Please returned to the topic at hand -schools returning to requiring standardized testing.
Users are free to start a new thread about need blind admission in the paying for college forum.
Further OT threads will be hidden. Thank you for your understanding.
I have never worked in college admissions so cannot pretend to know what happens during reading season.
BUT- I have worked for employers who ask for (and consider) standardized test scores as a screen prior to an interview. And scores are most definitely a heavy finger on the scale for kids from disadvantaged backgrounds or “not coastal privilege” background . Kid attended Oklahoma State. Kid has a 4.0. Lots of leadership on campus which is great, evidence that kid has taken advantage of every single opportunity to build her own rigor. Still great. Kid has a 1580 SAT score. That kid is likely to jump over the kid from the more prestigious school with the fancy EC’s (LAX team) and the 1420 SAT. Employers know a development admit (sometimes) or an athletic admit (sometimes) when they see one.
It’s not a perfect system. And trying to create a workforce that is up to the challenges at hand (one employer in particular required off the charts analytical skills- both math and verbal) AND isn’t filled with Princeton and Dartmouth Bros is a needle to thread.
But scores were an important component of our holistic review. And they definitely got kids from a wide variety of colleges a foot in the door. I remember my first hire from Albright college. Brilliant woman. During chit-chat I asked “how did you choose Albright?” and she said “it was my only affordable option, so that’s where I went”. And ditto several hires from HBCU’s (not Spelman, Howard, or the other elites…) and random Bible colleges with a student body of 800 in total.
What sort of industries or types of employers do this?
Take this for what it’s worth. It’s from Reddit
I was asked my SAT when applying for my first job in finance, more than 20 years after taking the test.
They were clear that not submitting SAT scores was increasing the likelihood that applicants would be rejected. They also said that this was worse for low income applicants. That means that submitting SAT scores were a determining factor, and it was more so for low income applicants.
Since Yale supposedly uses holistic review, that means that, if the rest of the profile was “sub-par”, a good SAT score would not have resulted in an acceptance. Ergo, based on what Yale announced, a noticeable number of low income applicants were being rejected specifically because they did not submit their SAT scores.
See what I quoted - Yale was very clear that not submitting test scores had a negative impact on admission chances of an individual, and even more so if that applicant was low income.
Anybody who knows anything about data analysis will tell you that this means one thing: “when controlling for all other factors, a low income applicant who did not submit test scores was more likely to be rejected than other applicants, including higher income applicants who did not submit test scores”.
If we take Yale’s words at face value, they are saying that low income applicants who didn’t submit test scores were more likely to be rejected than other applicants who didn’t submit test scores, regardless of other factors.
So Yale is admitting to discrimination against low income kids.
Or they may be “coloring” the truth.
Bottom line, the percent of first year Yale students on financial aid did not change because of the Pandemic, nor did the average amount (relative to COA). So Yale wasn’t accepting fewer low income applicants after they became TO. So being test optional did not result in it being more difficult for low income applicants to be accepted.
So either Yale is lying about the negative effects of TO policies on low income applicants, or they are discriminating against low income applicants. Only one of these can get them sued, and it isn’t lying.
My point is not that Yale is being discriminatory, my point is that test optional admission policies are NOT bad for low income applicants. They are bad for universities like Yale. Yale’s spin on why they are reinstating SAT scores makes in seem like they were discriminating against low income applicants.
While schools are pre then ever focused on SES diversity, this is different from not being need blind. Even majority wealthy public schools have students that can’t afford college and many students at top privates received significant FA. They are not at a disadvantage when applying to need blind schools.
Great. Then we are agreed that bringing back tests will benefit more low income applicants to Yale.
No, because if you read what I wrote, removing tests will not help low income applicants, because being test optional is not hurting them. If it was hurting low income applicants, there would have been a decrease in the number of students who are receiving financial aid after Yale went Test Optional, and there hasn’t been.
Seriously, the arguments supporting the claim that “requiring test scores helps low income applicants” all boil down to theoretical scenarios and unsupported claims. We don’t have numbers demonstrating that acceptance rates for low income applicants dropped, we have vague, unsubstantiated claims and imaginary scenarios,
“Imagine, if you will, that there is an AO, and that AO is perusing through an application by a low income applicant. Because the applicant is poor they are unable to afford fancy ECs and competitions, and the AO is unimpressed. However, imagine that suddenly, the AO sees the applicant’s stellar SAT score, and they now realize that this here applicant is, indeed, Yale Material!! So the AO decides to accept this poor kid based on their scores, and there is joy, happiness, and merry-making. All of this because the poor applicant was required to submit their test scores.”
If you claim that requiring SAT scores help lower income applicants, show me some data, not vague, self-serving claims by people whose job it is to spin the unpopular decisions of the administration.
As noted in my earlier post, highly selective colleges can largely admit whatever portion of low income students they want regardless of testing policy, so you see generally see little difference in stats upon switching to/from test optional. For example, across the 50 lowest admit rate colleges that switched during COVID, the median % Pell was 17% when test required and 17.5% when test optional. There were a few outliers, such as Caltech, which doubled % Pell when switching to test blind, but the vast majority of selective colleges had relatively little change in markers of low income.
Where you see a more notable difference in when you compare submitters to non-submitters at test optional colleges. Without exception every college that I am aware of that published such stats found that groups associated with relatively lower scores were more represented among non-submitters, including lower income students. For example, the review at https://nacacnet.org/wp-content/uploads/2022/10/defining-access-report-2018.pdf compares the demonstrated financial need between test submitters and non-submitters at 21 test optional colleges. At all 21 of them, the kids who submitted test scores had higher average incomes than non-submitters. The same pattern occurred at Bates and various other colleges that published stats. Some example numbers from the Ithaca study are below.
Applicants
Test Submitter Applicants – 10% Pell, 26% URM*
Test Optional Applicants – 17% Pell, 40% URM*
Admits
Test Submitter Admits-- 15% Pell, 22% URM*
Test Optional Admits – 29% Pell, 35% URM*
Enrolled
Test Submitter Enrolls-- 18% Pell, 19% URM*
Test Optional Enrolls – 30% Pell, 31% URM*
*Ithaca includes Asian students as part of their URM category. Only ~4% of Ithaca kids are Asian.
“Imagine, if you will, that there is an AO, and that AO is perusing through an application by a low income applicant. Because the applicant is poor they are unable to afford fancy ECs and competitions, and the AO is unimpressed. However, imagine that suddenly, the AO sees the applicant’s stellar SAT score, and they now realize that this here applicant is, indeed, Yale Material!! So the AO decides to accept this poor kid based on their scores, and there is joy, happiness, and merry-making. All of this because the poor applicant was required to submit their test scores.”
That applicant was me. 
For what it is worth, Dartmouth presented data showing that among disadvantaged students, the acceptance rate was a multiple for applicants who submitted a mid-1400s or higher test score, versus disadvantaged applicants who had such a score but did not use it. Their implicit claim is if such applicants are required to submit those test scores, they will admit more of them.
The thing is, when you do the math on how many applicants actually fit into that category–disadvantaged but mid-1400s or higher–it was not very many in comparison to the overall Dartmouth application pool. And then even if you assume it is true the acceptance rate would increase as per the data, that is still just a small fraction of that already small group whose result would change. Meaning even assuming the study data was predictive, the vast, vast majority were not being accepted under the old policy, and would not be accepted under the new policy either.
So if you are looking for some sort of large difference in the overall Dartmouth admit pool as a result of this policy, even if everything they claimed is true, you are very unlikely to see one. Because even taking them at their word, it would only end up changing the results for a very small group of applicants.
If only Dartmouth would share where/how they got test scores of non-admitted applicants who didn’t submit them (and the size of that sample)…then perhaps some would take their data with more than a health dose of skepticism.
So we know the answer to the first part of that question. According to Dartmouth’s white paper, they used “a specialized subsample of applicants in the test-optional cohorts who initially submitted scores but then asked Admissions not to consider them in the admissions decision.” In a note they explain Admissions did not in fact use the scores under those circumstances.
In terms of scale, their notes to Figure 6 say that “we have SAT scores both for students who submitted scores (blue lines) and for a small sample (19%) of applicants who chose to exclude their score from the admission decision but for whom we observe their scores ex post (red lines).”
I am honestly not sure what that 19% is supposed to mean–19% of what? But let’s say this is 19% of relevant applicants who sent them scores, but then 19% asked for them not to be considered (which sounds high to me, but who knows). Even that generous interpretation implies a very small sample size in each of their 50-point bins–there were only so many disadvantaged applicants with test scores in that range to begin with (a fraction of the total applicants with such test scores), which was only a fraction of overall applicants. And then only a fraction of that fraction of a fraction chose to have that score not considered, and then this is now a fraction of that fraction of a fraction of a fraction.
And in fact, if you compare the blue lines to the red lines in Figure 6, the blue lines are mostly pretty smooth and expected curves, whereas the red lines are pretty choppy. That is consistent with the idea the sample sizes for the red line buckets are getting so low there is a lot of statistical noise.
That being said, my two cents is I don’t really doubt that a disadvantaged applicant would usually have been well-advised to submit a 1450 to Dartmouth. The data is not perfect but that is a pretty reasonable hypothesis and the data does support it (imperfectly).
But then they do things like calculate, “Consider students with a score of 1450-1490 from less-advantaged backgrounds. These students increased their admission probability by a factor of 3.7x (from .02 to .074) by revealing their score.” I am very skeptical that the data actually supports a conclusion remotely that precise (3.7x). And in fact, you would need to control for other things that might correlate with test scores, the decision to submit or not submit, and so on.
Long story short, this is an old issue in this sort of research. Often the direction of a relationship is not so hard to figure out, but the magnitude is much harder to nail down. And I believe submitting a 1450-1490 had a positive relationship with admissions chances for disadvantaged applicants, but exactly what magnitude we are really talking about–meh, I don’t know.
And just to reiterate–even if you took that 3.7x as gospel going forward, you are still only talking about a tiny marginal group of applicants. Specifically, only 5.4% of a very small group to begin with (disadvantaged applicants with a 1450-1490 to consider submitting). The other 94.6% of this small group would not by this theory have a different outcome.
So even a generous estimate of the magnitude still ends up very small in the greater scheme. Although of course it could mean a lot to the few people who end up in that fraction of a fraction of a fraction of a fraction (I may be missing a fraction).
Good analysis!
Yep. And there was nothing to stop an AO from calling the HS counselors of TO students who they were most interested in, and said something along like ‘if this student does have test score of X or greater, we encourage them to submit it’. They could have also been much more transparent in general in all their admissions sessions, meeting with counselors, road shows etc in being crystal clear about test score submissions (so that low ses, etc students with high test scores and their counselors were confident in submitting scores.)
“And that little Boy was ME!”
Thank you for supporting my point. Instead of providing data to support your claim that “SAT scores are better for low income applicants”, you provide a story of how you imagine you were accepted to Yale.
That is likely true for Yale as well. There is probably not much real data supporting the claim that SAT requirements are beneficial for low income applicants. That are really just inventing excuses to justify a policy change which benefits Yale.
I mean, they could also have shown the data that there wasn’t an increase in the percent of students who were granted financial aid when Yale went test optional. However, that may bring attention to the fact that 44%-48% of the students at Yale are from families who can afford to pay $352,000 for their kid’s college. Since some 60% of American families cannot afford a 30 year mortgage on a $350,000 home, this would not be good PR for Yale.
It’s better to claim that SAT scores are good for poor people and leave it at that.
I think it comes down to the utility of SAT scores as predictors. If a school believes kids won’t be successful below a certain score, then they have a right to have policies that prevent such kids from being accepted. There are most definitely low income kids scoring very well. I can only tell you anecdotal evidence from our district.
I am generally for tests. It doesn’t need to be SAT. It could be a combination of APs that are similar in some ways to A levels in UK.
GPAs are increasingly meaningless. My son’s AP chem class last year had 15 kids. All but 4 of those kids got an A. Only two of them scored a 5. Seven of them got a 1. We are in a somewhat well off district but not one of those top districts like Palo Alto.
I’m not a fan, theoretically, but if high schools keep up the grade inflation trend, there may be no other solution but some sort of matriculation exams (like, as you wrote, A-levels).
Did anybody ever tell you that ad hominem attacks do not persuade? Actually, it was me. Yale lets you look at your admissions file. I did not imagine it–I read it.
The find-the-diamond-in-the-rough approach has a number of flaws, one of which is that colleges cannot admit students who do not apply. Requiring tests prevents two groups of otherwise qualified students from even applying:
Theoretically, a few “diamond-in-the rough” candidates may buck the trends to not only take the test but also have the wherewithal and sophistication to understand that their score may still be strong even though it pales in comparison to the published numbers, and thus apply. But the reality is that requiring the tests prevents a huge number of potential candidates from bothering to apply at all. One of the purposes of TO is to increase the number of potentially qualified candidates who apply, so it should be no surprise that requiring the tests works in the opposite direction.
By requiring tests, colleges are choosing to cast a much smaller net to bring in wealthier kids who are more likely to take the tests and (if they take the tests) more likely to have scores that fit with the published numbers (so they apply.)