without the read. However, since no applicant can be denied admission without a comprehensive review, applicants with less than the number of points needed for admission (even including all the possible points from the read, i.e., 1,250) cannot be denied admission without a read. The point hinges on what constitutes a comprehensive review. Since our ‘machine read’ includes a number of non-academic factors, we may be able to make the case that our comprehensive review may not include a read of the personal statement.
The following questions were raised at the BOARS meeting concerning our inquiry.
Is there a procedure in place so that readers of
the student’s statement can correct possible mistakes the student may have made
in entering data on their application?
An exhaustive
examination of the application is done when it is received to insure the data
presented is correct. This ‘front-end
review’ identified about 2,000 of the 32,000 applications to UCD in 2003 as
non-UC eligible. All of the remaining
30,000 applications were read.
Is there a procedure for the readers to bring to the attention of the admission staff an applicant they feel is exceptional?
No, but the readers
sheet can be modified to allow identification of ‘exceptional’ applicants.
The admission decision of only around 650 - 675 students are affected by our current selection procedure of awarding at most 1,250 points for the CR2 read. Most of the points were awarded for Leadership (27% of all admits received points in this category), talent (10.2%) and perseverance (8.2%). Admission is a ‘zero sum game’, so admitting a student because they received extra points from a CR2 read means that a previously admissible student would not be admitted. Therefore, the qualities in the CR2 read used to award points must be of sufficient value to the overall educational objectives of the University to justify exclusion of one student over another on the basis of the CR2 non-academic criteria.
There are definite overlaps among some of the CR1 and CR2 non-academic criteria. For example, 804 applicants (2.5%) received 1,500 points from EOP, Non-Traditional, UC Programs, and First Generation. The Committee is considering whether some of the criteria that are interrelated could be combined.
The Committee discussed how the admission procedure could be modified to better identify exceptional students with scores so low as to be non-admissible under the current procedure. Each campus is allowed to admit 6% of the class by exception. UCD currently admits about 3% (about 300 applicants) by exception. Admission by exception is currently limited to students who are not UC eligible. It may be possible to admit some reader identified exceptional students by exception, but such admission by exception may not be fair to other applicants. This matter is still being analyzed
We admit most of all ELC (Eligibility in the Local Context - the top 4% of each high school graduating class) applicants. The admission workload could be reduced if we admitted all ELC students. The discussion centered on how the selection procedure could be modified to reduce the number of admission folders given a personal read (i.e., CR2), while maintaining the transparency and fairness of the admission procedure. Around 98% of all ELC students are admitted under the current system that awards then 1,000 points. In Fall 2003, 4,318 ELC students applied to UCD. Only 78 were not admitted. Admitting all ELC students would not cause an undue burden on UCD, but it may admit a few students who are unprepared to succeed at UCD, and such arbitrary admission of one group may be unfair to other students. Increasing the points awarded to ELC students to 1,500 or 2,000 may accomplish the same thing if we can establish a threshold of CR1 points above which the CR2 read is not needed. The additional 4% (around 125 students) would not significantly alter the statistics of the admitted applicants.
The Committee considered modifying the admission procedure so that applicants receiving more than a set number of points from the CR1 would be admitted without the CR2 read of the personal statement. An analysis of the data indicated that there is an abrupt decline in admission from 100% to almost zero as the total CR1 points fall below 7,300 for L&S and Bio Sci, 7,600 for Agriculture, and 8,400 for Engineering. Admitting all applicants with CR1 scores above those values would reduce the number of files for the CR2 read from 32,465 to 13,161 for the Fall 2003 applicant pool. This procedure fits the intent of Comprehensive Review since no applicant would be denied admission without a complete comprehensive review. Additional analysis in being done before this new procedure is implemented.
Comments on Modifying the Course Major Articulation Agreement
The Committee discussed the UC Academic Senate Articulation Proposal to streamline the authority to articulate courses between UC Campuses and California Community Colleges. The Committee endorsed the proposal, but some felt it did not go far enough. The proposed streamlined process does not offer an enormous reduction in the workload associated with developing these agreements. During the discussion, the question was raised whether the number of campuses could be reduced from five to three to further improve the process. If 5 of the 8 UG campuses must okay each Course Major Articulation Agreement, then this only reduces the system-wide workload by 37.5%, even in the "perfect world" where no campus opts out of the otherwise-agreed-upon articulation agreement. Further, there is a measure of "overhead" associated with each of the 8 UG campuses reviewing, if not formulating, each agreement. Another possible problem with the proposal as written is that it encourages campuses to delay action, in the hope that 5 of the other campuses will do the work first.
Comments of the Part-Time Enrollment Proposal
The Committee was requested by the Senate to comment on the Part-Time Enrollment Proposal. The merits of the proposal were discussed and comments were solicited for inclusion into a memo sent to the Chair of the UCD Academic Senate.
This initiative would prohibit the classifying of any individual by race, ethnicity, color or national origin. There are provisions in the initiative for the State mandated collection of such information, and for the exclusion of medical research and the Department of Fair Employment and Housing from the provision of this initiative.
This initiative makes sense 1) if we are to live in a meritocracy (i.e., judging each individual on their merit and not on their race, ethnicity, color or national origin), and 2) if past racial based injustices can be remedied through equal opportunity.
The whole UC admission process is based on identifying applicants who will make the best use of their university experience and who will also contribute to the educational experience of their fellow students. We do not overtly use race, ethnicity, color or national origin in the current selection procedure. However, we do use racial characteristics to evaluate the impact of any modification to the selection procedure on the racial composition of an incoming class. The ability of various admission selection criteria (e.g., family income, first generation in college) to identify specific racial groups has been studied using racial data collected from previous classes. No criteria or combination of criteria has been identified that can substitute for information on an applicant’s race, ethnicity, color or national origin. Our inability to collect racial information would prevent us from analyzing the effect of selection procedures on the racial composition of the selected student population. If we wish to continue to have the ability to sculpt the racial composition of an incoming class through modifications of the selection procedure, then we will need to continue to collect racial data to examine and validate the outcome of revised selection procedures.
The Committee wishes to thank Darlene Hunter
for her administrative support, and Tom Lindholtz administrative analyst for
Undergraduate Admissions, for his expert assistance with data and analysis.
Christina Acham
Miroslava Chavez-Garcia
Karen Doyle
Mark Rashid
Mikal E. Saltveit, Chair
Gary Tudor, ex officio
Mary Bly, AF Representative