The claim was made that selectivity is the cause of a more holistic approach to admissions because of their correlation. I argued that it’s the other way around that holistic admissions actually increase selectivity, which accounts for the positive correlation.
I agree that both are imperfect. So why do away with one but not the other? They help validate each other for many colleges. IMO, the test scores at the minimum aggregately, if not individually, are better indicators. Take a look at CDS’s from many colleges. They don’t differ very much in the GPA distributions of their admitted students. The test score distribtuions are much more discriminating and better reflect the academic side of their student compositions.
I am not sure the chicken or egg analysis helps much. From what I have read and seen, and AO talks I have listened to (like from Bowdoin who has bee TO for decades), the AOs who look at scores filter for “good enough” - which is a lower number than you would think. But the kids who have the rest of the package just end up having higher scores than than “good enough”, too, dragging up the median score.
I’ve looked at enough scatterplots to know that the highest scoring/gpa kids from elite high schools are if anything less likely to get in than their slightly lower scoring schoolmates. Presumably because they have more going on in their lives - ie holistic admissions.
Places like Stanford and the Ivies actually have lower median test scores than the schools a notch below.
AOs also say they don’t compare grade ranges between schools, but rather they compare grades among kids from the same school. If they don’t have enough info about what a kid’s grades mean because no one ever applied from that school before, and the rest of the package doesn’t illuminate whether they can handle the work, then that kid is probably out of luck. A standardized score isn’t going to save them.
Standardized scores are just so darned flawed. From the correlation with SES, the ability to prep, the pressure for colleges to accept high scorers to keep their USNWR ranking - they really don’t add much to the equation. @data10, you have probably answered this question a billion times. What does the regression analysis say?
I would imagine that a typical AO, if not pressured by having to report standardized scores to the great ranking systems in the sky, would pick the go-getter with the lower score but high class rank any day and twice on Sunday. And if they were free to do that, they would have just as high performing students and not lose a lick of prestige.
4 Likes
The claim that similar grades are noncomparable should be easily proven then, yes?
Like, not by appeals to “common sense” or anecdote, but by the use of hard data.
I don’t think all As are equal any more than all high schools are equal. Even within a high school, one teacher’s A is harder to get than another’s.
The proof you are looking for is in how A students from underserved high schools perform in the same intro college classes that their elite prep school equivalents take. There was a great “This American Life” episode this pst weekend on UT Austin that addressed this very question. The A students from underserved high schools come to those classes less prepared. They aren’t less inherently capable or deserving of a spot, but their calculus class didn’t cover proofs, for example. With appropriate support those students catch up. But their calculus A does not equal an elite private high school calculus A.
6 Likes
The study that you mentioned sounds like what I’m looking for—do you have a link?
But I would ask that people stop making their point of comparison “elite” high schools (private or public). I think we can all agree that the tails of the distribution are, well, the tails of the distribution (as I’ve said before). I’m talking about the middle two or (I suspect) three standard deviations—that is, the source of education for the vast majority of high school students. Saying that the bottom 1% is worse than the top 1%—however defined—is outside the bounds of what I’ve been asking for.
You can search NPR on Itunes and listen to the latest episode of This American Life, as mentioned up thread. No link - as subscribed podcast.
2 Likes
There are many far more influential factors in selectivity changes than having holistic admissions. Regardless, plenty of colleges have had large increases in applications in recent years, prior to COVID. For example, Chicago applications increased by a factor of ~400% in the past couple decades to nearing 40,000, which began well before going test optional. Did Chicago decisions become formulaic and less holistic with the rapid increase in selectivity and applications? Can you think of any college that has had this change as they’ve had a rapid increase in applications or selectivity? Instead the opposite tends to occur. With a larger number of highly qualified applicants, colleges tend to place a greater emphasis on non-stat criteria rather than place an increased emphasis on finding applicants with the pinnacle of high stats. If the colleges need to hire more admissions readers to get through the applications, they do so. More applications means more application fees, which can support more readers.
Being imperfect is not a good reason to make an admission criteria required. While it may be your opinion that test scores are better indicators, the actual studies that have been posted do not support this. Instead they suggest that test scores at little beyond the other available admission criteria that is used at highly selective colleges. This other criteria includes far more than just looking at GPA in isolation. GPA in isolation being compressed in the CDS does not mean highly selective colleges have no other way to distinguish applicants besides what is typically small differences in SAT/ACT score.
1 Like
There have been many specific studies that have been posted in the thread whose results depend on the specific measure and conditions, but the general conclusion is fits well with your description – “really don’t add much to the equation.” I’d add the qualifier at the “top schools” mentioned in the subject header and first post. These “top schools” consider many additional factors besides just GPA in isolation in their admission decision.
1 Like
And so the stress level for high school students will increase even more, as application outcomes become even less predictable than currently, and everyone decides to try for Harvard.
Do the “hail mary pass” applicants experience stress at the thought of not getting in to Harvard? That would surprise me. I think many of them do it to placate parents, grandparents, etc. knowing full well that there are at least 20 kids in their HS class who have more of “the goods” (tests or no tests) than they do.
Since every now and then a Hail Mary application will work, yes, I think they probably do see themselves as having a chance. As do many kids with solid but not remarkable credentials.
I think this is a really good point. Students are very aware of how they stack up to their peers within their own schools.
I think it’s more of an issue for the top kids in HS, especially when the parents are telling them they are unique and the next best thing, forgetting that there are kids just like them in every high school across the country (and in some cases, the world).
Of course there’re other factors driving selectivity. However, the point is that holistic admission is a cause of selectivity, which leads to a positve correlation between the two. The other factors will surely also lead to positive correlations between them and selectivity, but they’re totally irrelevant to our discussion about the relationship between selectivity and holistic admissions.
Not all high schools rank. Not all students are aware of their position relative to their peers. And relative position is not necessarily determinative unless one is focusing on UTexas schools. We had unhooked applicants not in the top quarter of the class get into Ivies, and the valedictorian rejected.
1 Like
Even in schools that don’t rank, kids know who the super stars are in their classes. It’s like the kindergarten soccer teams that don’t keep score yet every single kid on the field knows who “won” the game.
Yes, there are always surprises, but I agree with Blossom in that those acceptances are happy surprises and not soul crushing rejections.
5 Likes
Not necessarily. They may know the presidents of school clubs, or captains of the school teams, but have little insight into nonschool activities. The kid from our high school who got into Stanford last year flew well under the radar but had published medical research done at a local university. Two years ago the Stanford admit had done remarkable work in Peru, we all found out later. And no one knew my kid was meeting with the governor to discuss voting rights. So just more uncertainty thrown into a very uncertain process.
1 Like
Ok, but they COULD rank. Or put into deciles.
I highly doubt adding a Hail Mary school to a list is more stressful than the testing situation is for these kids.
The arguments against standardized testing seem to be based on the following:
-
Test scores disadvantage students of low SES. If that’s the objection, then AOs can be asked to take SES into consideration (or even expliciitly adjust test scores for these disadvanged students as they do in some countries where standardized test scores are used for college admissions). However, even though nearly every objector claims the effect of test scores on low-SES students is one of the primarily considerations for her/his objection, I noticed some lack of genuineness or consistency by some of them. For example, one objector favors giving recruited athletes in some special, expensive sports special admission consideration. It makes one wonder whether s/he objects because her/his recruited athletic kid can’t clear the minimal bar on test scores for recruited athletes at some top schools.
-
High test scores are influenced by excessive test preps. The current versions of these standardized tests are more preppable because they were redesigned to satisfy some of the demands of these same objectors. If we were to make the tests less preppable (and they would be significantly harder as a result), the objection to the tests from the same objectors would be even louder.
4 Likes
I do think some of the kids with little to no chance of a highly selective admit experience stress at getting into a meets-full need school…for many of those kids that’s the only way they might be able to go to a 4 year college. And we know that budget is the number one factor in determining where (and if) many HS grads go to college.
3 Likes