How UC Admissions Changed After Going Test Blind

It's good news for weak applicants and bad news for strong applicants and those from private schools.

The University of California stopped requiring or considering standardized tests for admission in 2021. How has this affected applications and admissions?

I’m going to compare the most recent three years (i.e. Fall 2022 to Fall 2024) with the last three pre-pandemic years (i.e. Fall 2017 to Fall 2019). I’m going to ignore 2020 and 2021 because these were affected by the pandemic¹.

I’m also going to focus purely on applicants from California high schools (both public and private) because the system of A-G certified regular and honors classes makes their academic records more comparable. Additionally, the UC’s Eligibility in the Local Context and Statewide Guarantee initiatives mean that international and out-of-state applicants are held to higher standards.

Academics vs Storytelling

The factors UCs consider in evaluating applications can be grouped into three categories:

• Academics: classes taken and grades obtained

• Geography: where you live and where you went to school

• Storytelling: what you write about yourself in your application essays and extracurricular activity list. These are all self-reported and not verified so can be labeled as storytelling².

Standardized scores, when they were required, could be considered part of academics or a separate fourth category.

The UC publishes three types of data that speak to the academic quality of its applicants: the number of A-G classes taken, the number of honors classes taken, and the grade point average (GPA). Unfortunately, the only GPA number published is UC’s weird weighted, capped GPA and the comparability of grades across high schools is questionable. For example, there are some high schools in San Francisco where the number of ‘A’ grades exceeds the number of students who meet standards and others where the number of ‘A’ grades is considerably lower than the number who meet standards.

I’m therefore going to focus on the number of honors classes taken. This is obviously only a partial measure of academic quality because it doesn’t include any grade information but “rigor of secondary school record” is considered by all UCs³ to be a “Very Important” criterion for admission. UC publishes the number of applicants, admits, and enrollees who fall into three buckets:

• 10 or more Honors classes taken,

• 5.0 - 9.9 Honors classes taken,

• 0 - 4.9 Honors classes taken.

For clarity and convenience, I’m going to refer to the students who fall into these three buckets as strong, intermediate, and weak applicants respectively.

What should be the effect of going test blind?

One possibility is that the applicants’ academic record will become more important to fill the void left by the lack of standardized scores. Intermediate or weak applicants with strong SAT scores will now get rejected. Applicants with a strong academic record but low SAT scores will now get accepted. Overall, we would see a higher admission rate for stronger applicants and a lower admission rate for weaker applicants. Those with weaker academic records would get squeezed out.

Another possibility is that storytelling will become more important. No data is ever provided about the quality of applicants’ stories. UC publishes a breakdown of admissions by sexual orientation or gender identity (which in theory should have no effect on admission) but not on the quality of an applicant’s essays (which is a “Very Important” criterion everywhere) or even on something as concrete as whether a student would be the first in his or her family to attend college. In the absence of data, if storytelling becomes more important this will manifest itself as greater randomness. Weaker students will start getting admitted at higher rates.

Let’s see which actually happened.

UC Applicants Keep Getting Better

High school students are engaged in an academic arms race in a bid to stand out for college admissions. They’re not getting any smarter, and no one has claimed that the schools have improved enormously, but they are taking much harder course loads than their predecessors. The number of UC applicants is growing steadily but the number of strong applicants grew much faster from about 2012 on, and then jumped again in 2021 when the UCs first went test-blind.

Taking 10 or more honors classes used to be rare but, by 2024, these strong applicants constituted 27% of the total applicant pool. Intermediate applicants, at 42%, remained the largest group. Weak applicants made up the remaining 31%.

The composition of the applicant pool also changed in ways that I found surprising. Among public school graduates, there were increases in the number of strong and weak applicants but not in the number of intermediate applicants. I attribute the growth in strong applicants to the long-term strengthening of the applicant pool and the growth in weak applicants to the perception that abandoning the SAT would open up opportunities for students with weak academics and low SAT scores who are good at storytelling.

Among private school graduates, there was a massive 67% increase in the number of strong applicants but a drop of 14% in the number of weak applicants. I attribute most of this to the same long-term strengthening of the applicant pool. That there was no corresponding increase in weak private school applicants may be because they figured they would be bad at storytelling.

If we compare the application numbers with the number of high school graduates, we find that 27% of public high school graduates, and 41% of private high school graduates, apply to a UC. The private school graduates make up about 12% of those who apply to UCs from California, with the exact percentage varying from campus to campus from Merced at 9.2% to Santa Barbara at 14.7%⁴.

The result of these shifts is that the private school graduates who apply to the UCs are academically stronger than the public school graduates who apply. This wasn’t always the case. As recently as 2017-19, the proportions of private and public school applicants who had 10 or more Honors were roughly the same at every campus except Riverside and Merced, at both of which the public school applicants were stronger. By 2022-24, the typical private school applicant was much stronger than the typical public school applicant. At Berkeley, the proportion of strong applicants was roughly the same in 2017-19 (33.6% of private school applicants and 33.0% of public school applicants had taken 10 or more honors classes) but there was a 9 percentage point gap in 2022-24 (43.9% of private school applicants but only 34.6% of its public school applicants). The gap was 7-9 percentage points at every campus bar Riverside and Merced.

All campuses are not equal

Applicants of all strengths prefer the same campuses. Among strong applicants, 87% of those who apply to any UC apply to UCLA, 81% to San Diego, 76% to Irvine and Berkeley. Only 14% of strong applicants apply to Merced. Los Angeles, San Diego, and Irvine also attract the most applications from intermediate and weak applicants. 55% of weak applicants apply to Irvine and 52% to both Los Angeles and San Diego. Only 27% apply to Merced.

Some applicants apply to only one campus. Others apply to all nine. In 2024, strong applicants applied to an average of 5.4 campuses, intermediate applicants to an average of 4.6, and weak applicants to an average of 3.9. All these numbers have increased over time. In 2020, they were 4.9, 4.4, and 3.8. Private school applicants, particularly strong ones, used to apply to fewer campuses than strong public school applicants but, by 2022-24, the difference had disappeared.

Increase in Applications per Campus

The combination of more strong applicants, more applications per applicant, and a perceived increased randomness in admissions decisions, has led to an explosion in the number of applicants in recent years.

Between those two time periods, there was a 40% increase in the number of strong applicants (see the Systemwide columns). This, and the increase in the average number of schools to which each of those strong applicants applied, led to a surge in the number of strong applications at each school. Each campus saw at least a 50% increase in the number of strong applicants. At Santa Cruz and Merced the increase was over 100%. The likely explanation is that the perceived unpredictability of the admissions process led many strong applicants to apply to schools that they would not previously have considered.

There was a big increase in the number of weak applicants to the most competitive schools. The total number of weak applicants increased 15% but Berkeley saw an increase of 65% and Los Angeles and San Diego both received about 45% more. Evidently, many assumed that getting rid of the SAT would increase their chances of admission despite their weak overall academics.

The numbers above are for applicants from all California schools. If we focus on private school applicants alone, the growth is even more startling. The 57% growth in strong applications to Santa Barbara consists of 50% growth in public school applications and 101% growth in private school applications. The 100% growth in strong applications at Santa Cruz consists of 90% growth in public school applications and 176% growth in private school applications. Meanwhile, Berkeley’s 64% increase in weak applications is a mix of 73% growth in weak public school applications and a 49% decline in weak private school applications. These are dramatic changes over just five years.

Composition of the Applicant Pool

Since stronger applicants apply to more campuses, and some campuses are more popular than others, the composition of the application pool varies between campuses. Intermediate strength applications constitute the largest group everywhere except Merced but strong applications outnumber weak ones at six of the nine campuses.

Admission Rates

On every campus, applicants with a strong academic record had a greater chance of admission. The size of the admissions advantage was greater on the more competitive campuses but it existed everywhere. Only 21% of strong applicants were accepted at UCLA over the period 2022-24 but that’s still a 3.6x greater chance than the 4.9% admission rate of intermediate applicants and a 16x greater chance than the 1.4% experienced by weak applicants. Those 1.4% must be recruited athletes or exceptional storytellers. Somehow, they were found to be more deserving than the 79% of strong applicants who were rejected. A strong academic record is not as big an advantage at UC Santa Cruz where 79% of strong, 61% of intermediate, and 29% of weak applicants were admitted.

The admission rates for private school students are significantly lower than for public school students with similar academic coursework. San Diego and Santa Barbara are the toughest on private school students. San Diego admits 45% of strong public school applicants but only 34% of strong private school applicants. Santa Barbara admits 28% of intermediate public school applicants but only 17% of intermediate private school applicants. Merced manages the curious trick of having a higher acceptance rate for intermediate applicants (97%) than it does for strong applicants from private schools (94%). When the acceptance rates are this high, my thoughts turn to the few who got rejected. What did they do wrong? Were their grades terrible? Did they all apply to do engineering?

Change in Admission Rates

The private school penalty in admissions that we saw in the last chart is a new phenomenon. It did not exist before the UCs went test blind. The same chart for 2017-19, when the SAT or ACT was required, would have shown admission rates for public and private students being much closer with the private school students sometimes having a small edge. For example, over 2017-19, strong private school students had their biggest edge at UCLA (34% admission rate to 29%) whereas strong public school students had their the biggest edge at San Diego (56% admission rate to 52%).

Of course, the purpose of this post is not to look just at the current rates but to examine how these rates have changed since the UCs went test blind. The following chart shows the percentage change in an individual’s chance of admission between 2017-19 and 2022-24. So a drop from a 40% admission rate in 2017-2019 to a 30% admission rate in 2022-2024 is a 25% reduction in the chance of admission. A drop from 4% to 3% is also a 25% drop.

We saw earlier that there was a substantial increase in the number and percentage of applicants with strong academic records. A reasonable hypothesis is that these new strong applicants would squeeze out the weaker applicants, leading to a reduction in the admission rates of weaker applicants. At the most competitive schools, the additional strong applicants would be enough to cause the admission rate for strong applicants to decline too. This may be a plausible hypothesis but it’s totally wrong. While the admission rate for strong applicants fell nearly everywhere, the admission rate for weak applicants actually rose at many schools.

San Diego experienced the biggest transformation. The admission rate for strong public school applicants fell from 55.9% to 44.5% but that for private school applicants fell from 52.3% to 33.6%. For intermediate applicants, the admission rate slipped from 24.4% to 23.0% if they were from public schools and from 21.8% to 11.6% if they were from private schools. For weak applicants, the admission rate actually rose from 5.5% to 8.7% if they were from public schools but fell from 5.6% to 3.0% if they were from private schools.

At Los Angeles, Irvine, and Berkeley, admission rates declined across the board. Despite the lower admission rates, the sheer volume of strong applicants still squeezed out the applicants with weaker academics.

Davis and Santa Barbara, like San Diego, saw big increases in application numbers but somehow managed to increase the chances of admission of the weakest applicants. Getting rid of the SAT evidently allowed them to give greater weight to storytelling.

Merced, Santa Cruz, and Riverside increased everybody’s chances of admission but particularly those of the weakest applicants. Merced went from admitting 52% of weak applicants in 2017-2019 to 87% in 2022-2024. It has always admitted most strong students so the increase in their admission rate from 94% to 97% was quite small in percentage terms.

Let’s assume that UC applicants from private schools had higher SAT scores than UC applicants from public schools with similar academic coursework. Going test blind would then make private school applicants less attractive and lead to them being admitted at lower rates. It wouldn’t explain why the magnitude of the change at San Diego is so much greater than at Irvine, say. Nor would it explain why the percentage of weak students who are admitted went up at so many campuses while the percentage of strong students who were admitted went down. This can only be explained by greater weight being placed on storytelling attributes.

More students are applying to more campuses and are getting rejected more often. But how many get in to at least one campus and how has this changed? About 86% of strong public school applicants got admitted to at least one campus in 2022-24, very slightly up since 2017-19. Only 77% of strong private school applicants got admitted. Even though they applied to far more campuses than previously, this was not enough to counteract the lower admission rates. For intermediate applicants, the gap between the public and private admission rates rose from 2% to 10%. The most dramatic change was for weak applicants. Thanks to the huge increase in the admissions rate for these students at Merced, both public and private weak students were admitted at higher rates than ever before.

The Changing Admit Pool

I have been focusing on admission rates rather than raw numbers because the number of admitted students at any campus is also affected by the space available (either because of new construction or because changes in state funding affect the number of out-of-state students admitted) and the expected yield (i.e. the share of admits who will ultimately enroll). But it is also useful to look at the composition of the admit pool i.e. what share of the total admit pool comes from each group?

We saw earlier that, compared with applicants from public high schools, applicants from private high schools had significantly better academic records in 2022-24 than in 2017-19. But we also saw that admission rates for applicants from private schools fell by much more than the admission rates for public school applicants. It won’t be much of a surprise then to see that the share of California admits to UCs who are from private schools fell systemwide and at seven of the nine campuses. San Diego (11.9% to 9.5%) and Santa Cruz (15.0% to 13.2%) saw the biggest drops.

We’ve seen that the number of strong applications is up, particularly at the less selective campuses, and that the number of weak applications is up, particularly at the more selective campuses. We’ve also seen how the admission rates have changed in response to the higher applications: stronger applicants are getting rejected more often and weaker applicants are getting accepted more often. When we combine the application numbers and admission rates data and look at how the admit pool has changed, we see that the share of admits with 10+ Honors classes increased at every school and the share of admits with fewer than 5 Honors increased at every school bar Los Angeles. Although the admission rates for strong students declined everywhere, the increase in their application numbers was so much higher that it drove up their share of the admit pool. The group whose share went down everywhere were those with 5.0-9.9 Honors. Although they still get admitted at a higher rate than the weak applicants, the greater weight being given to storytelling has reduced their advantage over the weak applicants.

San Diego is the most extreme example. It is the only campus where the share of weak admits went up by more than the share of strong admits did (3.2 percentage points to 1.0). By this measure, the overall academic quality of the incoming class at San Diego has declined, which is quite an achievement given the long-term strengthening of the applicant pool.

Conclusions

Any analysis of admissions decisions using publicly available data is inherently limited. Most glaringly, there is no breakdown by school. Everybody knows that it is harder to get admitted to an Engineering or Computer Science school than to an Arts and Sciences school but UC publishes no data that would enable anyone to quantify exactly how much harder. To the extent that strong applicants, or private school applicants, are more inclined to apply for Engineering, this could explain some of the observed differences in the admissions rates described here.

There is also no data on students who are admitted as part of the statewide and local guarantees. Nor is there any quantitative data about what I’ve labelled the storytelling aspects of an application.

There are two factors for which we do have data but which I decided, out of consideration for the reader’s patience, not to include. One is the changing capacity of each campus. San Diego now admits more California freshmen than any other campus. A decade ago, it ranked 7th of 9. The other factor is the applicant yield i.e. the percentage of admitted students who actually enroll. To the extent that campuses engage in yield management (i.e. rejecting students they think unlikely to enroll even if they are highly qualified), this could affect the analysis.

Nevertheless, the available evidence suggests that getting rid of standardized tests in the admissions process did not lead to greater emphasis on high school academic records. Instead, it led to greater emphasis on non-academic storytelling.

1. Although all applications for Fall 2020 required the SAT and were submitted before the start of the pandemic, the admission numbers are clearly inflated for that year.

2. In the list of factors that schools might consider for admission, the Common Data Set groups Application Essays as Academic rather than non-Academic. I can’t even imagine the argument for this classification.

3. Except Santa Barbara, where it is merely “Important”.

4. These percentages are the average across 2022-24.