MCATs, Affirmative Action, and DEI, Oh My

[Derec]

Better at taking tests is not the same thing as being better at treating patients, which is what most physicians do.

How do you propose testing somebody's knowledge and reasoning other than with some kind of test?

You can't evaluate people trying to get into med school based on how they are "at treating patients" anyway.

 

[Derec]

Those are during medical school. People in this thread are whining about the testing regarding admission into medical school.

What would you base medical school admissions on instead?

GPA (and also transcripts) test how well a student did in their undergrad education. The drawback of that is that there is often a huge difference in grading and standards between schools, but the advantage is that it looks at the entire college career. The MCAT tests prospective med students in four categories: Chemical and Physical Foundations of Biological Systems, Critical Analysis and Reasoning Skills (CARS), Biological and Biochemical Foundations of Living Systems (Bio/Biochem) and Psychological, Social, and Biological Foundations of Behavior (Psych/Soc). The advantage is that everyone takes the same test, so the results are more comparable than GPA. The disadvantage is that it is testing performance over the course of one day on a set of topics that is more limited than the entire undergrad coursework one might take.

So, both of these metrics are important, and together they are a great measure of a student's academic preparedness.

 

[Derec]

I do think MCAT scores should be adjusted.

Adjusted to what? Race?

That said, in order to graduate with an MD, there should be an EXIT EXAM that everyone must pass regardless of race, gender, age, nationality.

There already exist licensing exams. You can't advance to clinical rotations without passing USMLE Step 1 (or COMLEX Level 1 if a Bone Wizard aka a DO). You can't graduate without passing USMLE Step 2 (or COMLEX Level 2 if DO). In addition, since USMLE is much better known by residencies than COMLEX, many DOs end up taking Step 2 and some even take Step 1 in addition to their required licensing exams.

MCATs are to GET INTO med school and since we know that public and private education systems are not equal, there would be unfair advantages that have little to do with actual intelligence or ability to be a doctor.

I do not see how MCAT is unfair. Everyone takes the same test - it doesn't get fairer than that. There are also very good state colleges and not so good private ones.

 

[Derec]

If you ‘just want applicants to be seen as individuals,’ why do you reduce them to test scores and race?

I am not reducing them to that.

I am looking at MCAT and GPA because these are objective measures of academic preparedness. I am looking at race to show that medical schools are discriminating by race. If they weren't, we'd expect similar average GPAs and MCATs across racial and ethnic groups.

Why do you not believe that medical schools are highly invested in admitting into their program only students they think will succeed?

I think they are indeed "highly invested in admitting into their program students they think will succeed". Note that I omitted "only" from the quotation. Sure, they are invested in that. But that's the only thing they are invested in. They are also invested in the ideology that says that certain groups should be given preferential treatment. Whether their reason is supposed redress of past racism or diversity, their stated reason will mostly be diversity because of the loophole given them in Bakke.
So, all things being equal, adcoms want students they think are more likely to succeed. But they also want to give a push to black, Hispanic and also American Indian students, resulting in the unequal distributions we see in MCAT data.

If MCAT scores are not sufficient, why not put your faith in the more rigorous USMLE exams which are given at different stages of the med students medical school education?

Is that a serious question? How can you base medical school admissions on a test that will not have been taken until more than two years in the future?

You have no evidence that students are admitted who are not up to the eulogies of med school and residencies.

Lower MCAT and lower GPA have a higher chance of not passing Step 1 and also of not graduating within five years.
And we also know that black students have a 250% higher chance of dropping out after the first year.

You just cannot stand that some black or brown person might be a better candidate than a white or good minority Asian.

Bullshit.
Of course there are plenty of "black or brown" students who are better than some other white and Asian students. And it is a disservice to them too that "black or brown" students with a mediocre academic are admitted just because of their race.

 

[Derec]

Not necessarily. Suppose, for example, the primary reason that people drop out is economic. Suppose also that most of the correlation you observe for MCAT/GPA and drop out is due to economics correlation with MCAT and GPA.

A lot of "suppose" here. Is there any reason to "suppose" such a correlation?

Then, in that instance, it's not the case. Have you looked into the economics and correlations with MCAT and drop out rate etc to investigate this?

You are the one making the supposition. Why don't you look into it?

But as to the data accessible to us from AAMC: let's look into a metric where economic factors do not play a role: Step 1 pass rates. Those people all completed their med school up to that point and signed up (and paid for) their Step 1 registration. But they are more likely to fail if their MCAT was lower. Can't blame economics for that!

 

[Bomb#20]

Thought Experiment

Take the above but ignore the bottom rows:
View attachment 55248

We observe small differences between the White and Middle Eastern/North African MCAT and GPAs, but an unexpectedly large difference in acceptance rate. Do we comment "Islamophobia is for realz!", "Whites clearly are rigging the system," or do we try to delve into how this is happening?

I tracked down a website that shows the averages for all applicants, not just for accepted students.

Ethnicity MCAT GPA
Asian 508.8 3.72
White 508.0 3.72
MENA 505.3 3.69​

So those 2.7 MCAT points difference are the prima facie explanation for the acceptance rate difference. There are undoubtedly other factors having an effect too, but I'm not seeing any obvious anomaly.

As far as "Islamophobia is for realz!" goes, keep in mind this is the first year MENA has been pulled out as a separate category -- in previous years MENA students were generally counted as White. So there hasn't really been time for institutional momentum to develop around using MENA numbers to virtue signal.

Likewise, for a comparison between Whites and Asians: Whites have a higher acceptance rate but slightly less GPA and less MCAT. Do we say this is due to racial preferences of propping up White people?

That's a weird way to look at it. Asians are discriminated against for the sake of propping up the ethnic groups the fans of racial preferences prefer, i.e., African Americans, Latinos and Native Americans. Deleting those lines from the table doesn't change the preferences of the people doing the preferring. The net effect of racial preferences on white people looks pretty negligible -- the admissions committees appear to be getting all the extra slots they need from Asians.

 

[Derec]

I assume you mean “non-academic reasons” in the start of your syllogism. Otherwise, it is a tautology.

No. Read it again.

It seems to me, you and your ilk think causes have distinct and separate influences on human decision making like admissions or dropping out.
I don’t think that assumption is valid. If influences intermingle or reinforce each other, your analysis falls apart. Don’s post gives an example of such an effect.

Just because influences may intermingle does not mean we can't look at different variables.
What is an alternative? Somebody else from your Ilk posted a link that used self-reporting about reasons for dropping out to make their point. That is surely far less reliable.

Since the data shows failure to complete medical school within 6 years and since there are many obstacles to reaching that goal, your analysis is dependent on the assumption that academic reasons are the primary factor in failure to complete in 6 years.

Five years. And of course there are other reasons (which is why even some high scorers fail to graduate in 5 years or less) , but there is no reason to assume that these reasons are dependent on somebody's MCAT score. There are reasons to think somebody's academic performance is strongly linked to their MCAT and undergrad GPA though.

Or look at Step 1 pass rates. We also see a dependence on MCAT and GPA, but I fail to see why there should be a non-negligible non-academic reasons to fail this exam.

Finally, even if your corrected syllogism is an accurate description of the actual failure to complete process, if the non- academic reasons subset of non- completes comprise a significant portion of the entire set, then it seems you’ve not shown very much. First example, if non-academic reason non-completes comprise 90% of the entire set, then the MCAT/GPA is doesn’t explain much.

You are making a lot of far-fetched assumptions here. All in the service to dismiss the obvious conclusion that people who do well in undergrad also tend to do well in med school.

 

[Toni]

If you ‘just want applicants to be seen as individuals,’ why do you reduce them to test scores and race?

I am not reducing them to that.

I am looking at MCAT and GPA because these are objective measures of academic preparedness. I am looking at race to show that medical schools are discriminating by race. If they weren't, we'd expect similar average GPAs and MCATs across racial and ethnic groups.

Why do you not believe that medical schools are highly invested in admitting into their program only students they think will succeed?

I think they are indeed "highly invested in admitting into their program students they think will succeed". Note that I omitted "only" from the quotation. Sure, they are invested in that. But that's the only thing they are invested in. They are also invested in the ideology that says that certain groups should be given preferential treatment. Whether their reason is supposed redress of past racism or diversity, their stated reason will mostly be diversity because of the loophole given them in Bakke.
So, all things being equal, adcoms want students they think are more likely to succeed. But they also want to give a push to black, Hispanic and also American Indian students, resulting in the unequal distributions we see in MCAT data.

If MCAT scores are not sufficient, why not put your faith in the more rigorous USMLE exams which are given at different stages of the med students medical school education?

Is that a serious question? How can you base medical school admissions on a test that will not have been taken until more than two years in the future?

You have no evidence that students are admitted who are not up to the eulogies of med school and residencies.

Lower MCAT and lower GPA have a higher chance of not passing Step 1 and also of not graduating within five years.
And we also know that black students have a 250% higher chance of dropping out after the first year.

You just cannot stand that some black or brown person might be a better candidate than a white or good minority Asian.

Bullshit.
Of course there are plenty of "black or brown" students who are better than some other white and Asian students. And it is a disservice to them too that "black or brown" students with a mediocre academic are admitted just because of their race.

Sigh. You are assuming that the students all apply to the same medical schools. You’re assuming that they all make the same number of applications, that all medical schools have the same criteria for selecting students.

This has all been discussed most eloquently by Don2 who is actually better informed than any of us.

 

[Derec]

They also explicitly say schools look at experience, attributes and other factors, not just test scores.

That's fine. But these metrics would have to be correlated opposite to MCAT/GPA in order to make up for the difference in the average MCAT/GPA for accepted students in the absence of racial discrimination. Do you really think black admitted students have, on average, superior experience compared with Asians?

Also, subjective metrics can be gamed to accomplish "race conscious admissions". Remember Harvard's infamous "personality scores"?

Where Derec loses me is going from “these groups have different average MCAT/GPA scores” to “therefore this proves racial discrimination.”

In the absence of what I described above, that is certainly true. Especially when med schools are not exactly coy about believing in "race-conscious admissions". The Ilk here is tripping over itself to deny something that people who do it are not even denying they are doing.

It might give you a reason to ask questions, sure. But I don't see how the averages alone prove what caused them. Don2 has pointed out, and I agree, that lumping together schools with different applicant pools, missions, state preferences, standards, etc. hardly explains anything.

I do not think Don's argument holds water, because all applicants apply to a variety of schools and select them based on their GPA/MCAT compared with school averages. It's not like all Asians are gunning for T20s exclusively.

Emily argued, and I agree, that different group averages can produce different results even if everyone is being run through the same threshold.

Emily has shown that with an example of two distributions where one was much wider (bigger SD) than the other. It does not work with the admitted student distributions because SDs are similar, but means are far apart.

I don't buy the argument that MCAT doesn't matter much because almost everyone who gets into med school graduates. That's swinging too far the other way. It clearly matters. I just don't think a few extra MCAT points automatically means “better future doctor.”

It matters to academic preparedness and aptitude which in turn makes it easier to master the med school material. But of course, that is the necessary condition for a good future doctor. There are obviously skills not captured in the MCAT or undergrad schooling that is important.
But it's impossible to assess those things in undergrads.

And if you're going to use subjective stuff like interviews, experience, leadership, “fit,” etc., then IMO that stuff needs to be as structured and transparent as possible. Otherwise it can very easily produce town criers. Right Derec?

Unironically, right. What's wrong with transparency?

Anyway, we're only dissecting these MCAT/GPA averages because they're being used as evidence against affirmative action and DEI. Fine. But the averages themselves don't tell us what caused them. Maybe AA/DEI played a role. I'm not saying they didn't. I just don't think pointing at different racial averages gets you from “here's a disparity” to “AA/DEI exists" to "AA/DEI caused it.”

In the absence of a viable alternative hypothesis for the large and persistent discrepancy, and with known history of med schools (along with other academic institutions) practicing "race-conscious admissions", I do not see why this conclusion is premature. If anything, denying it is just willful blindness.

 

[Derec]

What do YOU mean by ‘less qualified if preferred races? I mean, historically that has always happened.

Why do you always keep harping on what happened a long time ago?

FFS, women long wrote and published under male pseudonyms or disguised themselves as male to gain entry into educational establishments.

Oh boy. We are back in the Victorian era now with George Elliot and them.

FFS, Sandra Day O’Conner could not find work as a lawyer after graduation and finally found a job—without pay! and shared an office with a secretary, writing memos. Surely she was as well qualified and far more talented than the majority of her male classmates.

Even she was born almost a century ago.

So whenever I read/hear people complaining that ‘less qualified’ candidates get hired over better candidates, I understand the ‘ better candidate ’ is white and male—because he is male and white. Or perhaps Asian.

No. We are saying that it is wrong to discriminate on the basis of race or gender.
Discrimination in the past was wrong. Discrimination today, whether it is justified as "payback" or "diversity", is just as wrong.
You are accusing people of ugly motives because you still live in the past and can't comprehend just how radically things have changed, especially in academia.

 

[bilby]

How do you propose testing somebody's knowledge and reasoning other than with some kind of test?

Why do you think knowledge and reasoning are the only, or the most, or even significantly important indicators of which already qualified candidates are likely to become the best doctors?

You can't evaluate people trying to get into med school based on how they are "at treating patients" anyway.

Sure; But you can evaluate them based on their interpersonal skills and their interactions with other humans - one time honoured way to do this is by interviewing them.

Success at interview isn't only about answering the interviewer's questions in such a way as to be technically correct.

Interviews are not a great way to assess candidates - but then, nor are standardised test scores.

 

[laughing dog]

What do YOU mean by ‘less qualified if preferred races? I mean, historically that has always happened.

Why do you always keep harping on what happened a long time ago?

Homo sapiens have been around for 300,000 years or so. So, 30 20 to 60 years past is not "long ago". And whether you like it or not, racial bigotry is still alive and kicking. The resurgence of white nationalism and the election of Mr Trump are two of the more obvious signs.

 

[laughing dog]

I assume you mean “non-academic reasons” in the start of your syllogism. Otherwise, it is a tautology.

No. Read it again.

It seems to me, you and your ilk think causes have distinct and separate influences on human decision making like admissions or dropping out.
I don’t think that assumption is valid. If influences intermingle or reinforce each other, your analysis falls apart. Don’s post gives an example of such an effect.

Just because influences may intermingle does not mean we can't look at different variables.

It means that properly interpreting them is much more complex than your analysis.

What is an alternative? Somebody else from your Ilk posted a link that used self-reporting about reasons for dropping out to make their point. That is surely far less reliable.

The alternative is to be more careful in one's analysis.

Since the data shows failure to complete medical school within 6 years and since there are many obstacles to reaching that goal, your analysis is dependent on the assumption that academic reasons are the primary factor in failure to complete in 6 years.

Five years. And of course there are other reasons (which is why even some high scorers fail to graduate in 5 years or less) , but there is no reason to assume that these reasons are dependent on somebody's MCAT score. There are reasons to think somebody's academic performance is strongly linked to their MCAT and undergrad GPA though.

Wow. You admit that the data on failing to complete in medical school is not a clean indicator of academic performance, yet persist in looking at MCAT and GPA scores. Your hobby horse must have pretty good padding on it since you've been riding it so hard.

Finally, even if your corrected syllogism is an accurate description of the actual failure to complete process, if the non- academic reasons subset of non- completes comprise a significant portion of the entire set, then it seems you’ve not shown very much. First example, if non-academic reason non-completes comprise 90% of the entire set, then the MCAT/GPA is doesn’t explain much.

You are making a lot of far-fetched assumptions here. All in the service to dismiss the obvious conclusion that people who do well in undergrad also tend to do well in med school.

I made no assumption. I pointed out that your analysis ignores many possible causal factors because of your obsession with race. Unlike you, I don't pretend to know the underlying causes, but as many others have pointed out, the requisite date is either insufficiently granular or nonexistent.

 

[ZiprHead]

That sounds like you deny the effects of generational racism exists. It's they only way for your point to work.

Whatever these effects are, and I think they are significantly exaggerated by the left, I do not think a solution is to keep holding certain people to a lesser standard based on their race. I do not think it is really helping blacks in the long run either. Racial preferences have existed in one form or another for 60 years or so, and they have not fixed anything.

Really? Black people today are still treated as they were in 1966???

 

[Emily Lake]

Pretty sure the descendants of enslaved people’s, native Americans, Chinese Americans and Hispanic people and Catholics and Jews and more recently Muslims are t terribly fond of being held back, pushed down, lynched, murdered and discriminated because of the sins of others.

Do you have any actual real evidence of this happening in any material numbers? Or is it just plain old narratives?

 

[Emily Lake]

Me thinks you’re reading “nobody” much more broadly than I meant it. I’m talking about this discussion and the people making the AA/DEI argument here, not claiming that nobody anywhere complains about nepotism, favoritism or White people benefiting from connections. My point is that here, when people talk about somebody receiving an opportunity they supposedly didn’t earn, the example keeps becoming a Black or Hispanic applicant taking a spot from a White or Asian applicant. Of course people complain about nepotism and favoritism involving White people. That actually reinforces my point. We recognize all kinds of reasons someone might get an advantage over a supposedly “better” candidate without immediately turning it into a demographic grievance. What I’m questioning is why, here, the outrage over AA/DEI preferential treatment has been so specifically attached to Black and Hispanic people as the “spot takers.” Your hypothetical explains why an explicit racial preference could create suspicion. Fine. I’m against that kind of preference too. But it doesn’t really answer the selective framing I was talking about.

This is exactly why I tend to avoid debating you. You seem to have a real problem paying attention to the details of what I actually wrote. Bomb does the same thing, he took my specific point about White women, where race was part of the point, and quietly turned it into an argument about women in general. It’s a silly way to argue. If you don’t want to address the point I actually made, fine. Just don’t address it. I’d much rather that than have you rewrite my argument into something easier for you to respond to and then argue against that instead. It's not like I'm not willing to concede when my argument is flawed or when I’m wrong. But I can’t do that when the argument being criticized isn’t actually the one I made.

Your post was unclear, and did not seem to be limited solely to giving specific posters here a hard time.

Be more clear.

 

[Emily Lake]

Every straight white guy in America who is ever told "no" believes firmly in his heart that it was because of discrimination against straight white guys.

And in some specific situations, you consistently side with the straight white guys, and present them as the horribly oppressed group.

 

[Don2 (Don1 Revised)]

Not necessarily. Suppose, for example, the primary reason that people drop out is economic. Suppose also that most of the correlation you observe for MCAT/GPA and drop out is due to economics correlation with MCAT and GPA.

A lot of "suppose" here. Is there any reason to "suppose" such a correlation?

Then, in that instance, it's not the case. Have you looked into the economics and correlations with MCAT and drop out rate etc to investigate this?

You are the one making the supposition. Why don't you look into it?

From your perspective, I am talking suppositions, but in reality I am already aware of the correlations. I was asking you to be open-minded and stop trying to fit everything into single variable narratives. Life is complex and multivariate. Variables are continua or spectra. More often than not, ideology creates oversimplified narratives of single variables.

I don't want to spoon-feed you because I wanted you to consider your own methodological flaws and learn from them. But here you go...

Few medical students drop out—but when they do, here’s why

Medical school graduation rates exceed 95%. When attrition happens, it’s usually tied to life outside the classroom, not academics.

www.ama-assn.org

Most significant reasons for dropping out: career changes, life changes, financial pressures. Don't think in absolute terms either. None of these are meant to indicate that there is only ONE reason a person drops out and sometimes in less frequent circumstances that could be tied to academics, too. I mean, why not? When people are being rational, they take time to think about their decisions and weigh pros and cons, costs and benefits.

MCAT scores are correlated to SES, which is related to financial pressures:

Finally, according to SES: EO-1, 498.6 (10.6); EO-2, 501.1 (10.8); EO-3, 503.2 (10.4); EO-4, 505.3 (10.3); and EO-5, 507.2 (9.9). Score distributions at the two extremes of SES (EO-1 and EO-5), which together comprise 48.9% of the sample, are depicted in Fig. 1. Score distributions for the remaining three SES groups (EO-2, EO-3, and EO-4) are included in the Supplemental Digital Appendix.

 

[Emily Lake]

Perhaps Asians apply less frequently to schools with no or lower requirements. In fact, HBCUs on average have lower thresholds and very little Asian enrollment. I think it's like 1%. HBCUs are only a subset of such schools.

Crazy idea... but the fact that a college or university promotes itself as historically black in a way that very clearly communicates that they prefer their attendees to be black... maybe that would discourage people who aren't black from applying in the first place?

 

[Emily Lake]

Is there an unspoken assumption here that absent any sort of discrimination, enrollments would be exactly equal between men and women, exactly proportionate to racial disposition in the general population, and so forth? That seems deeply unlikely to me. What would be more concerning to me us if there is evidence that a broad populational group is submitting applications, but being disproportionately turned down from jobs, loans, housung, educational opportunities, grants, or other consistently measurable social goods. I have FAR more female than male students, and more white than non-white students by a slim margin, but that isn't because men or non-whites are being turned down for a seat. It can't be that, because we have open registration in the community college system, no one who manages to complete the application is barred from attending. Presumably, different social and economic forces are being exerted on the various social classes that encourage or discourage enrollment. It drives me crazy when administrators talk about rates of first enrollment as something faculty could conceivably "fix".

I think the issue is that your "unspoken assumption" is being made by those that push DEI policies - that a difference in proportionality of outcomes implies discrimination.