AP Use in Admissions: A Response to Geiser and Santelices | 1 AP ® Use in Admissions: A Response to Geiser and Santelices 1 Wayne Camara and Michalis Michaelides March 11, 2005 Overview In their 2004 paper “Te Role of Advanced Placement and Honors Courses in College Admissions,” Saul Geiser and Veronica Santelices of the University of California, Berkeley, address the use of Advanced Placement Program® (AP®) and honors courses as a criterion for admission at the University of California system and suggest that the policy for awarding bonus points to such courses “has little, if any, validity with respect to the prediction of college outcomes” (p. 24). Tey find that the number of AP or honors courses taken is not a statistically significant predictor of college outcomes, while performance on AP Examinations is strongly related to college performance. Policy of Awarding Extra Weight to AP, Honors, IB, and Concurrent Community College Courses In 1982, the University of California instituted a policy of awarding one bonus point to AP and honors courses taken in the last two years of high school. Similar policies of considering advanced courses in college admissions exist in other institutions, particularly selective ones. Admissions officers may consider students’ advanced-level courses by examining the number of these courses on transcripts and/or through the bonus weight given to them in the calculation of the high school GPA, as in the case of the University of California. Te authors identify a number of reasons why this policy is an important issue that deserves scrutiny. Some of them are discussed below: (a) Access to AP or honors level courses may not be equal for all students; there may be disparities, ofen related to socioeconomic variables. Results from Table 1 in the paper show that the disparities in access to advanced-level courses at the school level are “not as great as perhaps might be expected” (p. 8); for example, schools in the upper API quintile offer on average 14.5 AP courses as opposed to 10.2 AP courses offered by the lowest API quintile schools. No statistical tests are carried out to compare advanced-course offerings across the school API categories. Results from Table 2, which is based on the self-reported responses of the California SAT® population, indicate that the representation of minorities and less advantaged populations in the categories with five or more “AP/Honors subjects taken” is less than that of their counterparts. About 19 percent of the sample report taking five or more advanced-level courses; 26.3 percent of the sample take one to four such courses, and less advantaged subgroups are slightly overrepresented compared to their share in the overall sample. Tey are also slightly overrepresented in about half of the sample that reported no AP/honors course work. However, the authors cite research from the CSU Institute for Education Reform that points to schools’ internal policies for AP participation (e.g., tracking) rather than the availability of the AP courses as a reason for the observed disparities. (b) Te policy itself may encourage schools to offer more rigorous courses, and students to enroll in them, but the resources available to schools may not be adequate to ensure high quality of these courses; or students may casually take the course, without evidence of mastery of the material—in the case of AP courses, for example, enrolled students may not take the end-of-course exam. For admissions purposes, enrollment in such advanced- level courses during the senior high school year suffices, because admissions applications and decisions are made before the end of those courses. Tus, there is no control over student performance and no guarantee that the student had a “truly” rigorous, college-level experience in the course. (c) More importantly, the authors cite the lack of research on the validity of advanced-level courses as an admissions criterion. Te AP Program was developed to enable placement into sequent college courses and/or for granting college credit, and research continues to support this use (e.g., Dodd et al., 2002; Morgan & Crone, 1993; Morgan & Ramist, 1998); however, the use of AP, IB, honors or concurrent community college course work in admissions decisions has not been validated. Te authors’ attempt to fill the gap on the predictive validity of such courses in the literature is commended. It should be noted that the authors’ purpose is to examine the use of AP and other honors courses in admissions to the University of California system with data from the particular, selective population of students who enrolled in the system. The Predictors Multiple predictors are used to build the models presented in the study. Most of them are known to be positively correlated: 1 We฀would฀like฀to฀thank฀Rick฀Morgan,฀Neil฀Dorans,฀Shelby฀Haberman,฀and฀the฀AP฀Program฀staff฀at฀ETS฀for฀their฀input.