Racial standards in SAT and college admissions - ideal threshold?

It’s been observed that people of certain racial groups need to have higher SAT scores than some other racial groups in the college admissions process.

Now, while it’s impossible to identify a specific math formula that will determine how much more diversity a student body will have, what, in your opinion, would be a fair and proper ratio of how high one group of people needs to score? Bear in mind also that the SAT has recently changed.

It is impossible for there to be perfect fairness, nor for all groups of people to have their scores weighted equally in admissions…but how large a gap is too large?***

If Race A has to score 200 points higher than Race B, is that an acceptable gap, or too large? What about 350? 120?

Any gap is too large. Discrimination based on race is wrong.


How could we possibly quantify the difference when it’s a brand new test?

Personally, I think all students should have to be over the threshold that the institution has determined indicated that the student has the fundamental skills needed to succeed in their program, and after that threshold, test scores should have a much diminished role in admissions decisions. So, if MIT has found that students who score over 700 have the basic math skills needed to be successful, they shouldn’t really care that much about whether Johnny has a 710 and Bobby has a 770–they should be looking at other factors. Some of those other factors might include developing a class full of diverse perspectives. I’m even okay with one of those perspectives being “The kind of kid who can get a 1600 on his SAT”, because that’s a genuine variation, but it’s not because that kid has more merit or is more prepared, honestly.

True, but on the old test, how large a gap was too large?

Again, it’s about meeting the minimum. No one should be let in who can’t handle the work, but in a lot of schools, a kid who got a 1750 is not less capable than a kid who got a 2310: both can handle the work, so now it’s time to look at other things. The 2310 kid does not have more merit on that basis, nor has he somehow earned something the other kid has not.

Here’s an analogy. If you take a kid out of the finest track program in high school, trained to a fare thee well since he was ten years old, with an personal trainer and nutritionist, he’s going to have a pretty good time in the 400 meter, lets say 00:46. If you take a kid from a crappy high school, where the English teacher doubles as the track coach, and they practice indoors because they don’t have an outdoor track, he might turn in a time of 00:46.5. If I’m a college coach, I’m going to say that the first kid is trained up as fine as he can be, and will never get any better, but the second kid, in a good program, will end up running rings around the first.

Now take a kid from one of those rich suburban high schools. He’s had enrichment programs since he was seven years old. His nearsightedness was diagnosed early and he got corrective lenses.

His (two) parents put him in summer reading programs at the local library, he had a whole spectrum of elective courses, and he’s taken both the Sylvan and the Kaplan SAT training courses. His senior year he went to the community college for entry level college courses.

He ends up getting 1250 on his SAT and a 3.5 GPA.

Now, here’s kid number two. He went to an inner city school, where the bulk of the teachers are entry level educators getting enough experience to qualify for a job at a better school. They turn over 30% of the teachers every year, and the administrators are all just trying to eke out a few years until retirement and usually last around 3 years in the job.

He was nearsighted since he was eight, but because he never saw an optometrist his vision was not corrected until he was twelve. He lives in a single family home, and it’s one of the large number of homes in America without a single book. He never went to the library until he was in high school, and his school not only has no electives, but some of the required courses are held only every other year for lack of instructors.

His SAT score was 1180 and he got a 3.0 in high school.

If I’m an admissions officer, I’m going to think the right thing is to say, “You know, that second kid, if in a good program, has a lot more potential than the first one, who’s already at the limit of his abilities.”

I don’t know if that’s the way it always works, and one problem I have with affirmative action programs as executed is that in practice they seem better at making it easier for rich black kids to get into school than for poor kids of any color, but you can’t just look at the scores and say that one is higher than another, and therefore this student is better than that. Otherwise, why have admissions officers at all? Just add up the scores with a spreadsheet.

I am against any sort of positive discrimination because it discriminates against those who are trying extremely hard but recieve no help and those who aren’t part of a minority ethnic group.

In the UK, we have several ethnic groups from a third world background. Hindus and East Asians are stereotyped as middle class while Muslims (Pakistanis) and Blacks are lower class; Hindus and East Asians do well regardless of income or cultural upbringing, and many of them come from conservative households.

We get everyone from Japanese immigrants working in banks, Gujarati farm owners who sell their farms in Gujarat and move to the UK, Sri Lankan Tamils fleeing war, illegal Chinese immigrants who pick cockles on the beach, etc but at least the 2nd generation are all middle class. It would be unfair to discriminate against them since many of them have troubled upbringings.

Most of the lower class come from Africa and Pakistan, though we also have a more prominent population of lower class whites. It would make no sense, and probably cause an outrage, if we did something like AA because it would be racist and make the end result worse.

I’m going to go with Manda Jo’s answer.

The SAT is not complete bullshit. Just like the IQ score isn’t bullshit. But if I were a K-12 administrator, I’d only use IQ scores to help me predict whether a kid is going to have learning problems–likely someone whose score falls below two standard deviations of the mean. I would not use IQ to tell me whether a student would benefit from arts enrichment or whether he or she should be encouraged to do a science project.

So it seems to me if I were a college admissions officer, I’d use the SAT to identify students who are remedial. Maybe I have a few spots available to remedial students who offer something special (like athletic ability), but a minimal SAT score would be an easy, guilt-free way of winnowing the stack of applicants.

Even schools that are highly selective offer programs that are not the most intellectually rigorous. At my undergrad, very few (if any) brainiac types majored in management, for instance. If you set your minimal requirements too high, you risk leaving some programs under-utilized. So diversifying the student body isn’t just a feel-good goal. It benefits the university as well.

That assumes that, if you don’t use AA, you won’t have enough applicants for the less rigorous majors. It isn’t usually how AA works.

You are quite correct that, if the minimum is set too high, you won’t have enough applicants. AA comes into play when you have enough applicants who can meet the minimum standard, but they aren’t diverse enough.

IOW university X sets a minimum SAT level of 1250 and has 500 slots. They get 500 applicants, but they are all or almost all Asian or white. So they admit 500 applicants, and the brainiacs who scored 1450 go into engineering and chemistry and the rest go into management. But AA proponents would say that this is insufficiently diverse, which is not a problem of under-utilization of university resources. So lowering your standard does not address under-utilization, which is not a problem unless you can’t find 500 applicants with an SAT of 1250.

The issue with AA is that if you want diversity, you will need to lower the minimum standard, either in general or for selected races.

Lowering it in general means that, if 1250 was too high, then those admitted with lower SATs would graduate at the same rates and score just as well in GPA and other measurements. That typically doesn’t happen - those who score below the minimum tend to drop out more and do less well. So the 1250 was not wrong - it was the standard that found out the students who thrive.

Lowering it for selected races means that only those selected races are going to underachieve, drop out, etc., as well as major in less rigorous areas like management. That will not affect enrollment in the less rigorous majors - there were always sufficient applicants for those programs - it is just that, under AA, the less rigorous majors will be filled up with students who fail more than would do if the mimimum standard were kept at 1250. But again, that has nothing to do with under-utilization.


The problem I have with this holistic approach that uses all sort s of other criteria is that it has by and large been used primarily for the benefit of rich white kids. Blacks and American Indians get into schools under pretty explicit affirmative action programs. White males get in under all these other holistic criteria and crowd out better qualified candidates.

When they passed race blind laws in the UC system, UC Berkeley saw its black incoming class drop from ~500 to ~250 but the real big drop was a 15% drop in white admissions from ~3000 to ~2500. I can make a reasonably good argument for why we should try to boost black admissions to top schools, I don’t know what the argument is for a system that boosts white admissions to these schools. Asians were the primary beneficiaries of race blind admissions.

Almost 20 years later, the black population has recovered in the UC system as a whole but their numbers are Berkeley and UCLA are still lower than before prop 209.

Some Berkeley admissions officers were shocked at the results. They didn’t realize they even had this bias against Asians in the admissions process.

Mayor DeBlasio (a very liberal mayor of NYC) recently tried to inject holistic criteria into the science high school admissions exam thinking it would improve underrepresented minority admissions to these science high schools. He basically ended up making the argument that a bunch of old rich white guys were engineering the test for the benefit of poor immigrant Asians and got laughed out of the room. Holistic criteria benefits people who can pay for mission trips to Mozambique, art study programs in Paris, etc. If you want race based affirmative action then just have race based affirmative action. If you want socioeconomic based affirmative action then just have it. Don’t try to camoflauge it as holistic admissions because in the end, people with money will game those holistic criteria and you will end up giving advantages to rich white kids.

I see your point, but I’m okay with one of the holistic goals being “building a racially and socio-economically diverse class for everyone to experience”. My point is that it’s faulty thinking that a higher SAT score = more highly qualified for college across the whole range. There’s an idea that if we just looked at test scores and maybe grades, we’d be taking the kids that “deserve” it because they are the ones most likely to be successful, and that other criteria are unfair to those kids. But it’s not that simple.

What is the gap now?

For race-based AA, absolute score differences is not the way to approach it.

Better schools get a much better cross section of candidates with high scores, for example, minimizing the score differential across self-identified races.

A better approach is to set aside a target percentage for (self-identified) race-based categories and make sure you fill those minimums regardless of the average score differentials between races.

The targets should drive some sort of reasonable approximation of those races in the general community, and especially so with black students, who struggle the most to have admission scores that are competitive with their socioeconomic peers.

So, for example, if you set aside 8% of your admission slots for black matriculants, then you take your top black applicants, period, until the target is reached. Their evaluation is independent of any other group. The black applicants filling the 8% compete only against other black students for admission. Once the minimum quota is filled, additional black students can be admitted if they are competitive against the entire field of applicants.

Well, shit. Nothing is THAT simple. But if you are going to replace a system that uses largely objective criteria with a system that uses largely subjective criteria because the objective criteria are imperfect, then I think you have to show that your subjective criteria are more perfect and not just different.

I’m pretty sure that the sort of quota system you describe is unconstitutional. Depending on where the quotas are set, I would be OK with a quota for blacks and American Indians and perhaps a softer quota for socioeconomic status but its unconstitutional.

It depends on the school. More competitive school have larger gaps.

What are your standards for “largely objective”, because that is certainly debatable. In fact, that is the crux of the issue given those criteria are what most proponents of AA are arguing are not particularly, or demonstrably objectively indicative of merit. They just appear that way because they are quantitative measurements.

Further, if you really are of the mind that test scores and grades of THE most important thing, why even have an admissions department? Just establish a cutoff similar to how the National Merit Scholarship committee does, then accept the first x number of people whose scores and grades exceed those numbers. Why do you think Harvard and the like don’t do that? Do you think they are that committed to spending millions of dollars to advance a idea that has no merit or positive benefit, and to accept people who don’t “deserve” to be there?

How are test scores NOT objective measurements? Aren’t you falling into the same pitfall as NYC mayor DeBlasio? Are these old white test makers writing tests to help out Asians?

Nobody thinks that test scores are the ONLY thing that matters, especially at the top end of the academic spectrum. But the disparity in test scores and grades is stark.

They are not always objective measurements of merit because there is not equal opportunity and preparation. It’s not some conspiracy against Asian people, it’s largely the reality of the education system in this country. I am not saying the test is biased per se, but rather that they are not really good indicators of the qualitative differences between students.

Why don’t they believe that scores and grades and other quantitative measurements are the only things that matter if they want to switch to largely “objective” measures? If you are of that mindset, why would you care about anything else?

Why is that of any great consequence even if I agree that a given school is accepting students who are qualitatively worse?

They are good indicators of the qualititative differences between students. They say nothing about the source of those differences, but they are clearly demonstrations that differences exist.

Because you have denied opportunities to students who are better qualified to take advantage of them. You are getting less bang for the buck, in other words.

You are also, by systematically overmatching less-qualified students at more competitive universities, set the less-qualifed up to fail. And ensuring that the bottom students are going to be disproportionately AA beneficiaries.

That’s what’s called “diversity”.