"Race" in College Applications FAQ & Discussion 12

Applying to College
4081

So to summarize, none of the recent posters here agree with the premise of this article (https://www.outsidethebeltway.com/why-more-americans-dont-major-in-the-math-and-science/) that to major in math/science is tough and lots of folks switch out of math/science in part because they may not be capable of completing the major and in part to have an easier time in college and get better grades.

Also none of the recent posters are astonished, like the writer of the article that the MIT grad with the B average in STEM did worse than NYU english major. Data10, reading over the phrase “B average in STEM”, I see now that it is possible that this meant that the MIT grad had a B average overall, but I interpreted this to mean a B average in STEM courses.

The whole point of the article is to talk about averages–to analyze trends, to talk about typical cases. Typically, STEM courses are tougher than nonSTEM courses. Of course, there are exceptions, but the author is talking about general trends, No one, including the author of the article would be surprised if the NYU english major did well in law school admissions if he/she had recently written 2 award winning novels, and was in the top 1 percentile in LSAT scores. Of course there can be incredibly intelligent students at all schools. Perhaps, there’s even an out-of-school janitor who’s also a math genius roaming the halls at MIT who can correct mistakes in the math proofs that have been scribbled in various classrooms.

Also of course, in the example given in the article there is not enough information to know if that ONE specific MIT grad is more capable than that ONE specific NYU grad. That’s not the point. It’s not the premise of the article, nor the seed of my question. The premise/question was a what-if based on the average/typical case. To rephrase and be more specific: would it be surprising if a cohort of 100 consecutive MIT grads majoring in STEM with a mean GPA of say 3.3 (with a science/math gpa of 3.0), ON AVERAGE, did less well on law school admissions than a cohort of 100 consecutive NYU english majors with a mean GPA of say 3.8? (The article does not say the NYU grad had a 4.0, but merely implied his/her gpa was higher than the MIT grad). The author of the article and I would say, no, this is not surprising, in large part, because of the importance of the GPA in professional school admissions, regardless of the difficulty of the curriculum.

The implication in the article, however, is that again on average, the 100 MIT STEM grads are at least as capable (and perhaps more so) than the 100 NYU nonSTEM grads with higher grades. The article after all describes the phenomenon of relatively bright students who can’t complete a STEM major who then do better (in terms of GPA and future professional school admissions) by switching to easier non-STEM majors. There would be no point to the article if the message was simply, it’s obvious that students with higher GPAs are just more capable that students with lower GPAs regardless of the major and school tier.

Based on graphs from the reference below, it appears that students that start off as science majors are close to 3 times more likely to change their major compared to those who start off as humanities majors
https://www.businessinsider.com/science-majors-and-grade-inflation-2013-7

4082

More capable of what?

As a graduate of a top law school, I would not agree with the premise that 100 MIT STEM graduates would be as capable at law school as 100 NYU nonSTEM grads with higher grades, nor that they would be particularly likely to become more capable lawyers. The cognitive strengths required for law school and lawyering are very different than the cognitive strengths required for most STEM fields outside the behavioral sciences (psychology, sociology, etc.). This kind of depends on how you define STEM – NSF would include social sciences like anthropology & economics as well – but I think most people on CC are using STEM to mean math, lab sciences, and technology.

But the point is, I wouldn’t hire an engineer to handle my legal affairs, nor would I consider an undergrad STEM degree to be of any relevance in choosing an attorney. In specific fields of law --sure – probably a useful background for a patent lawyer and might help with litigation or contracting tied to other legal specialties.

So I do see it as rather dumb assertion.

I’d also note that I started out college on a STEM track but only took one quarter of chem . Chem was hard but it was not my most mentally challenging course. The first year course that really left my brain spinning was History of Western Philosophy. Kant, to be exact.

And in 4 years of college, I only once dropped a course because I was concerned about the grade & my ability to pass. That was a sociology course. (I had a B+ in my chem 1 A class. I wasn’t worried about the grades when I dropped chem 1 B – the problem was a combination of lack of interest and the large lecture-hall format – once I discovered that I could enroll in most upper-level non-STEM courses without having to worry about prerequisites, that was pretty much the end of overcrowded intro level coures for me).

And my first year at law school was far more challenging mentally than anything I had ever been exposed to in undergrad. Including chem. Because the process by which laws are made, interpreted and applied is analytically very different and I think far more complex than scientific analysis. In law you can’t run a formula and come up with a number to solve problems. In science, the rules don’t change depending on state borders – the laws of physics in Alaska operate the same way as the laws of physics in Alabama.

I mean, why on earth would you expect law schools to look in any way favorably at hard-STEM coursework or graduates of tech schools? As opposed to all the students who have majored in stuff that provides such a great foundation for law, such as history, classics, philosophy, poli sci, rhetoric, linguistics, etc?

4083

Mr. Spock did a very nice job in the trial portion regarding Captain Pike in the Menagerie episode of Star Trek. He was pretty STEM. :-??

4084

Well, actually I’d be happy to give a critique of all the things the scriptwriters for Mr.Nimoy got wrong from a legal perspective* – but t.v. & movie depictions of lawyers tend to get just about everything wrong in any case. I think most lawyers like the trial scenes from My Cousin Vinny the best… because though not realistic, that at least manages to depict an inexperienced lawyer getting things right. (plus it’s also a good portrayal of an expert witness being totally flummoxed because of real-world facts outside the scope of his expertise).

  • (Though, of course, my legal education didn't include 23rd century inter-stellar legal procedure... but I seem to remember at least one rather large gaping legal plot hole in that particular episode. Although that might have been resolved by the disclosure in the end that the entire "trial" had been a sham and a ruse)
4085

@calmom :slight_smile:

4086

If somebody questioned that many switch out of math/science, I didn’t see the post. In previous posts in this thread, it’s been well established than many students switch out of math/science type majors, even at highly selective colleges. This switch out of STEM rate is often especially high among URMs and women at highly selective colleges. The part that people have a problem with is being “astonished” that an English major at NYU could have better law school admission success than a B student at MIT. Or assuming that a B student at MIT should be a more capable attorney than an A student at NYU.

In the most recent MIT post graduate survey, fewer than 0.6% of students said than had or planned to apply to law school. It’s not a common path to go from MIT to law school. The few MIT law school applicants likely have a unique history that involves switching to a less STEM centered field than they had planned on during high school, which can involve negative issues as well as positive. For example, maybe the MIT kid struggled in STEM in a similar way that the article describes. Had he attended HYPS… he would have likely switched to another major. However, it’s not as easy to avoid STEM at MIT, as ~95% of students at MIT major in STEM, and the few MIT humanities majors involve a large number of STEM classes. Maybe after the lack of STEM success at MIT, he decided CS/engineering type career wasn’t for him and instead applied to law school. However, with a mediocre GPA and mediocre LSAT, his law school acceptances were also mediocre. Some kids from HYPSM… type colleges do not have stellar LSATs. Note the lower end of the LSAT range in the table below. All we know is the article claims that a B average kid at MIT didn’t have great success in law school admissions. This is far too little information to be “astonished.”

Given the NYU kid’s law school admission success, I’d expect both GPA and LSAT were quite high. We don’t know how high, but near perfect scores are possible. In the 2 most recent years, the maximum LSAT score from NYU undergrads was 179/180, which is >99.9th percentile. If we look at averages, the average kid from MIT probably has a higher LSAT than the average NYU kid, but I wouldn’t assume the same among MIT kids with GPAs towards the bottom of the class.

If anyone is interested, stats for the most represented undergrad colleges among ABA applicants are below for 2017.
Highest Mean LSAT

  1. Yale – 95th percentile (Range 40th to >99.9th), Median GPA = 3.73
  2. Harvard – 94th percentile (Range 37th to >99.9th), Median GPA = 3.69
  3. Princeton – 93rd percentile (Range 37th to >99.9th), Median GPA = 3.55
  4. Chicago – 93rd percentile (Range 21st to >99.9th), Median GPA = 3.60
  5. Stanford – 92nd percentile (Range 12th to >99.9th), Median GPA = 3.71
  6. Dartmouth – 92nd percentile (Range 23rd to >99.9th), Median GPA = 3.67
  7. Columbia – 92nd percentile (Range 1st to >99.9th), Median GPA = 3.70
  8. Duke – 92nd percentile (Range 21st to >99.9th), Median GPA = 3.69
  9. Penn – 91st percentile (Range 18th to >99.9thh), Median GPA = 3.68
  10. Tufts – 91st percentile (Range 21st to >99.9thh), Median GPA = 3.62
    …
    NYU Arts & Sciences – 84th percentile (Range 6th to >99.9th), Median GPA = 3.58

The MIT post graduate survey suggests less than 10 students per class from MIT apply to law school, so 100 is not a realistic sample size. Ignoring the sample size issue, the table above indicates that NYU CAS had a median GPA of 3.58 among ABA applicants and mean LSAT of 84th percentile. 3.58 GPA is also almost dead on the expected mean GPA for all NYU students based on reports from GradeInflation.com . A 3.8 GPA is 90th percentile at NYU CAS, according to CAS Latin honors cutoffs. GPA has some degree of correlation with LSAT score, so the average LSAT will no doubt be higher among 3.8 GPA kids at NYU than among 3.58 GPA kids. We can only guess at how much higher LSAT than 84th percentile, but I’d guess likely exceeding well over 90th percentile… exceeding the listed mean LSAT for HYPS… and likely MIT. Similarly it’s safe to assume the kid from MIT with a lower GPA would on average have a lower LSAT than the MIT mean. If the MIT mean is anything like the HYPS mean, then this does not bode well for the lower GPA MIT kids’ average law school admissions.

4087

race shouldn’t be considered in the application process. it’s very unfair.

4088

Any field of study that would fall for the Sokol Hoax and its protege would be suspect. I think Anderson was alluding to them when he mentioned how some disciplines exist for the sole purpose of giving As. His “fatherly advice” section gives some very vivid examples.

This is true only if you decide to group them according to N=150+. If I were to group them by LSAT scores of 160+, Every single one of them exception 2 are in STEM, with Classics at 9th and Industrial Relations at 12th. I am very surprised that philosophy and economic are not among them, btw.

The real lesson for looking at LSAT scores is that it demonstrates something that industrial psychology has known for a long time- that it is not necessary to test specific skills with specific tests; a general test of cognition would do. How else can one explain the outstanding performance of STEM in an area that is their “weakness”? Aren’t English majors suppose to blow everybody away?

To get a fairer assessment of relative ability, i.e. bringing number skills into the equation, and for a larger sample size, I turn to Prof Mark Perry’s GRE scores by major for verification:

http://mjperry.blogspot.com/search?q=GRE+scores

Just as I would expect based on what I see from the LSAT scores.

If anyone has a newer version of this data, please post it.

4089

I’ve seen similar types of studies over the years (but can’t find any links). The results of these studies are fairly consistent and unsurprising. In fact, if non-native speakers, most of whom are STEM majors, were taken out of these studies, STEM majors would score even higher with better verbal scores. Even the orders within STEM fields are not surprising. Physics, especially the theoretical side, is generally the most challenging of all fields. Even the most elementary physics at the level of AP Physics C is commonly considered the hardest courses in most high schools. And the difficulty increases exponentially from there. Most engineering fields, by comparison, are cakewalks.

It’s also not surprising that STEM majors will do reasonably well on LSAT. They generally have good comprehension and reasoning abilities. They may not have the same verbal skills as the English majors, however.

4090

I don’t know if the LSAT has changed over the past few decades, but it seemed to me that it was mostly logic. I have long thought that high scores associated with certain majors – or undergrad institutions, for that matter – indicate a correlation rather than causation.

4091

Seriously.

What is a STEM kid or liberal arts kid?

Is there a genetic test I missed when my daughter was born?

Obviously being facetious, but my experience with her as a so called “smart” kid - she was encouraged, had societal influence and received praise for choosing STEM. Especially as a young women, it’s a thing.

So if our more gifted students are steered this way - as a percentage and not all of course - is this another example of correlation not causation.

Many of these students may actually have excdllent writing skills too. They are bright. And they are good test takers.

Not sure what this has to do with AA and or race preferences in admissions, I am not sure.

For the record I do not believe it’s fair. I do believe that it works to deprive other potentially qualified kids. But hold on !

But I still think it’s the right thing to do for the time being.

It’s good for the country. It’s also may be a small step in helping our inner cities break the cycle of poverty and violence.

But I do have a caveat.

You lose the preference of you’re wealthy at say x income level as a family. Period.

4092

There has been a suggestion that URM kids are steered to non-STEM majors and that this is a Bad Thing.

4093

@OHMomof2

I think there are so many variables in that conversation.

How many crappy schools serving underserved communities have great support or teaching in stem.

I don’t know that answer but I would guess not so much.

4094

That is because the college courses that AP physics C emulates have prerequisites higher than most high school students (even those taking AP courses) have, so high school AP physics C curriculum and courses need to work around missing prerequisites.

AP physics C mechanics: emulates calculus-based physics 1, which typically has calculus 1 or AP calculus AB as a prerequisite and calculus 2 or the rest of AP calculus BC as a corequisite.

AP physics C E&M: emulates calculus-based physics 2, which typically has calculus 2 or AP calculus BC as a prerequisite and calculus 3 (multivariable calculus) as a corequisite.

Physics is often easier to understand if you have enough math knowledge, but AP physics C is commonly taught to students who are just starting calculus concurrently, not the optimal time to be learning calculus-based physics.

4095

Math is just a tool. Mastering math helps understanding physics, but it’s far from sufficient. Truly understanding the physics (of physics) is where the difficulty lies. At the AP Physics C level (or the college freshman/sophomore levels), the problem may not be obvious, but the students who rely mostly on their math ability without comprehensive understanding of underlying physics will find themselves struggling beyond these basic courses in physics.

4096

I was writing that the math is necessary, not that the math is sufficient, to understand the physics. Certainly, many students with enough math will have difficulty with physics, but students who have less than the necessary math are likely to struggle in physics even if they would have understood it if they had enough math.

The way AP physics C is set up in high schools pretty much guarantees that most students will be playing catch-up with the necessary math instead of being able to apply math that they know to try to understand the physics.

4097

The official GRE scores as listed by ETS are the first link in a Google search - https://www.ets.org/s/gre/pdf/gre_guide_table4.pdf – and they are completely different than the very old blog post. For ties, I ordered by the additional sections.

**Highest Total GRE/b

  1. Philosophy – 434.8
  2. Religious Studies – 427.0
  3. Political Science – 427.0
  4. Physics – 425.7
  5. Humanities – 425.2
  6. English – 425.0
  7. Art History – 423.2
  8. Economics – 421.6
  9. Foreign Languages – 420.3
  10. Library Science – 419.3

**Highest GRE VR/b

  1. Philosophy – 152 to 166
  2. English – 150 to 164
  3. Humanities – 149 to 165
  4. Library Sciences – 150 to 164
  5. Art History – 150 to 164
  6. Religious Studies – 150 to 164

**Highest GRE AW/b

  1. Philosophy – 3.5 to 5.1
  2. English – 3.4 to 5.0
  3. Religious Studies – 3.4 to 5.0
  4. Political Science – 3.4 to 5.0
  5. Humanities – 3.3 to 4.9
  6. Art History – 3.3 to 4.9

**Highest GRE QR/b

  1. Mathematics – 157 to 169
  2. Materials Engineering - 157 to 169
  3. Physics – 156 to 168
  4. Finance – 153 to 169
  5. Economics – 153 to 167
  6. Engineering Other – 153 to 167

Note the different scores by section above. The mathematics majors tend to really well on QR, but not as well on the other sections; while the English majors tend to really well on VR and AW, but not as well on QR. A single general test would not capture these specific subject strengths.

4098

Steering back towards the topic of this thread, the URM percentages among bachelors degrees in the 10 highest average total GRE score majors as listed in the previous post are below for the most recent year in the NCES chart – https://nces.ed.gov/programs/digest/d18/tables/dt18_322.30.asp?current=yes . There are definitely racial statistically significant racial differences by major, but it doesn’t follow average total GRE score well. There appears to be more a STEM/Vocational vs Humanities type pattern than GRE score pattern. Asian percentages tend to be highest in STEM/Vocational type majors, particularly CS and Engineering, while Black percentages tend to be highest is less STEM centered majors, as well as more interpersonal majors (law enforcement, social services, …).

Racial Percentage in 10 Highest Average GRE Majors
Philosophy + Religious Studies – 67% White, 9% Black, 6% Asian
Physical Sciences – 65% White, 5% Asian, 9% Asian
Humanities – 61% White, 15% Black, 4% Asian
English – 69% White, 8% Black, 4% Asian

Foreign Languages -- 56% White, 5% Black, 6% Asian

Average of All Majors – 61% White, 10% Black, 7% Asian

4099

Interesting, since they replaced the old analytical section (when I took it in the 87-ish I think) with the analytical writing section, the stem students don’t do as well. It was not uncommon for many stem students to breeze through the GRE in the good ole days (math/verbal/analytical, 2400 scale). tbh, I think the writing does balance things out. Students would take the test without any prep since we were taking classes, and no I was not one of those students, otherwise I’d be humble bragging about it.

“There has been a suggestion that URM kids are steered to non-STEM majors and that this is a Bad Thing.”

If you think gender diversity is bad in high tech, racial diversity is even worse. There may be 3-4% urm in tech, and of that, probably 1-2% in actual product groups (pdt mgmt, engr, dev, etc) that can influence what products are built, how products are designed.

4100

I didn’t realize GRE has changed so much either. It’s been so long that I have little recollection what the actual tests look like. Perhaps it shouldn’t be so surprising since SAT has also changed so much over the years.