UC Faculty Report recommends continued use of SAT/ACT in admissions

[Mwfan1921]

Long awaited faculty report (by exec committee only) posted today, note report is preliminary. It goes to the full academic senate next, then final recommendations go to UC President Napolitano in April, who in turn makes her rec to UC Regents (due to vote on the issue in May).

Good summary here:

https://www.latimes.com/california/story/2020-02-03/uc-should-keep-sat-and-act-as-admission-requirements-faculty-report-says

Full report (it’s long), “Report of the Academic Council’s Standardized Testing Task Force (STTF)” accessible here:
https://senate.universityofcalifornia.edu/under-review/index.html

[Lindagaf]

I am in two minds about this. Interesting. Thanks for the links.

[ucbalumnus]

Looks like the report is suggesting that the HSGPA versus SAT relative prediction value has reversed since the Geiser studies, apparently because of HS grade inflation and HSGPA compression of admitted/enrolled students due to increased competition for admission.

[Mwfan1921]

@ucbalumnus

Looks like the report is suggesting that the HSGPA versus SAT relative prediction value has reversed since the Geiser studies, apparently because of HS grade inflation and HSGPA compression of admitted/enrolled students due to increased competition for admission.

Yes, but student success predictive model is highest when using both HSGPA and SAT/ACT (and looks like the essay might be nixed as an application requirement).

Models that combine HSGPA along with standardized test scores predict student success better than ones that only use one or the other. In contrast, the addition of the SAT/ACT writing tests did little to improve the prediction of student success above and beyond HSGPA and SAT Reading/Math and ACT composite scores.

A not insignificant piece of the analysis focused on explaining why underrepresented groups are underrepresented, and what role standardized testing plays in this. It is shocking how many in-state students do not complete the A-G requirements. Here are some select findings:

The conclusion from this analysis is that the comprehensive review admissions procedures at UC, of which standardized test scores and HSGPA are two components, is not the main determinant of the gap between the racial mix of high school seniors and the racial mix of UC admits.

The Task Force concluded that multiple factors contribute to this underrepresentation at UC. Roughly 75% of the opportunity gap arises from
factors rooted in systemic racial and class inequalities that precede admission: lower high school graduation rates for URMs, lower rates of completion of the A-G courses required by UC and CSU, and lower application rates. The most significant contributor was lack of eligibility as a result of failure to complete all required A-G courses with a C or better. Roughly 25% of underrepresentation was due to UC admissions decisions overall. Test scores play a role in those decisions, and thus account for some of that
25%, even if they are not the primary barrier to admission. It is also unknown to what extent low scores on the standardized tests deter some students from even submitting applications.

[bluebayou]

A not insignificant piece of the analysis focused on explaining why underrepresented groups are underrepresented,…

Which the politicians conveniently ignore when they compare the race/ethinicity of UC students to the state high schools. UC is designed to educate the top ~8% of the state’s high school grads. Unfortunately, the race/ethnicity mix of that top ~decile varies widely from that of high school grads as a whole, many of whom do not even complete a-g requirements to be eligible for Merced. Years ago, someone from UCI was interviewed about this disparity and he quoted the ‘UC eligibility’ numbers of high school grads by race. (part of which is due to test scores.)

[ucbalumnus]

@bluebayou

UC is designed to educate the top ~8% of the state’s high school grads.

Top eighth (12.5%) is the target, not top 8%.

UC eligibility for California residents was once defined with a sliding scale of GPA (2.8 to 3.3 or some such) and test scores, in addition to a-g subject completion. More recently, it is 3.0 GPA with a-g subject completion. (GPA as recalculated the UC way.)

[ucbalumnus]

Meanwhile, a University of Chicago study found that HS GPA was more important than test scores for predicting college performance (generally for Chicago area HS students going to any four year college, not University of Chicago specifically)…

http://talk.qa.collegeconfidential.com/college-admissions/2174485-uchicago-consortium-study-finds-high-school-gpas-outweigh-acts-for-college-readiness.html
http://news.uchicago.edu/story/test-scores-dont-stack-gpas-predicting-college-success
https://journals.sagepub.com/doi/10.3102/0013189X20902110

Perhaps it could be that whichever of HS GPA or SAT/ACT scores that a college emphasizes less becomes the more predictive factor, because the factor that the college emphasizes more gets compressed toward the top while the factor that the college emphasizes less is more widely distributed across the range among the college’s students.

An extreme example: suppose a college uses HS GPA only and admits only 4.0 students, but with any SAT/ACT score. Then HS GPA predicts nothing (because they are all the same), but SAT/ACT score has some predictive value for college GPA. At the other extreme, suppose a college uses SAT/ACT score only and admits only 1600/36 students, but with any HS GPA. Then SAT/ACT score predicts nothing (because they are all the same), but HS GPA has some predictive value for college GPA.