Based on how they define “need”.
Still nice to see some objectively rated data injected into the process. I am so very tired of the aggrandized extracurriculars with meaningless nonprofits, charitable work and related awards that litter most of the applications I see. The poor really do not benefit from the focus on extracurriculars.
Everyone I speak with has their hardship story ready. Actually, the most disadvantaged often do not promote it as much as the privileged.
12 Likes
that’s true, but folks looking for a full ride will be pleased with the need-based aid. Sure, Dartmouth is not as generous as HYPSM, but then no one else is either.
2 Likes
My state’s public schools got rid of their own standardized test, on which a certain (low) score was required in order to be graduated, and instead made the SAT mandatory, but there is no required score to be graduated. The PSAT was already offered to every student. So effectively, this gave every student the opportunity to take the PSAT and SAT for free. I would hope that every public school in every state would follow this lead, and offer both the PSAT and SAT/ACT for free at least once to each student, and strongly encourage or mandate that the student take it.
My own kid, having already taken the ACT and gotten the score that they wanted, slept in on the day of the SAT, if I recall correctly, and fortunately, still received their diploma, probably as part of pandemic exceptions.
7 Likes
I’m uncertain how they define each group, but I tend to agree with this view.
1 Like
The data presented in the pdf shows a highly correlated linear relationship between grades and SAT. So, as an isolated variable, their research says best single indicator. clearly there is not one magic variable.
I would, however, suggest that’s it’s a far better indicator than the number of clubs one belongs to. Or whether they were voted class president.
6 Likes
I do think that far more URM or poor rural students who have high GPA’s but low SAT/ACTs will be identified and excluded, than “diamonds in the rough” included. Dartmouth is trumpeting the positive - that they could have identified and taken high GPA kids from low-performing high schools - but they’re not stating the obvious and appropriate corollary, that with a mandatory SAT/ACT, they’ll also be able to identify and reject the high GPA students from low-performing high schools who are not ready for Dartmouth. The fact is that in many poor rural or URM inner-city high schools, the student who attends class, is not disruptive, and hands in his homework/assignments is going to get straight A’s, because that puts them at the top of the class, even though, for whatever reason, they haven’t yet mastered advanced high school level English and math. Many of these highly-ranked students, best in their high schools, will score, not just under 1400, but under 1200 on the SAT. Scores like these indicate that they are not ready for high level college work. This is why MIT quickly moved to require a standardized test - not because they wanted to take the 800 math scorer over the 760 math scorer, but so that they could filter out the A student whose math SAT score indicated that they just were not ready for college level math.
Students with low standardized test scores are better served by attending colleges which meet them at their level of preparedness. Community colleges, state colleges, and less selective liberal arts colleges offer classes starting at the remedial level, all the way up through higher college level standards. The simple fact is that there are students who are at the top of their low-standards high schools, who are not ready for the level demanded of freshmen at highly selective colleges, and a low standardized test score is a way of identifying this.
I strongly suspect that Dartmouth will wind up using standardized test scores “holistically”, looking for a >1500 score from applicants who didn’t face academic barriers, while accepting >1400 scores from applicants who did, just as they view other application criteria.
4 Likes
The studies I have seen on this subject suggest at the undergrad level, the debt problem is concentrated among people who attend for-profit institutions, particularly if they never actually finish their degree.
If instead you are looking at people who attended non-profit colleges and got their degrees, few of them have any sort of long-term debt issues, and indeed the typical “return on investment” is strong across different fields, including the oft-derided arts and humanities.
3 Likes
Yes, it was never likely these colleges were going to just rank and accept applicants by SAT/ACT score.
1 Like
But now they will - and they will decline - by ACT/SAT score.
I think they may have a better prepared student body, but I don’t think it will be as diverse. I just don’t see how they are going to find all these diamonds in the rough given how few low income students take the tests in the first place, let alone score a 1400+. I’m not suggesting this is the wrong move, btw, just trying to inject an element of realism into the idea that suddenly all these 1400 scoring, FGLI/lower SES students are going to materialize when they haven’t in the past.
3 Likes
No, what it shows is a correlation between the SAT and the first year cumulative GPA.
Other studies have demonstrated that the first year cumulative GPA is not a strong predictor of graduation rates, or even of the four year GPA.
However, SAT supporters keep on using this statistic because it is the only statistic which regularly correlates with SAT scores. SAT supporters also like creating graphs that remove most of the variance in the graphs, in order to create a visual illusion that the correlations between SAT scores and first year GPA is stronger than it actually is.
7 Likes
I note Dartmouth is using binned scatter plots. They divided the 4051 Dartmouth students in the sample into 16 equal-sized bins, and then plot each bin as a single point. The virtue of such a presentation is it makes it easier to see the linear relationships, which is fair enough. But used incautiously people can mistakenly conclude there is a much higher correlation than would be apparently if you actually plotted each of those 4051 individuals.
Table 1 reports the actual R2, which is 0.222. Which is definitely information, but I am pretty sure it would look a lot less impressive to the eye without that binned approach.
So I would also caution people to understand Dartmouth has not in fact announced they are going to abandon holistic review.
From what Dartmouth has said in various places, it wants more test scores to use for the purpose of initial academic screening. Beyond that, it is undoubtedly still going to use all that other stuff when making actual admissions decisions.
And indeed, take observations like this:
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. The test-optional policy thus led to Admissions not identifying these high-achieving applicants as highly prepared.
OK, so being identified as academically highly prepared is important–but you still only had a 7.4% chance of admission at that point! Even the top scoring disadvantaged students only got up to around 17%.
So the other parts of holistic review still play a very big role.
5 Likes
So, you are saying that Dartmouth is using this statistic as a visual illusion in order to argue something that isn’t true? wow.
MIT was able to maintain its diversity numbers after going back to requiring tests, and their math cut off is likely higher than Dartmouth. As others have noted, there will have to be increased outreach to low income schools. I think many colleges have already started doing this, off the top of my head have heard of or read about UVA, GaTech (which also requires SAT), and Hopkins increasing such outreach, in the case of the first two, particularly in rural areas
1 Like
Colleges are pretty risk averse. Few highly rejective ones will take a chance on a kid from an unknown ( or known to be awful) public high school who just presents the usual 4.0 gpa and varsity sport, but add in a decent score (1300 or more I think) and that can do a lot to calm concerns about academic readiness.
I asked one student ( a relatively recent migrant to the US from Africa) from an awful school if he was in any clubs. He said it wasn’t safe to be in school outside of class and he tried to avoid the troblemakers. There are lots of reasons kids don’t have stellar extracurriculars. He is a non native English speaker, so I am not sure about his essays. I am sure his teachers don’t write effective letters. A score would really help.
7 Likes
To strengthen a weak argument. Very typical for people with weak arguments to “clean up” their visual representations of their arguments.
Similarly, they will repeat “highly significant”, and focus on P values rather than on R square values.
They also will select the one statistic that seems to support their argument, misrepresent what that statistic actually means, and equate correlation with causation.
3 Likes
But, your contention is that Dartmouth is manipulating data in order to present something that doesn’t exist? Why would they do that? There is zero reason to reintroduce the SAT if they didn’t see real value in the results.
4 Likes
Personally, I wouldn’t go that far, but I do think Dartmouth is highly aware of the scrutiny it gets for anything like this, understands these tests have drawn strong criticisms (many well founded), and is trying to present this decision in a way calculated to minimize the blowback.
1 Like
Ya, my point is versus not including it at all. I’d choose the SAT over club participation, for example.
I think there is a lot of additional information provided by the SAT but agree that there is not one magic variable that predicts success at the next level and I don’t believe the SAT is a panacea. On the other hand, it does help ID kids that otherwise might not get a look.