More Colleges Backing off SAT and ACT Admissions Rule

Applying to College
361

If you don’t like the report, many other studies have been linked showing the same trend, including the one in my previous post that you neglected to reply to. The trend of females having a higher average GPA than males occurs in grade school, high school, in first year college GPA, and in cumulative college GPA. It occurs in engineering majors, as well as in non engineering majors. I can keep posting more studies that show this conclusion if you like, but I expect this is not going to change your opinion on the subject.

“Keep noting” implies stating more than once. This was not the case. In one post, I stated:

And in another post I stated the following. Note that I am talking about the confidence related conclusions.

362

@foosondaughter - I should have qualified that statement even further that it is just my opinion that women applicants are being granted moderate preferences in MIT admissions, but I thought it was implicit.

Admissions is a competitive process, especially so at the most elite colleges. Agreed that mathematical aptitude is not the sole criterion for picking the “best” class at MIT or elsewhere (you can insert any type of relevant aptitude here, depending on the institution). But neither is the concept that MIT - or any school - is picking those students it believes will be most successful at the institution (or thereafter) or will most contribute to a “strong” class. Who knows, but I am guessing that if those were the criteria, double or even triple the actual number of admits would “pass.” That is, if those were the sole criteria, the institution would not be able to pick the actual admits from many supremely qualified applicants.

I take it as axiomatic that schools admit those students who match the school’s institutional priorities (these can include legacy, development, athletics, URM, etc. although most of those are supposedly not utilized at MIT). The existence of the diversity report that we are all discussing makes clear that one of MIT’s institutional priorities is admitting more women. I am sure that there are more recent numbers, but the differential admit rates are striking:

“The most recent federal data show 13 percent of female applicants were offered entry to the elite school in Cambridge [MIT] for fall 2014, compared to 6 percent of male applicants.”

https://www.washingtonpost.com/news/grade-point/wp/2016/03/13/want-an-edge-in-college-admissions-see-the-schools-where-women-and-men-have-an-advantage/?utm_term=.3cd7710f6714

IMHO, that’s too large of a difference in admit rates, especially in light of the clear disparity in mathematical preparation and aptitude as shown in the diversity report, to be explained in the absence of preference. Again, just my opinion!

363

And here’s one offering a possible explanation why females gpas are higher: a possible tendency to protect them more than men. This study discusses behavior regarding the economics major. It also states that female gpas are higher, but notes that male grades in the intro course are higher than females. It also shows that females have a higher tendency to drop econ as a major if their grades in the intro course are not good:
https://scholar.harvard.edu/files/goldin/files/claudia_gender_paper.pdf

In addition, female grades in the more mathematical versions of intermediate econ courses are lower than males, while the reverse is true in the less mathematical versions of the courses. The gender ratio in the less mathematical courses slightly favor females vs the more mathematical courses, which again hints at potential self-selection that females employ more readily than males.

I’m not dodging your Rowan paper; it supports your view. I’m replying in the only way that I can given that by showing other examples indicating otherwise. The Caltech and Harvard papers are interesting counterpoints from where I stand.

So in your mind “they give various details about the process” and “They explain their methods clearly” are not repeating… ok, I read it as such where you’re showing the upfront nature of their presentation as reason to believe them. I just would like to see the data on gpa by major before believing them… I don’t know why that is considered such an outlandish request.

364

@Data10 - “The trend of females having a higher average GPA than males occurs in grade school, high school, in first year college GPA, and in cumulative college GPA.”

I don’t know why people have such a tough time understanding or accepting this. Males and females are different, and on measures of maturity, conscientiousness, focus and others, girls and women consistently show advantages as compared with boys and men. On average, of course - we are always talking averages. These abilities are plausibly related to academic performance advantages in the classroom, whether we are talking English class or engineering.

On the other hand, studies (and SAT scores) have consistently shown that males (once past puberty) have higher quantitative and visuospatial reasoning than females. Males also exhibit higher variance in these abilities. The result is that at the highest levels of mathematical abilities, males enjoy an overwhelming advantage over females. These abilities plausibly relate to achievement at the higher and highest levels in quantitative fields, both in college and beyond. In some ways this is the mirror image of the cumulative large advantages that females show in the middle of the distributions for conscientiousnes, maturity, focus, etc.

There is some evidence that females enjoy some advantages on verbal intelligence, but we are talking engineering here, so…

Newsflash: men’s and women’s brains are neurobiologically different: http://www.sciencemag.org/news/2017/04/study-finds-some-significant-differences-brains-men-and-women. We are learning more about these differences, which are encoded at the genetic level, every day. Haier’s “The Neuroscience of Intelligence” is a good place to start your reading if you are interested.

Just as we shouldn’t take female GPA as evidence of structural bias against males in the edutocracy, we shouldn’t view female underperformace on quantitative measures like the SATM as structural bias either. Again, for the hundredth time, we are talking about averages. IMO we should embrace the true diversity that acknowledges our differences, rather than the phony diversity that always seeks to prove we are all the same.

365

@saillakeerie

"Its interesting to me that you think testing optional schools are doing so to cut off the money train or to relieve kids of the stress, hassle, etc. of taking the tests. If those schools determined it was in their own best interests to require testing, the next class of admits would be required to take the tests (the money train and stress/hassle of the kids be damned). "

Um… I’m not certain I ever argued that that is why all schools are doing it, but no matter why they are doing, I support them putting an end to, or at least making optional, what is a “test to nowhere” make-work project that costs in some (many, most?) cases unneeded money, time, stress and even confusion (which test? When? Subject test in June? Do I have to send? Superscore?.. etc.)

It sets up any number of unneeded barriers to entry, whether financial, time constraints, psycho-emotional, informational or even locational (how many test are within a close range of each student.)

This discussion is literally titled more schools are back off standardized testing. It’s interesting the # of folks coming by to tell us those schools are wrong. Ok.

I do think it’s worth reading Bates 2009 publication on their 25 years of test optional admissions and what they had observed and learned. (and note that Bates allows schools GCs to send test scores so students don’t have to pay Princeton.)

https://www.bates.edu/admission/optional-testing/

(Link at the bottom of the page.)

366

@SatchelSF

“I have zero doubt that females are being granted at least moderate preferences in admissions at MIT. Ensuring that more females are admitted and succeed at MIT is obviously an institutional priority, and keep in mind there is always a subjective component to any grading system.”

If this is true, they should be giving MORE preference in admission to F as they are performing better than their M counterparts. Even if your (uh… “unsupported?”…) claim about GPA is true, ain’t nothing subjective about graduation rates. You either get the credits or you don’t. (Yeah, I know, all the guys that aren’t graduating quit to start Geocities.)

Grad rates are almost always slightly better for F in almost every academic environment, so it’s not surprising.

Really, evidence suggests that MIT is not putting a big enough thumb on the F scale!

367

As far as I’m aware, MIT is only explicit in not giving preference to legacies.

I just don’t think that the admission-rate difference is itself compelling, even with the difference in mathematical ability. In fact, the lopsided academic skills of many male applicants – strong in math/science, but weaker elsewhere – may well be a source of their disproportionate rejection. This is the flip side to boys scoring higher on the Math SAT: those girls who do have high math scores are more likely to also have high verbal scores.

368

@CaliDad2020 - “Really, evidence suggests that MIT is not putting a big enough thumb on the F scale!”

You keep focusing on endogenous variables like GPA and graduation rate as measures of institutional “success.” By logical extension, MIT should go all female!

What you miss, and if you read in the literature of intelligence and behavior you will see it, is that males have higher variance in behavior and intelligence. This variance in behavior may perhaps explain why men fail to finish their degrees in higher numbers than women. However, male variance in quantitative intelligence means that most of the “halo” science and math achievements will be by men. Again, MIT makes a huge deal about how it has placed in the top 5 in the Putnam team competition in 16 out of the last 20 years. (It truly is an astounding achievement for an institution.) That is the kind of “halo” achievement that attracts the best and brightest, especially from overseas. MIT had 42 opportunities to pick a female since 2003 (and it is a choice by faculty) to be on a team, and yet chose exactly 0.

It’s like Pareto optimization. In any group 80% of the achievement is going to come from 20% of the group. Take males out of the equation and MIT will cease to be relevant at the elite level. Again, don’t shoot the messenger - just my opinion!

Some people might find this interesting (I think I have linked before but not sure): http://www.lagriffedulion.f2s.com/math.htm

369

13,131 men applied to MIT last year while only 5889 women did. For whatever reason, MIT is far more appealing to men than it is to women.

MIT accepted a roughly similar number of each - around 750.

It seems very few women want to apply to MIT in the first place.

(according to http://web.mit.edu/ir/cds/2017/c.html )

370

This seems to be more popular every year. https://www.insidehighered.com/admissions/article/2017/10/16/move-let-applicants-self-report-sat-and-act-scores-spreads

The linked list of schools allowing it is at 46 (not 25 as stated in the article).

371

Spent one summer giving statistics lesson to some graduate engineers and was surprised to learn how upset one civil engineer was to learn that a bridge, designed strictly by the book, might collapse if the underlying assumptions about the severity of an earthquake were changed. All the available data could offer for an honest answer was “probably.” Underlying assumptions are always there, but often ignored. Ignoring what we don’t know about a process involving human behavior does not make it “objective.”

Our understanding of human behavior cannot be reduced to the certainty found in a testing lab where the civil engineer’s concrete block finally breaks at a known, measured force. I was trained as an economist. Economists must work with statistics and list their assumptions.

We have not reached the point where we can know with certainty that one applicant is a better investment than the other. Welcome to the world of behavioral science. College admissions, selection of a spouse, selection of a career and even the purchase of a vehicle embody a lot of untested assumptions. When behavioral scientists test their stochastic models, they try to measure the strength and nature of the relationship between variables. This is why linear algebra and “big data” are important subjects. For more fun read Isaac Asimov’s Foundation Series. We are not their yet!

As stressful as the process is, you might want to give college admissions a break! :-t

372

Not sure how self-reporting of test scores (with verification by official score report if the student is admitted and matriculates) really affects this discussion. While it is desirable in lowering costs for applicants, a college doing test scores this way could still be test-required or test-optional independently of doing it this way. Indeed, using this method may be preferred by a college that wants to be test-required but not want to have the cost of the score report be a barrier against low income students applying.

373

^it kind of removes one hurdle being held against testing

374

Here you are the one saying the testing required schools are wrong. The tests are unneeded because you say so? Or a given group of colleges have made the optional (for a number of reasons)? Not sure I understand the reasoning here. Because X schools do something every school should do it?

I don’t see people saying testing optional schools are wrong. Just different. You seem to be the one saying that because some schools are doing it and view tests as unnecessary, all schools should do so.

The Bates article is interesting. But again, different colleges take different approaches. That doesn’t make one right and the other wrong. And what is in the best interests of one isn’t necessarily in the best interests of all.

375

@SatchelSF

Yes! I got me some MIT grad amigos might have a little issue with your 20/80 breakdown, but I’m gonna use it on them for sure next time I see them! “Hey, Mr. 75%… should gone to UMass Amherst like I suggested and saved a few bucks! I think your ‘halo’ was a ‘failo!’” lol.

I know you love the “tippy top award winning” pointy head stuff - and more power to you, but that is not at all what we are talking about here.

We are talking about the other 1,599,000 kids who take the SAT every year to whom just getting defered at MIT would be a halo worthy of Bondone. And maybe MIT should be all F - they’d have a better grad rate and climb a point or two in USNEWS!

We are all just digging our trenches deeper for the most part around here. We know what we believe and now are just shading the argument to favor our stats and positions.

But I agree with @saillakeerie that choices are the best thing and the more kids have the option of going test optional, esp those that know their GPA is strong and expect it to stay that way, a few more mornings for Saturday morning sleep-ins, or watching cartoons, or getting out of town for a hike up a nice cool canyon.

376

The Caltech reference you linked about men having a higher grad rate than women, discussed the 1980s and 1990s. There have been many changes in regards to women in STEM in more recent decades, and I expect you know that in more recent times women at Caltech have a higher graduation rate than men because the reference below was the first one that came up in my Google search – https://iro.caltech.edu/gradrates . It found the following average graduation rates across all years by gender:

Female Average Graduate Rate – 94.1%
Male Average Graduation Rate – 91.3%

I would not assume economics majors at Harvard are a good representation of the general trends for gender based GPA differences. There are many examples to the contrary. For example, the first link in my Google search for ‘first year gpa gender’ is https://clas-pages.uncc.edu/rootsofstem/wp-content/uploads/sites/529/2013/04/SSS-2013-first-year-grades-presentation.pdf , which discusses gender GPA differences among 21,339 students at NC public schools. It found the following first year GPAs by gender.

Female Overall GPA: 2.67
Male Overall GPA: 2.42

Female STEM GPA: 2.35
Male STEM GPA: 2.19

Female non-STEM GPA: 2.83
Male non-STEM GPA: 2.54

Sure it’s theoretically possible, that the women were all taking easier first year courses to protect their GPA… even though women had higher GPAs than men in the “harder” STEM classes, but this seems unlikely at a typical public university, where women typically have more rigorous HS curriculum, with less need for remedial classes. This seems especially unlikely when you consider how the overwhelming majority of similar studies/reports came to similar conclusions.

If you reread your comments and my reply, it should be more clear, but it’s not worth elaborating here.

377

@saillakeerie

Sure - except to the point that the evidence does not indicate what the SAT suggest it does.

I am under no illusion that all Ivy, Patriot and NESCAC will be so moved by my fact-based rhetoric they will suddenly drop testing requirements… but I think they could!

378

@CaliDad2020 I see you’re still fighting the good fight against testing. Have you thought of joining forces with the anti-transcript revolutionaries?

https://www.insidehighered.com/news/2017/05/10/top-private-high-schools-start-campaign-kill-traditional-transcripts-and-change

379

The rates I posted were engineering gpas, the ones you cited were general. That is why I chose that paper. (btw women were doing better in the 80s than the early 90s according to that paper)
I chose the Harvard paper since it was more clearly addressing issues noted here than the others. I didn’t post them so that you’d agree with them… I only presented to demonstrate such results exist.

Finally, I know you were referring to two different aspects of that paper. My point is that viewing those actions from a different perspective, to me you were attempting to lend more credibility to the paper on two fronts, both by mentioning that they were explained well despite the fact that the data wasn’t shown. But you’re right that this adds little to the discussion; I too will drop elaborating further.

380

LOL @LadyMeowMeow that was my high school. Pass/no pass grades only…we also got about a page of handwritten comments from each teacher.

FWIW we did just fine in elite college admissions.

But that article describes something else - a list of abilities (“masteries” I guess), not courses and grades. Really just a different way of describing the same thing.