Highly competitive high schools produce a large number of students with incredible test scores. College counsellors at these high schools realize that submitting a test score will invite comparisons to the other applicants from that school or similar competitive schools, and while a 1520-1540 SAT is a great score, it may pale in comparison if bunch of applicants from the same school scored at 1570+. If the applicant can otherwise demonstrate their academic excellence, submitting the score won’t help and may hurt the application.
When I first became aware of this, my reaction was that it was silly logic, and that the colleges will just assume that those who don’t submit test scores got low scores, and reject those kids. But the reality seems to be more nuanced. While colleges will use the test scores if they are provided, at least some colleges aren’t penalizing for non-submission of scores, provided that the applicant can otherwise demonstrate their strong academic ability and potential (through grades, course rigor, school rigor, essays, recommendations, etc.)
(Again, though, this is all dependent upon your child’s particular circumstances, including the relative strength of the applicant pool in your child’s school and demographic. If the counselors are competent, you should discuss it them rather than following whatever might be said on CC.)
One should also consider correlation vs causation when drawing conclusions from such statistics.
For example, suppose MIT admitted their entire class of 2024 by pure random lottery without considering the application at all. The stats of applicants (not admits) suggest that if the applicant pool remained the same, MIT’s 25% ACT score would have been ~33, rather than actual 34 – not a huge difference. ACT scores and the application were not considered all in the decisions, so applicants certainly didn’t need a high ACT score for admission in this system. Yet the 25th percentile ACT was still a 98th percentile score of 33.
The high 25th ACT score in this random lottery system occurs because MIT’s class of 2024 was self-selecting. Kids with high ACT scores applied to MIT, so kids with high ACT scores were admitted to MIT… something of a self-fulfilling prophecy. Colleges for which applicants have a high 25th percentile ACT score will almost certainly have admits with a high 25th percentile ACT score.
Looking deeper, the admission decisions at MIT were not just random and did have a correlation with score. Kids with higher ACT scores had a higher admit rate. Specifically the admit rate was 10% for 34-36 vs 4% for 31-33. That is a clear statistically significant difference in admit rate, but that doesn’t mean the score was driving the difference in admit rate. The first sentence of MIT’s score page links to a page on MIT’s website that explains this well, in which MIT claims there isn’t a significant difference in their admission process for the 2 groups and instead the higher admit rate primarily relates to other things that are correlated with score, rather than the score itself. The 34-36 ACT kids are more likely to have other traits that MIT values than than 31-33 kids, including things like averaging higher GPA, higher course rigor, better LORs, better ECs/awards, etc. It’s unclear how the difference in admit rate would be among 2 applicants that were equal in all metrics except ACT score.
Now, I and others are on the record as saying that we admit people, not test scores, and that in any case there is really not a difference in our process between someone who scores, say, a 740 on the SAT math, and someone who scores an 800 on the SAT math. So why, as the commentor asks, is there such a difference in the admit rate? Aha! Clearly we DO prefer higher SAT scores!
Well no, we don’t. What we prefer are things which maycoincide with higher SAT scores. For example, a student who receives a gold medal at the IMO is probably more likely to score an 800 on the math SAT than a 740. But if we take an IMO medalist (with an 800) over random applicant X (with a 740), does that mean we preferred an 800 to a 740? No. It means we preferred the IMO medalist, who also happened to get an 800!
This distinction is particular relevant in recent admission cycles since all of the above colleges were test optional this year, and some may continue being test optional in future cycles. For example Caltech has committed to being test blind (not optional ) this coming year. Even though Caltech is test blind, I expect that if they collect ACT scores of entering students for statistical purposes, the ACT scores will be extraordinarily high. I’d expect Caltech’s scores under a test blind system will be as high or higher than the HYPS scores quoted above under a test required system.
Test blind by definition means that scores will not be considered, yet I expect high scores because Caltech’s applicant pool is self selecting, and scores are correlated with other criteria that Caltech values. The increase in applicants upon going test blind this year suggests that a significant portion of applicants are outside of Caltech’s usual score range, so not as self selecting as usual. However, I expect a much lower admit rate among low scoring applicants to Caltech than their overall average, in spite of scores not being considered in the application process. This relationship occurs because the scores are correlated with other criteria that is considered in admission decisions. The scores will no doubt be lower than Caltech’s usual >99th percentile average, but will still probably be extraordinarily high, such as >98th percentile.
Going back to the OP’s comments, if submitting test scores, yes, it is good to have scores that are near or above the college’s median score – not necessarily required, but they are good to have and are likely to improve chance of admission. If not submitting test scores, a low score is not going to eliminate you. However, things that are correlated with the lower score may eliminate you. And if you get a lower score, there is a good chance that you have these characteristics. In short don’t just focus on scores – focus on the full admission process and consider what the college is looking for in applicants.
I am highly skeptical of going test optional. I think “test optional” is primarily a tactic for schools to spike their application numbers to look more competitive. For what it’s worth, US News and World Report does also, having removed acceptance rates from its ranking criteria.
Unless the rest of a candidate’s application is so strong that the candidate would have gotten in with a score well outside of the middle 50%, I think that candidate should submit scores. There is no evidence that schools are not penalizing candidates that don’t submit scores.
Is there any evidence that schools are punishing students who not do submit scores?
If a school like Cal Tech believes it can identify well-qualified candidates without relying on test scores all, what make you think that other schools couldn’t do the same for those who don’t submit test scores?
US News removed acceptance rate as part of the ranking criteria in 2018, well before the pandemic. They said they removed it to make room for the social mobility indicators, but regardless, I agree with the title of the following article
There are a lot of people on this board that assert with a lot of authority that someone should only submit SAT/ACT scores if they are above the median for that school. I do not think there is any data to support that advice, and in fact, some of the schools that have released data about acceptance rates for test optional applicants seem to indicate that there is a penalty for not submitting tests.
I think school Admissions Officers are smart enough to figure out all of the signaling involved in test score submission, so if the score is not submitted, I believe that AO’s will assume that the scores are bad.
Which colleges are these? In another post, from several months ago, I listed some specific numbers, which rare requoted below. Is this type of figure you mean? The first figure is 44% of applicants to Vanderbilt were test optional, but only 39% of admits were test optional. This implies non-submitters had a slightly lower admit rate than submitters, so does this show that Vanderbilt had a penalty for not submitting scores?
Total Class (ED/EA + RD) Vanderbilt Total – 44% of applicants test optional, 39% of admits test optional Tufts Total – ~50% of applicants test optional, ~41% of admits test optional Wellesley Total – 60% of applicants test optional, ~50% of admits test optional
Only ED / EA
Penn ED – 38% applicants test optional, 24% of admits test optional
Amherst ED – 45% of applicants test optional, 39% of admits test optional Notre Dame REA – 49% of applicants test optional, 31% of admits test optional
As stated in my earlier post, you need to also consider that scores are correlated with other criteria that colleges value. Kids with lower test scores also tend to have a lower rate of ALDC hooks (especially true for early applicant subgroup), lower grades, worse course rigor, worse LORs, worse ECs/awards, etc. Considering the worse rest of the application, a similar admit rate between submitters and non-submitters would suggest non-submitters are favored.
For example, suppose the applicant stats were as follows (hypothetical example, not real stats). A similar admit rate between the 2 groups is not expected, regardless of whether scores or lack of scores is considered.
Vanderbilt Submitters Applicants – Higher rate of ALDC hooked, Averaged 3.7 GPA, averaged 5 AP classes, 15% had top LORs, 15% had top ECs/awards
Vanderbilt Non-submitters Applicants – Lower rate of ALDC hooked, Averaged 3.5 GPA, averaged 4 AP classes, 12% had top LORs, 10% had top ECs/awards
We had “debates” on test optional policies on numerous threads before and I’ve stated my positions which I won’t rehash. However, the point that @CTDad-classof2022 raised whether an applicant should submit her/his score that is below the median to a school does deserve more exploration. If the score is in the bottom quartile, most would probably agree that it isn’t a good idea to submit. If the score is above the bottom quartile (but below the median), what should the student do? Would a college assume the student’s score is in the bottom quartile? Would it look at its applicants as two separate pools (test submitters and non-submitters)? Or would it look at all applications without the assistance of test scores first and then check the scores for confirmation?
It seems to me that colleges are deliberately opaque as usual with regard to their test optional policies. Perhaps it’s a necessity because they’re still unsure themselves. But I’m certain submitters and non-submitters wouldn’t be treated exactly the same, because it simply isn’t possible. Caltech is the only private elite that I’m aware of that has chosen to go test-blind instead of test-optional. It did so because it believes that a test-optional policy would give some applicants an advantage over others and is inconsistent with its core belief that no one shall take unfair advantage of any other.
I agree test scores between the 25%ile and 50%ile is a gray area for some schools.
Not all schools lack transparency wrt to when to send test scores though. For example, Tulane AOs directly state in admissions sessions not to send if below the previous year’s median. ND AOs generally are also open with candidates about test scores, and which round of admission might be best for them. Many direct admit nursing AOs are helpful as applicants are deciding whether to submit test scores.
I agree some schools could be more transparent, but IMO they can choose who they want to fill their classes. Advice that is regularly missing from these boards is for applicants to email or pick up the phone and ask. Just ask. Often one will get an answer.
Lastly, several Cornell schools are test blind again this year. Will be interesting to see if test blind policies increase, or not.
There’s another issue that I didn’t mention in my prior post: the published test scores for all these test-optional schools are rising, and significantly in some cases. An applicant can’t even be sure in most cases what the score distribution really means. The optional nature of test reporting will lead to fewer and fewer applicants submitting their scores, as the 25%tile numbers rise each year. Another self-reinforcing spiral in college admissions.
I agree and it can be a problem…some test score ranges for class of 2025 are very high. For example, Penn’s mid 50% admitted ACT range was 34-36, SAT 1500-1560.
Every applicant is unique…basically if the test score strengthens the app, send it.
I have not listened to a Tulane session yet this year, but need to do that to see what direction they are giving applicants. But their numbers didn’t creep up much…class of 2024 median ACT was 32, this year’s mid 50% range was 31-34, so median up about half a point.
For Georgia publics, seems more straightforward to ignore class of 2025 data and use 2024, as they are back to requiring tests. Not perfect for sure….I tell students to reach out and ask for direction. Some do, some don’t.
If I was cynical (and TBH, I can be), I would argue that schools like test optional for 3 reasons:
Spikes applications to make the school look more competitive;
Allows the school to hide ALDC and other candidates (fine arts, etc.) where the primary criteria for acceptance was never test scores and now the schools don’t have to report those lower scores; and
Because of 2, the overall scores will go up, making the schools look more competitive.
Put another way, when a school is telling a candidate to NOT provide useful information about that candidate, there has to be a benefit to the school.
I think also think that is reasonable to assume that the non-reporters all had lower test scores than the reporters, or maybe there was some overlap. Someone with a test score above the prior year’s median is very unlikely to go test optional. The simple math of this would seem to indicate that going forward, the reported test scores will just represent the top % of prior year admits equal to the % that reported scores.
Or maybe I am wrong, and the schools are just going test optional for the heck of it.
I obviously don’t know any school’s reason for going TO, but most schools were not TO prior to the pandemic. Seems like if there were inherent benefit for the schools to go TO, many more would have done so prior to the pandemic …when they were basically forced to because a not insignificant proportion of students couldn’t sit for tests. The reality is that some schools have successfully been TO for 50 years.
We have covered this ad nauseam on other threads, but GPA is the primary predictor of college success. When adding a test score to GPA, the increased predictive power is de minimus. This data is what many TO schools used to initially convince their trustees and faculty to go TO (again for those who were TO prior to the pandemic)
If the tests are useless, then more schools should be going test blind. The U Cal system went test-blind primarily for political reasons, and I think the result will be exactly the opposite of what the state wants. CalTech is a universe of one or very few among the prestige private schools going test-blind.
Grade inflation and EC resume padding, especially at prep schools and wealthy suburbs, will just get more exaggerated if there is no way to benchmark kids across high schools.
As a result, I think there is quite a bit of disagreement among whether high school GPA is the primary indicator of future academic achievement. More importantly, whatever people’s personal opinion about SAT/ACT, I think posters should be more circumspect about telling applicants not to submit scores when it is not clear where the cutoff to go TO actually is.
Like anything, people have to do their own research about submitting test scores…there’s much info out there. Do admissions sessions. Ask their AOs. Ask their GC. Ask their independent college counselor if they have one.
If they do post on CC, they do need to wade through responses and decide for themselves which ones are relatively more informed and/or are more in line with what they are hearing from schools/AOs/GCs.
Not just responses. They need to wade through numerous threads, often populated by the same posters making the same tired arguments that they did on countless other threads. My hope is that this thread will not devolve into yet another circular argument on TO.
But yes, it is incumbent upon applicants to do their research and make their own informed decisions.
Some went test optional or test blind because they realized that many potential applicants could get no test scores at all due to COVID-19 cancellations.
Whether they continue is probably the subject of study by their institutional research departments.
I haven’t followed any of the old threads on this issue, but I haven’t seen anyone on this thread “telling applicants not to submit scores.” On the other hand, you and a number of other posters seem to be advising applicants to submit scores even though “it is not clear where the cutoff to go TO actually is.” Shouldn’t your admonition apply both ways?
Moreover, you and others seem to be making a number of assumptions that that aren’t necessarily supported by any data. For just one example, you write:
Based on my limited observations, I am not sure either of these assumptions are reasonable, but there may not be enough actual data to support a position either way.
College admissions is complicated and sometimes counter-intuitive. Depending upon their respective circumstances, a great score from one kid might not be considered a great score from another kid. If these two kids with the same test scores apply to the same school, it may be advisable for one to submit the score and for the other not to submit.
Maybe those aren’t the only options. Maybe the schools are going test optional for reasons you don’t completely understand or haven’t fully considered.
I am not going to convince anyone to submit scores that does not want to. My S22 is applying next year so I hope many applicants go test optional to the schools he is applying to.
