I enjoyed that article. This excerpt was interesting:
Caltech aside (as I thought the discussion here covered that well), the above quotation from Friedman notes the variation across institutions that we’ve conjectured here. He seems to imply that the utility of testing is significant when selectivity is high enough that the HS gpa loses sufficient utility. And when that is, the institution needs to determine for themselves. Not sure how that works with a place like Bowdoin so that implies that while testing has sufficient utility in certain places, those places are manifested in a very small number of schools (but the ones many are attracted/addicted to).
He is very clear to say that both can be true, he just notes that there is a tension between the two. This tension is clear because we’re nearly 850 posts deep into a thread on the subject. I think the point he is making is that while grades aren’t the only factor considered, they are the predominant factor (and rightfully so), and this was true before TO. My understanding is that test scores were never used in isolation or as the dominant variable in a student’s application.
And yet, just because a course is labeled “Honors” or even AP, course rigor, pace, grading style and instruction quality can vary from school to school, and teacher to teacher. So we’re back to the fundamental disagreement about whether test scores are “worth the squeeze” to shed light on these inevitable discrepancies. Reasonable people can disagree…and clearly they do.
Much of the apparent tension is a product of the exaggerated level importance that many give these silly tests. In other words, by presenting the false choices (SAT vs. grades alone) and flawed comparisons (MIT and just about everywhere else) he is drumming up tension which doesn’t necessarily even exist. Expanding access doesn’t necessarily decrease quality. And in the process he ignores the abundance of evidence that TO works just fine at schools that have been using it for a long time.
He’s also ignoring some real ugliness that underlies the view that expanding access decreases quality. Some people just don’t believe that certain segments of society belong in college.
College is not for everyone. Could we bust that myth? It isn’t 13th grade, part of a basic minimum education all need to master. It is for those interested, motivated, and who can succeed from the course of study. There are alternatives.
From the tables in your post, your daughter’s high school has more Asian Americans applying to UCs than all other races combined. Only a dozen or so places in the US have such a high concentration of Asian Americans. So no, your experience/community is far from being typical of the US at large. 98% of my children’s “flyover country” schoolmates have no idea test prep centers exist . I believe a large swath of the country didn’t know as well.
Yes, I believe my city’s population is approximately 30% Asian or maybe just under that. The Black community is very small (also reflected in the UC application data). So there are some demographic specificities. So, sure, this may not be at all typical of a town of similar size in say, Kentucky, but just pointing out that there are students who put hours of prep time into the SAT. Not all skim the book and take it on the fly. And of course for those who choose to do that, it comes with opportunity costs in terms of ECs, or just having fun with friends or getting more sleep. But some prioritize this which, of course, is their prerogative.
I’m in the SF bay area, so related demographics. I’d say that most of the students in our area do prep, but I guarantee it doesn’t prevent them from any of their activities. It feels equivalent to two weeks of practice for a musical instrument when you include the practice tests? Or soccer practice for a couple of weeks? It’s nothing compared to what they sink into sports/ECs although it may bite into their screen time. Which I’m willing to sacrifice. I don’t feel too much stress in the students I’ve seen but that is admittedly a sample population.
Your continued mischaracterization of my initial post is puzzling. I mean, you have seen my posts for years, and you know that I understand holistic admissions quite well. So I’m confused, why would you think that I believe that colleges admit using grades alone?
And regarding Jake Vigdor’s model, note that the words I used were “mathematically correct”. Apparently you misunderstood that to mean the only valid model. I can see where you might make that mistake, but that’s not what it means at all. There are always many “mathematically correct” models. Finally, note that Vigdor’s model includes both gender and race. While including those can be useful for creating a model, note that race can no longer be directly used in admissions, and it’s unclear whether gender can be as well. So while it’s a mathematically correct model, in its current form it’s not a useful model for admissions.
If any part of this wasn’t clear enough, let me know.
2 weeks? The kids in your area must be much smarter.
I’m in SanDiego and I have heard many stories kids signing up for summer prep courses that starts in June and ends Aug, and these are everyday affair. Several kids have taken these classes multiple summers to get a 1600 super score. My nephew did one at cost over $5k. After he and many of his friends with 1500+ scores fail to get into their top 5 choices, I told my own kid don’t bother.
Tutors who charge $150-250hr are also not rare in these area and yes they guarantee 1500+ after so many hours.
The CSU weighted capped recalculation means that this 3.9 GPA likely corresponds to an unweighted 3.6 by a student who took at least 8 semesters of honors courses.
As one of my former professors once said (and I think he borrowed from someone else), “There are no solutions, only tradeoffs.”
I haven’t seen anyone on this thread say that “certain segments of society don’t belong in college.” I see people saying that there are possible tradeoffs involved in TO policies. Again, most of us have a fairly balanced view here. No one is saying tests should be the only criteria or the main criteria, just one criteria to support grades (after a rise in grade inflation post pandemic).
To quote from the Inside Higher Ed article posted by @pilate:
Friedman [John Friedman, co-founder of Opportunity Insights and the lead researcher on the study on which the NYT story was based] also clarified that, unlike Leonhardt, he holds no strong opinion about whether his research makes the case for testing requirements. He just wants the data to speak for themselves.
“I don’t think there’s a right or wrong decision … there’s nothing wrong with admitting students who you know are going to struggle,” he said. “But ignoring that there is a decision to make is not the right approach. We need to be honest with ourselves about what the trade-offs are.”
In related news Vandy extended their test optional policy for three more years, thru Fall 2027. Reps had previously stated they would continue to analyze whether TO made sense for them. Clearly they do think that by extending the policy, would be nice if some data were to follow.
Vandy has a maximum of 61% of its enrolled class for fall 2022 having applied with scores, at least 39% without scores. I wonder whether the portion submitting/not submitting plays a role in a college’s decision to continue the test optional policy.
Of course you don’t believe it. It is because you don’t believe it that your point of comparison (test scores vs. grades alone) makes no sense whatsoever. You are using statistics to advance arguments that do not apply in the real world, even though you know better.
As for Vigdor, that is not all you said. You are fully aware that, as he pointed out, the value of the test scores is greatly diminished when additional factors are considered. And all AOs use many other factors besides just grades. You ignore this in your initial post and throughout. Even though you know better. Why knowingly create the false impression that these scores provide more predictive value than they actually do?
Why do you keep referring to these tests as silly? On an SAT math test, you interpret word problems, compute percentages, analyze graphs and tables, and solve systems of equations. In a freshman math class for STEM and many non-STEM majors (e.g., business, economics), you do the exact same thing at a slightly higher level. So why are these tests silly?
Regardless the word one wants to use, the SAT is designed to test what students have learned. I expect we all agree many K-12 schools in the US are lacking in academic offerings and quality of instruction. Basically, some students who take the SAT are being tested on concepts they have not yet learned.
I don’t think that happens in a freshman math class, does it?
Here is a podcast with Caltech admissions talking about their decision to go test blind. I found it to be very informative and helpful in understanding the thought process of this highly selective school.
I’m not sure if this is accurate. My understanding is SAT math covers pre-algebra, algebra I and II, and geometry, which students should have learned by the end of their junior year, before they take the test. I’d be surprised if College Board would design a test that tests material students have yet to learn. There would be an uproar from schools and parents.
Now let’s assume this is indeed the case. If students somehow have not yet learned such basic math concepts and are admitted to colleges – even elite colleges – where freshman math classes progress at a greater speed, then are AOs not setting these students up to fail? Or are college professors who are busy with their research so magical that they could achieve in a semester or two what high school teachers couldn’t in years, dramatically elevating the math level of these students?
. I believe a large swath of the country didn’t know as well.