UC needs to be transparent about student admissions
It’s the time of the year when the University of California’s nine undergraduate campuses release their fall admission decisions to hundreds of thousands of anxious high school seniors. Regardless of the decision, students and parents would agree the process is extremely arduous, stressful and even costly. It begins, in fact, in the freshman year of high school or even middle school, when parents and students need to set in motion a complex set of academic and nonacademic efforts to have a chance at just meeting the admissions criteria.
As I go through this process now with my eldest daughter, and despite her having been accepted to some universities and not others, we are left with questions. What more could we or she have done in the past three years? What factors on her application did the various university admissions teams consider?
For many parents, these questions may not deserve much mindshare: Their child will probably get admitted somewhere, if not at a UC. “It all works out” is a common phrase of hope uttered among commiserating parents.
I am not placated by this eventuality, though. I have two more kids with whom I will soon go through this process. I am proud of the academics offered by our state’s universities and colleges, but I am increasingly skeptical of their administrative bodies.
We, as parents, want to believe that playing multiple sports, taking music lessons and doing well in elective classes are all adding value. Yet, it is quite likely a waste of time and resources.
What is of concern is the toll this is taking on our students.
For most of the UC campuses, fewer than 1 out of 3 applicants is admitted; for UC Berkeley and UCLA the rate approaches 1 in 6. When you factor in that the UC system takes in about 23 percent out-of-state undergraduates, your child’s chances as a Californian are quite meager. In fact, for most of the UCs, a non-Californian applicant has higher odds of being accepted than a resident; conversely, California students have as high as an 80 percent acceptance rate at top public universities outside of the state.
While gaining admission to a UC campus is becoming “more competitive” (as the campus administrators say), the real issue is that UC is grossly misleading students on what it takes to get accepted. The reason is clear: It cannot meet the demand. For example, in 2018, UCLA (the most popular college campus in the nation) had 113,000 applicants, compared with 55,397 a decade prior and 32,792 in 1998.
We’re telling very smart kids that they are just not smart enough — which is ridiculous, if you compare them across national averages or even to past graduating classes. In the 1990s, 75 percent of UCLA admissions had a 4.0 grade-point average; now it is near 90 percent, and the top 25th percentile are fast approaching the unicorn 5.0 GPA.
UC Berkeley has explanations for this on its undergraduate admissions website — just the sheer number of disclaimers and hyperlinks is daunting. For instance, you read that due to the number of applicants, UC Berkeley selection criteria “exceeds the UC admissions criteria.” This is followed by various linked pages to criteria, such as weighted GPA, pattern of grades over time and test scores, as well as nonacademic criteria such as sports, volunteering and work. The information is bookended by a reminder that the applicant’s essays (each applicant writes at least four) are also considered. You can then visit a page about Berkeley’s “holistic” review process that purportedly factors in “personal qualities of the applicant, including leadership ability, character, motivation, insight, tenacity” and “likely contributions to the intellectual and cultural vitality of the campus.”
What UC is saying is very appealing: We’re not just looking at your GPA and test scores, we’re looking at many other factors. But how does that actually play out? If we were to analyze the students who were admitted, we should see a nice distribution across these various factors but, in fact, we see the opposite.
I can’t help but draw an analogy. About 10 years ago, I worked on a biotechnology that combines blood biomarker values to get a disease prediction more certain than produced by any single marker. For instance, while fasting blood glucose alone is a decent indicator of Type 2 diabetes, if we analyze other markers with it, we can predict your chances of getting diabetes five years earlier. Similar predictive models can help determine if a transplanted organ would be rejected or if a cancer drug would work on a specific tumor type.
It is important to know how each biomarker in the formula is weighted.
For instance, if there are four markers being analyzed and A is given a weighting of 1,000, while B, C and D are each weighted as 1, then regardless of the combinations, variable A will always drive the result.
This is the problem with the UC admission criteria.
It’s not clear what weighting, if any, UC gives the variables it claims to factor into the admissions decision. UCLA even seems to boast that its “criteria carry no preassigned weights.” Because UC publishes data showing how many applied, how many it admitted, and what the range of GPA and ACT/SAT scores were for those who get in, we can reverse-engineer the algorithm. And here is what we see: Despite all the variables UC says it considers, the vast majority of students are accepted on GPA and test scores alone. The other variables are what in the biotech world we call “junk markers.”
Junk markers are unethical because they drive up health care costs (along with collecting personal data for no good reason). If glucose is as good of a predictor of diabetes, then why should you pay for five other blood tests? Yet an “algorithm” diagnostic appears way more cutting-edge (and lucrative) than a half-century-old fasting glucose test.
Is the University of California system engaging in the same sort of marketing deception?
The UC system was founded with the intent to enable all qualified California students a higher education, yet most qualified applicants do not get in. Every year, the UC campuses move the goalposts closer together. The minimum bar for the ACT composite test score is no less than 26 at most UCs — yet the national average for high school students is just 21. At UCLA and UC Berkeley, the ACT composite score approaches a near perfect 34 (out of 36) at the 75th percentile of admitted students.
If UC was factoring all, or even just some, of the variables, you would not see a continuous rise in average GPA and test scores year after year. Yet UC continues to market a “holistic” admissions approach; this is as deceptive as charging a patient to test for markers that add no value to their diagnosis.
I am not saying that I regret my child participating in high school sports, music and summer internships. Those are great life experiences.
What I am saying is that if these factors do matter in UC admission, then make all the applicant data (depersonalized, of course) available for public analysis.
And if none of these other variables really matters, then stop saying they do — because as any parent of a high schooler will tell you, it would probably change the advice we give our kids when they ask if it’s more important they go to swim meet or study for their math quiz.