Yes, absolutely, I don’t think I have read a “What we look for” page yet from any of these sorts of colleges that does not in some way imply they highly value active participation in their college community.
2 Likes
University presidents are focussed on politics, policy, and fundraising, which is their job.
If the president had wanted to seriously consider whether SAT score predict student success, they would have commissioned a long-term study which checked whether there was a correlation between SAT scores and graduation.
Choosing the one academic factor which is most likely to be correlated with SAT scores, rather than choosing actual graduation rates, GPA at graduation, GPA in major courses, or even chance of obtaining a job after graduation, is a strong indication to me that they knew what result they wanted to have. Did they actually do a study to figure out whether the students who did not submit SAT scores graduated at lower rates? No. Did they test whether the students who did not submit SAT scores have a correlation between their high school GPA and graduation rate? No.
So I think that you are wrong to assume that President was coming from a Neutral perspective, and we don’t even know what the Board of Trustees wanted. The Boards of Trustees at colleges are invariably more conservative in their views (I’m not speaking now of partisan politics of the party they endorse, but of willingness to change their view of the world). They all grew up in a world in which SAT scores ruled, and they are much more likely to see SAT scores as being the end all for testing intellectual abilities. I wouldn’t be surprised in the pressure the reinstate SAT scores came from them, rather than being a policy that originated from the president.
Well, them and admissions. Requiring test scores may reduce the number of applications, and will make their job simpler.
I do not see my skepticism of their use of this statistic as being “nitpicking”. The graph is minor compared to that, and I just use it as another indication of what I see as a lack of honesty in their presentation of the reasons for their decision.
If they would present a study in which, after controlling for the effect of income and for high school GPA, they demonstrated that SAT scores were a strong predictor of graduation rate or even GPA at graduation, or demonstrated, after controlling for family income, that TO students had a lower graduation rate and that the high school GPAs of these students did a worse job on their own in predicting this, I would consider that to be data-driven support of requiring students to submit an SAT scores.
At the end of the day, though, they are a private college, and I’m not emotionally invested in their admission standards as I am in the admissions processes of public colleges. I just am constantly annoyed by the fact that they, and other such low admission private colleges, keep on putting in standards that benefit the wealthy and powerful, and claim that these make their admissions “more fair”.
PS. The reason that it doesn’t bother me a lot is that I believe that private colleges really shouldn’t be required to fix a situation that is being created by the states which refuse to invest in the education of low income kids in their state.
4 Likes
Umm, I believe the U-Chicago would like a word…
By focusing only on graduation (“one” factor), aren’t you doing the same thing for which you are chiding Dartmouth?
2 Likes
I mean university and higher ed politics, not local, state, and national election politics.
No, I wasn’t complaining that Dartmouth used a single outcome. I’m complaining that they used an outcome which doesn’t signify what they claimed it signified. Ultimately “success” in college = graduating successfully, so using that is legitimate, even as a single outcome.
1 Like
Right, per your definition, and many of those who are critical of standardized testing. Others, might have a different view; For example, ‘thriving’ at college, which is more than just graduating. For example, someone who is prelaw is not well served to attend a highly selective (and competitive) school where, if they were underprepared, might struggle with low B’s first year when they could attend, say a Cal State instead and earn a 4.0. Or, someone who is pre-engineering but with weak math skills so they end up ‘graduating’ with a Studies major instead.
Yes, they earn a degree, but not necessarily in their interest.
Yes, I get that many kids change their majors, but what about a study to see how many of those TO students earn Latin Honors or are in the top half of their class after 4 years?
4 Likes
I appreciate you engaging with the question. Here are my thoughts:
First, only part of Dartmouth’s argument for bringing back test requirements is about outcome: quoting from their announcement -
Their illuminating study found that high school grades paired with standardized testing are the most reliable indicators for success in Dartmouth’s course of study.
There are a ton of studies out there we’ve all seen from various schools and groups of schools that show varying levels of correlation between SAT scores (by themselves and with other factors) to GPA, long-term graduation rates, etc. The data on the success rates in a Test-Optional environment is, I think we can all agree, much more sparse, and of course newer TO schools can’t do that for a while…
Of course they didn’t, because they can’t yet. The first class admitted TO matriculated in 2021. They haven’t graduated yet. If they want to do a “long-term” study - let’s say five years? - then they’d remain TO until the class of 2026, and really they would wait until that class graduated so as not to be swapping back and forth, so they’d be TO until 2030. It’s reasonable to see that a school that saw TO as a potentially-temporary adjustment due to a global crisis might want to use data earlier to evaluate that decision, even if there are reasons as to why that data might be imperfect.
And, of course, graduation rates are an imperfect tool for this, especially at highly-selective schools. Dartmouth’s four-year graduation rate is 88%, six-year is 96%, so the likelihood of a substantive differentiation here is unlikely. The much longer testing thread has lots of other arguments for and against graduation rate as a reasonable metric, and I stopped reading that 800 posts ago, so I won’t rehash them. I would just say that graduation rate is one reasonable measure of outcome, and first-year GPA is another.
Here’s the second point, though, which is that Dartmouth said it wasn’t just interested in measuring matriculated student outcome, but also interested in regretted student declines, based on a lack of relevant information:
They also found that test scores represent an especially valuable tool to identify high-achieving applicants from low and middle-income backgrounds; who are first-generation college-bound; as well as students from urban and rural backgrounds. It is also an important tool as we meet applicants from under-resourced or less familiar high schools across the increasingly wide geography of our applicant pool. That is, contrary to what some have perceived, standardized testing allows us to admit a broader and more diverse range of students.
Even if we time travel and discover there were no discernible differences between matriculated students who submitted test scores and didn’t, this argument still holds up: if Dartmouth thinks it is losing out on a tool to identify students (for lots of reasons, including misinformation/counselor misunderstanding or disengagement/the Reddit effect/whatever), this could be enough of a reason by itself to remove the TO policy.
I wouldn’t be surprised either - I’m not naive enough to really think the President of any school is without opinion or influence - but I don’t think there is reason to believe that the Dartmouth Board of Trustees is substantively more conservative than the ones at, say, Princeton, Stanford, or Harvard, all of which are TO for at least 2025, if not longer.
Where I end up is pretty straightforward: if we’re going to be fair to the schools making these decisions, we are best off taking them at their logic and their word (and seeing if we can learn from it), rather than supporting the ones that confirm our (possibly-well-formed) bias and questioning the motives of the ones that don’t.
4 Likes
I don’t disagree. But that means similarly accepting the data/reasoning that some schools have used to inform testing policy decisions that are different than Dartmouth’s. Some have chosen to remain test optional for a longer time, like Harvard, or some have chosen to stay permanently TO (Columbia, many LACs).
Seems schools should do whatever is right for them and their institutional strategy. I will note they are all targeting the same group of FGLI and/or URM students so in some sense their strategies/priorities do overlap.
2 Likes
Why? I personally don’t care why schools choose to go TO or Test Blind or Test required. As long as they are not discriminating against protected classes, they can and…
If counselors have an issue with standardized testing, then they can just not recommend Dartmouth & MIT.
btw: Similarly, are you suggesting that Yale and other schools that require supplemental essays show thier data on the reasoning of using such essays?
I agree, and I don’t have a need to see or vet any school’s data or reasoning for their policies. IMO they can do whatever they want within the bounds of the law.
I am somewhat skeptical of Dartmouth’s claims that they really want to increase the proportion of FGLI and/or URMs. Many highly rejectives have had this goal for a decade or more, and for most of that time D has required tests…why weren’t they getting more of these students then? What’s different now?
I do wonder why, given that goal, would D want to make the target pool relatively smaller, which requiring tests does, while many of their competitors will have a larger pool of FGLI/URMs to choose from. It’s an interesting marketing strategy. Time will tell how it works.
It is extremely common.
5 Likes
Even based on the Dartmouth analysis, students who had an SAT of 1200 are not getting “low B’s”, their average is B to B+.
Moreover, here is a little factoid - grade inflation at Dartmouth is far larger than grade inflation at Cal state. To the assumption that a person who will get “low B’s” at Dartmouth will have a 4.0 at Cal State is just that, an assumption.
It seems to me that the assumption that an A is easier at Cal State than at Dartmouth would be pretty classicist, especially considering that the grade inflation at Dartmouth is higher than at Cal State.
I cited a study above that demonstrated that high school GPA is better at predicting graduation rates from college than SAT scores are.
I think that SAT scores are not a bad tool in aggregate, so a high schools which has increases in the GPAs without a similar increase in their SAT scores indicates thing like grade inflation. However, rather than the absolute scores, the changes are more important.
Now, to be honest, if somebody comes up with a test which actually measures mastery of the material learned in high school, without there being benefits for those who can pay more, I would also accept that as an admissions requirement.
However, I don’t think that the SATs do this, as they are formatted and as they are administered today. There are too many benefits for students who can afford to pay somebody to train them specifically how to do well on the SATs, rather than learning the material, and there are too many advantages for those who can pay for accomodations, better testing conditions, and multiple tests.
My issues are not with testing as a concept (after all, much of the high school GPA is testing) but with the SATs as they are today.
3 Likes
The SAT Subject tests were very valid measures of subject mastery and college readiness. You used to have to submit 3 to an Ivy or equivalent. Too bad they are gone. AP results show the same readiness but isn’t it hit or miss on how many take them?
1 Like
I assume CollegeBoard decided to phase them out to focus on AP exams, which are often school or district supported and bring in more revenue.
More exams should be available though - the AICE program should offer its AS level exams in more states for instance, or IGCSEs…
The State of California (students) was CollegeBoard’s biggest customer, and when UC dropped the Subject tests, a bunch of revenue went with it.
Sounds like Yale is thinking about bringing back the SAT.
The rationale people use when explaining why SAT scores are correlated with income have traditionally focused on ability to pay for SAT prep courses. Today, thanks to the good people at Kahn Academy, there is a level playing field. I found Kahn Academy SAT to be excellent, personally. The reason income and SAT scores are correlated is due to hereditary characteristics. Smart parents tend to be wealthier and have smart children.
The reality is that grade inflation exists to varying degree across colleges and high schools across the country. That’s a reason why there are standardized tests - to give everyone an equal playing field to which to assess competence. That’s why there’s a Bar exam, MCATs and so on.
Straight As mean different things whether from Dartmouth, Cal Poly, Princeton or my local community college… it’s certainly true, likely to a greater extent, when comparing your local high school in North Dakota, South Central LA or rural Arkansas.
5 Likes
The playing field is not level, not even close. SAT correlation with wealth goes far beyond the ability to pay for test prep…it includes a lifetime of superior K-12 education, greater vocabularies from exposure to highly educated adults, level of stress in one’s home setting, etc.
This article is interesting, doesn’t look like I can gift it though. Many public libraries have free access to NYTimes. New SAT Data Highlights the Deep Inequality at the Heart of American Education - The New York Times
5 Likes
Yes, and depending on your school you may not have all that many by the time application season comes around. At our HS in MA, APs are capped and not available until junior year so at most kids might have 4 scores.
Sorry, I meant level playing field for test prep, not across the board. Kahn is as good as any of the expensive prep services, IMO.
2 Likes
While those other things may be factors, the greatest thing these kids have received from rich parents is inherited intelligence and genetics.
I agree with this.
3 Likes