I find it curious that Geiser is critical of UC and the College Board for not using student demographics*, when some of those are specifically excluded from admission criteria by law (bolded) or the data does not exist. Yes, UC does give an admissions bump to first gen to college and those that click the low income button, but there is no way verify the accuracy of those claims. And Geiser has to know that. Most of his “data” on income is self-reported by 17 or 18-year olds. (Even CB recognizes that their parental income ‘data’ is suspect.) The only hard data Geisler would have access to is from those students who apply for federal loans.
*College Board: “gender, ethnicity and SES…”
Geiser’s abstract: Compared to high-school grades, SAT/ACT scores are much more strongly correlated with student demographics like family income, parental education, and race/ethnicity.
In other words, we’d have a better model if we can use data that is precluded by law and/or data that doesn’t exist and is not collected or verified.
While that might be true I haven’t seen a lot of evidence that lots of affluent TO students are being admitted over kids with test scores. If anything, I think the expectation is that you should have a test score if you come from an upper class community. I don’t think TO is really TO for everybody. Most kids applying to top schools submit tests - often those who don’t submit are “hooked” kids (athletes, donor kids etc) - and to be blunt, prior to TO some of those students were already being accepted with much lower test scores. As far as “redeploying” time, a kid would have to be spending a crazy amount of time prepping for it to take away from other activities.
But it all correlates with wealth and privilege. The reason the UCs, Cal Tech, and other schools have gone test optional with regard to the SAT/ACT is that, given their various academic missions, the benefits of using the tests don’t outweigh the costs of requiring the tests. In other words, more is lost than gained. The calculation may be different with the APs.
I am wondering if we will see more of Purdue’s strategy. This is similar to the position schools took on the SAT subject test, moving them from “test required” to “test recommended.” The idea was to not penalize someone who truly did not have access to the test.
“With Purdue’s “test expected” approach to first-year admission, 44,000 Early Action applicants applied with a test score, and 97.2% of the students offered admission through Early Action submitted test scores.”
Maybe they should use the test scores as a reverse indicator? I can see it now - Apply with us because we admit based on those with the lowest SAT/ACT scores…
The reason I chose “A” over “B” is because the cost/benefits of “A” outweighed those of “B”.
Sure.
But the reason this is controversial in the first place is because lots of folks question the cost/benefit analysis that the UCs are using and, in particular, feel it’s a way to get a little bit of pseudo-affirmative action effect (not directly legal in CA, IIUC), at the cost of a lot more noise introduced into the process (of selecting highly academically capable students for the most challenging UCs).
I guess I’m a little more of a stickler and did frown upon UC’s jumping ahead to test blind (instead of temporary test optional or such) without the further exploration the academic senate proposed. And as far as I’ve heard the new assessment system was looked on skeptically from early on; seemed dead in the water from the start.
I was aware of the recommendation for creating a new assessment, but due to the above felt the real contest was between (a modified) SAT and test-blind.
For test optional, the more relevant question is what SAT adds beyond the criteria that would be used to admit test optional applicants.
I agree with this. And to add mtmind’s comment to it, is the benefit worth the cost? And I think the schools mentioned in the article have indicated their views on it (granted without all lifting the hood for us to see), with the most crowd stirring being Yale’s.
So is the result of all this basically that each school must decide for themselves whether testing is relevant? I’m curious if the small/LAC vs med/large school hypothesis can hold up
Without data on hand, I cannot argue, but I don’t quite see it that way. APs seem worse in terms of access and scores (i.e. teaching) with respect to correlation with wealth/privilege.
I don’t think basing a decision on a cost benefit analysis is tautological. My point is simply that not all potential data points are going to have the same costs or benefits, so it makes no sense to say if you look at AP Subject tests you have to look at SAT tests, or vice versa.
On the other hand, I do think there is something illogical when parents essentially say that X should be the standard because my kid excels at X.
As for the supposed “pseudo affirmative action effect” here is the breakdown of the first time admits to UC Berkeley. Can you point out who is benefiting from this supposed affirmative action?
This bit feels tautological: “The reason the UCs, Cal Tech, and other schools have gone test optional … is that … the benefits of using the tests don’t outweigh the costs of requiring the tests.”
I mean, sure, the reason for any binary choice is that the decision-maker has decided the benefits of the choice made outweigh the costs. But others can (and should) question/debate those assessments.
No doubt it enters, consciously or unconsciously, into folks’ thought processes (as does the reverse). But I don’t think anybody in this thread is explicitly arguing that, so it’s a strawman here.
I don’t want to de-rail this thread too much into topics that the mods want confined elsewhere. That said, yeah, I could point them out (or take some guesses anyways), but won’t.
Perhaps I didn’t explain well: If we are trying to enhance the demographics of the student body and SATs are in the way, I can understand wanting to get rid of it. But if APs are going to similarly affect the demographics, then I don’t understand using it while banning the SAT. I believe more schools do not use AP scores for admissions than schools that do not use the SAT for admissions for the reason that AP scores are more problematic than the SATs.
My kids excelled at testing. Each took it once and were done.
But for years I have said that the test scores matter least for kids like mine, who came from a well-respected public high school where each top college gets dozens of applicants from our school, admits some of them, and where the admitted students almost always thrive in college. Because the colleges know our school well, the college can determine academic fit just from the high school transcript.
And for years, I have said that it matters most for students from high schools where a student rarely applies to elite colleges, by showing that they significantly outperformed their peers and therefore can likely handle the rigor of elite colleges. I am glad to see confirmation of this from places like MIT, Yale, and Brown.
You have made a good point that quite often these students never apply because they look at the 25%-75% and think they have no chance, even if they are scoring 300 points above their school mean. But the flip side is that without test scores, they do apply, but the college is reluctant to admit because they wonder if this student can thrive at the college. One option might be to eliminate public reporting of 25%-75% scores and tell everyone that test scores are encouraged.
I don’t understand the cost-benefit argument. The cost for taking SAT/ACT is minuscule compared to the cost for attending college, even after aid. The former is also cheaper than taking AP exams. And there is no covid now. So why is cost not justifying benefit used as a reason to go test optional/blind? Am I missing something?
Data clearly shows that affluent kids are submitting scores at a higher rate than less privileged students, although in my corner of the world, I still know many who applied test optional (and were admitted, unhooked). So while I initially thought that it was basically an expectation that affluent kids would submit scores (which would make sense), I’m not sure. Obviously, though, this is anecdotal and your larger point stands–I think most affluent kids applying to competitive schools continue to study for and take the SAT/ACT and see it as an important part of their application. I also know that many are delighted to not have to worry about it.
You’re also right that athletes and donors have a long history of being admitted with scores that fall below the school’s average, so that reality is unchanged in a TO environment. But at least they had to take the test and perhaps sweat it out that their scores were good enough, even if they could be admitted with lower scores than an unhooked applicant.
The odds are still better than if they don’t apply. And most in that situation don’t apply. Most don’t even take the test!
DEMING: For children from the bottom 20 percent of the income distribution, only about a quarter of them take an SAT or ACT test. Among those, only about 2.5 percent score 1300 or higher. You can see right there what we’re up against in terms of economic inequality in college admissions and success in college and in life. Crimson
That is a “cost” of requiring test scores, one that eliminates capable students from even sending an application.
Another cost is the resentment and confusion by families of high scoring students if scores are viewed in context. That isn’t how the public thinks “standardized” tests should be viewed.
So as not to mislead any applicants, or get their hopes up…Purdue is test required, unless a student can’t take a test:
*Purdue University expects applicants to have SAT or ACT scores. Recognizing that in some exceptional cases, applicants may not have been able to take a test, we allow submission of applications in those cases via the Common App.
To highly rejective schools?
This is true and data confirm it. I haven’t heard any of the test proponents on this thread address how to counter this. I also don’t know how many students this really impacts, but as one who has worked with plenty of low income and or URM students I have seen it many times.
This is all important context for what was going on at the time. It does seem they were kicking the can down the road with the idea that it would take 9-10 years to develop their own UC testing system.
I found this statement towards the end of Geiser’s critique a bit—to use his own phrase—“spurious,” in that holistic review is not usually a way to consider getting more academic information about applicants, but rather to consider other factors:
While predictive-validity studies examining only test scores and grades may have made sense at one time, when UC admissions relied almost exclusively on those two factors, this is no longer the case. Holistic review has substantially expanded the body of both academic and socioeconomic information considered in admissions decisions, with the result that SAT/ACT scores have become increasingly redundant.
Furthermore, although “mtmind” linked to Geiser’s critique of the UC Academic Senate task force’s report twice, they didn’t look into responses to Geiser that occurred contemporaneously, such as the following that calls into question reliance on Geiser’s critique as the last word on the validity of the report:
The UC Academic Senate task force members stood by their work.
“What can I say? Saul unfortunately is wrong,” said Li Cai, a UCLA professor who directs the National Center for Research on Evaluation, Standards, and Student Testing.
Cai said the task force analyses “very much included the purportedly omitted demographic variables, through a more transparent means.” He said task force members chose to use a different and simpler model than did Geiser so the public would more easily understand their findings.
He also said Geiser’s model did not reflect how admissions decisions are actually made at UC campuses. Admissions officers compensate for the discriminatory impact of SAT and ACT scores by discounting their weight while increasing emphasis on grades in application reviews of underserved students, the report found.
UC initially gave campuses a choice of test optional or test blind in the face of COVID-19 related test availability issues.
Then UC lost a lawsuit claiming that test optional was unfair, so all campuses had to go test blind. With that lawsuit outcome, UC’s future choices are test blind or test required.
Many states REQUIRE SAT or ACT - I’d guess that, pre-COVID, that they alone accounted for >25% of the HS grads in the country.
Lower income “smart” kids are not idiots, by definition. Yes, their counseling (from parents, HS staff and teachers, and peers) is weaker than the wealthy, but are you suggesting there’s a large (75%!!!) pot of poor but smart HS grads, pre-COVID (2019 or thereabouts) who would have scored >1300 on the SAT but didn’t take either the SAT or the ACT? I find that strains credibility.
I don’t know the details of the lawsuit in question or the law(s) that are apparently at stake. But I’m not aware of other states claiming that due to federal law, they are legally confined to test-mandatory or test-blind. Thus I’d suspect that the lawsuit at issue addressed a state law, and state laws can be changed.
California has many social problems at the moment that are not faced to the same degree by other states. Sometimes there is a claim that things MUST be that way because of this or that state law/court ruling. But again, laws can be changed. Blaming poor government/regulation on poor laws is obfuscating things - legislatures and governors exist for a reason.
SAT/ACT scores…
