IT IS NOT JUST AT BATES. Dammit - a person at Bates looked at 33 different schools.
and your own link shows this quote in the opening abstract
So the SAT does not necessarily correlate to college success.
Again, see figure 25 in the study: http://api.viglink.com/api/click?format=go&jsonp=vglnk_14320957607598&key=ce270d38103a630c3181db372a25a052&libId=i9w8q2d2010004p6000DL1c62941v&loc=http%3A%2F%2Fboards.straightdope.com%2Fsdmb%2Fshowthread.php%3Fp%3D18377991%26posted%3D1%23post18377991&v=1&out=http%3A%2F%2Fwww.nacacnet.org%2Fresearch%2Fresearch-data%2Fnacac-research%2FDocuments%2FDefiningPromise.pdf&ref=http%3A%2F%2Fboards.straightdope.com%2Fsdmb%2Fnewreply.php%3Fdo%3Dnewreply%26p%3D18377982&title=Asian%20American%20groups%20accuse%20Harvard%20of%20racial%20bias%20in%20admissions%20-%20Page%202%20-%20Straight%20Dope%20Message%20Board&txt=http%3A%2F%2Fwww.nacacnet.org%2Fresearch%2Fres...ingPromise.pdf
All students SAT scores: 1135
Submitters’ SAT scores: 1130
Non-Submitters’ SAT scores: 1138
Do you find that difference significant?
Note from the same table: when you compare two groups with significant difference in SATs: 1137 vs 895, the graduation rate drops from 67% to 56%. And that is for ALL colleges in the study, we don’t know what they are. They could be dinky colleges where graduating is not that big a deal.
Now - show me an elite (you know what, top 50 will do) school and give me a graph that correlates SAT ranges with graduation rates. As I said, if such a graph is flat, that finding would actually be non-trivial.
But what the Bates experience does show is that you could just as well at predicting success in a liberal arts college without looking at a standardized test result as knowing it … choosing based on GPA and the “holistic” factors does as well.
Of course that follows from what you observed: if you choose based on GPA and those ephemeral other factors then you will be choosing for those who would score similarly on the SAT as well … on average anyway. The SAT therefore adds little to the decision process. And the bigger study further shows that the SAT as a stand alone measure is less predictive of college performance than HS GPA is.
In terms of why I would suspect the SAT would show a difference at Bates between submitters and non-submitters is due to one difference they did find - future standardized test scores (say for application to med school) apparently differed between the groups.
Truth be told I am still modestly skeptical … the groups generally went into different sorts of fields of study and the SAT may still be more predictive of success in maths and hard science performance. But again, a liberal arts college is not only students going into math and the hard sciences.
Taking standardized tests well is a particular academic skill and not an unimportant one. But it is not the definition of an overall higher quality student nor required to identify one even if that skill, like being able to shoot free throws well, is one useful skill for a student/player to possess.
As far as the graph you want Terr easy to create: the range of SATs among those getting into Harvard, just in the class of 2017, goes down to 1660 and up to quite a few with 2400. Essentially no one fails out of Harvard … a majority graduate with honors … the graph is pretty perfectly flat.
I don’t see the graph. If the upper is 99% and the lower is 85%, (with it averaging to 90%) that would definitely not be flat.
Also, when dealing with elite schools (and, I think, medical schools), graduation rates are less of an indicator. My wife is an MD, and she told me when we were talking about her school that it was basically almost impossible to flunk out. The administration almost forced you to stay in. The graduating class rank probably would be a much better measure.
Move goalposts much?
Oops.
Y’know I’ll sit this fight out. Knock yourself out.
I acknowledged that I “moved the goalpost”. But even with the old goalpost, you couldn’t provide the flat graph.
You’re both saying the same thing: a disproportionate number of blacks admitted to Harvard and the other elite schools are from wealthier families.
But why should this be surprising? Is it not the case that the lower-income demographic on average scores lower on SATs and tends to have lower GPAs and fewer advanced courses and, no doubt, fewer opportunities to achieve the kinds of personal benchmarks that the schools look for? Blacks and Hispanics may get a boost from AA but the admissions process is still highly competitive, including among the minorities themselves. And, in the end, they do have to be successful once they are admitted, not to mention successful once they graduate.
I’m not saying this isn’t a problem. It is. It just isn’t Harvard’s problem, and it’s not fixable with yet more tweaking of admissions criteria. It’s much more basic than that, and needs to be approached on a much broader basis and much earlier in the educational system.
Just as in the alleged but non-existent anti-Asian “bias”, why does everything have to be blamed on some nefarious conspiracy of class-warring rich white supremacists? The solution to these inequities isn’t to keep hammering away at admissions criteria to admit more of the unqualified, it’s to help more of the disadvantaged actually become qualified.
Cambridge is not exactly a good example for merit-based applications. In the UK, you take exams for A levels when you’re about 18. Candidates for the Russell Group universities often have four A-levels at A or A* grade. Beyond that, there is no ranking; no way to tell if someone got 450 in their SATS or whatever because they don’t exist at that age - if it’s an A grade, it’s an A grade.
Those exams are compiled and tested by national examining, btw - it’s not like a school says “he’s an A student.”
Oxbridge still tends to admit fewer students from comprehensives who have excellent grades than students from private schools who have the same grades or less. (Link: Insights - Complete University Guide) To their credit, they are trying - a sixth-form near here has a wonderful Oxford Uni connection that provides tutoring and a quiet study room, though this is paid for by the Local Education Authority, with all its leather-lined chairs and Oxford branding.
Comprehensive schools are the norm in the UK. I can’t, via Googling, find a really good source for this but someone else maybe can. I have known people from comps who went to Oxbridge colleges but they did notice that they were a minority.
So the fact that Oxbridge tends to underadmit students from comps with identical grades ad students from private schools does not, actually, make them wonderful and something for everyone to emulate.
Some general comments from the perspective of one who has sat on an admissions committee (med school; couple decades ago;) for five years.
-
Standardized scores are a necessary, but not sufficient, screen to get students capable of high performance in STEM fields. A hard-working student with good grades from a mediocre high school/college and with lower standardized scores will struggle in STEM fields much more than a student with mediocre grades and stellar standardized scores.
-
The vicissitudes of the workplace, and the qualities required beyond standardized scores mean standardized scores are much less useful for success prediction in non-STEM fields.
-
Preferential hiring (employment AA) obscures success rates. If asians who are highly qualified get a job in which they succeed, that success is likely related directly to ability. If a preferentially-admitted student gets a job in which they succeed, but the job is attained through preferential hiring, you cannot use “success rates” to say the standardized scores (or grades) didn’t mean anything. The reason they didn’t mean anything is that they were discounted.
-
There is currently no mechanism by which to achieve race-based balance other than pure race-alone based preferential consideration. Correction for socioeconomic privilege does not create a balance; at every socioeconomic decile asians will be highly over-represented within the highest academic performance groups for that socioeconomic decile.
-
The idea that “other considerations” are what drive preferential admission for those who are admitted with lower standardized scores is pure bullshit. It’s a public-facing masquerade on the part of universities behind which to hide. Asian students are no less likely to have the softer criteria under consideration save one: their (self-assigned) race.
-
100% of public and private institutions of higher education use race-alone considerations, and particularly so at elite institutions. Because elite institutions have such a plethora of applicants with the triple combination of high scores/grades/other, only race-alone preferences will keep a student body tuned for race-based diversity.
-
When institutions talk about “diversity,” they mean one thing and one thing only: diversity of race. The rest of the diversity talk is window dressing. Elite institutions can, and do, preferentially admit highly privileged black students with (relatively) low scores over poorly privileged asian students with high scores. There is simply no other mechanism by which to maintain a race-based diversity. Harvard (and the rest of us) will matriculate a wealthy black immigrant with no US enslavement history over an asian from a poor background even if the asian has better scores/grades/other. The asian competes for admissions against his fellow asians; not against blacks.
-
Oversight legislation and a few court cases have attempted to intervene in the admissions process, with various degrees of success. And of course “success” depends on which side of the AA boundary line you stand on. However the admissions process is generally complex enough, and soft enough, to dodge most efforts promoting straightforward meritocracy that simply combines prior academic performance along with socioeconomic opportunity and “other” as the winnowing factors. It is not that difficult to hide a race-alone preference for any given student, and only when large patterns are analyzed does it become obvious that asians are at a marked disadvantage in the sense that, for equivalent scores/grades/other, they are much less likely to be admitted.
-
For the forseeable future, race preferences will need to continue for both educational institutions and any jobs for which screening examinations are administered for either hiring or promotion. Without a deliberate race-based double standard, the workplace will become as disproportionately represented by race as is the NBA.
The idea of diminishing returns comes to mind with all admissions criteria, but especially the SAT. I think we can all agree that a student who scores in the 60th percentile on the SAT is less prepared for college than one who scores in the 90th percentile.
But if we’re comparing a student from the 90th percentile from a 99th percentile, why would we expect the former to be a worse student than the latter? You don’t get extra points for turning in an exam early. You can’t do better than a A+ on an essay. Being able to juggle 19 credits doesn’t make you a more valuable student than one who can only juggle 16 credits at a time. Scoring astronomically high on the SAT does not mean you will be more creative, innovative, or tenacious.
To use a “holistic” analogy, it’s like saying that a student who founded an African orphanage, holds a patent for a Smart Phone app, and is the concertmaster of her city’s youth orchestra is more qualified than a student who started a pitbull rescue group, published an online cookbook on Appalachian-Asian fusion cuisine, and gives poor rural kids free piano lessons. Obviously both of them are interesting students. If you’ve only got one seat left at your institution, you’d be better off flipping a coin between them rather than pretending you can pick the “better” kid.
I have a bit of an issue with this one point. We know that African-American applicants to the elite colleges have significantly lower SAT scores than the average, and that the preferential treatment is driven by societal mandates that most of us believe are worthwhile. I don’t think there’s any “masquerade” about it, or any claim that these candidates have some abstract meritorious attributes that account for it. And it would be silly, just on the basis of simple math, to claim that this doesn’t disadvantage others, including whites and Asians. Just like preferential treatment of legacy applicants and star athletes impacts the whole applicant pool. Some of it is race-based, lots of it may be unfair, but none of it is the question here. The question is whether there’s overt racism directed against Asians.
And here the story is entirely different. Asian applicants have marginally higher SATs than the average of the applicant pool (from the study that was linked earlier) and all things considered (from the same study) they are disadvantaged by the equivalent of about 50 SAT points (out of 1600). Much of that is accounted for by the preferential categories just mentioned, which makes everyone else, by definition, also take a hit. Is it such an incredible stretch to imagine that the rest is accounted for by the kinds of soft criteria that the Ivy Leagues use to assess, say, a person’s social abilities or other abstract qualities that correlate with success?
Remember that Asians are over-represented in the elite admissions stats. They just aren’t over-represented quite as much as their SATs alone would indicate. Should they be, when this is just one number among innumerable quantitative and qualitative criteria? The DoE Office of Civil Rights certainly didn’t think so in 1990. Why would it be different today?
The premise behind this is that the correct level of “representation” is entirely based on the racial makeup of the country, which is a false belief.
Harvard agrees with you – there are too many Asians and there must be racism targeted at them in order to keep Harvard white. Those who do not think in terms of inherent “right” numbers of particular races disagree with you, and Harvard.
That’s not the premise at all. “Over-represented” is a statistical concept, not a value judgment. It just means that there are more Asians in the cohort than there would be if the selection were purely random. It tells us that the deviation from random is strongly in favor of Asians, which makes it rather stunning that they’re complaining about being discriminated against. The false premise is assuming that the correct level of representation must exactly match some arbitrary synthetic number like an SAT score.
No one expects or desires the selection to be “random.”
AA benefits Asians by reducing the amount of white students in schools and establishing a diverse quota of racial enrollment. What part of that are you not getting? Do you honestly think AA only helps blacks? 
Again, you’re not getting it. You asked in post #247: “At what point in time do you believe there was affirmative action in favor of Asians at Harvard?”
My response was: “When such quotas were created, reducing the dominance of whites in these schools and creating a diverse student body”. I haven’t forgotten anything but it seems like you realized you couldn’t win the argument and went off on a tangent. To spell it out for you and Velocity: AA benefits non-whites when it was created because whites were overrepresented in schools. Reducing the number of whites helps every non-white person.
Now you want to say that I’m arguing to give less qualified whites over Asians? No, sorry, you don’t get to conflate 2 different arguments. You asked a question about how AA benefits Asians and I answered, the end. If you want to discuss how things are now with AA and its impact, ask it separately and don’t try to tie it into a question that’s been asked and answered
And I’ve answered the topic. I’m fine with AA, being a tool to enforce racial diversity, being used to reduce Asian enrollment in order to increase underrepresented races, in this case whites.
AA doesn’t just help blacks. You don’t seem to understand that. Until you do, you’ll continue to ask me pointless questions. AA in this case is helping whites, which is fine with me. Its hilarious that you don’t know what AA does yet argue about it like it affects you personally
Some might have, but not enough as represented by the proportion of their population. You do realize that at one point, pretty much all top schools were like 99% white males right? You do know this country’s racial history don’t you? In that case, of course AA’s going to help Asians. And blacks. And women. And anyone else who’s not a white male.
And now we have a lot of Asians and they want to get rid of it because it no longer helps them specifically. That’s why I called that selfish. If 5% more Asians can’t get into Harvard because that 5% of spots is reserved for non-Asians, I’m totally fine with that. Completely.
So you’re doubling down on “racism is good” and refusing to answer the question of when there was affirmative action for Asians at Harvard?
You’re using an affirming-the-consequent fallacy. The mere fact that a racial minority increases as a percentage of university enrollment by no means proves that AA was the cause. There could be many possible causes.
Opening admissions to people on the basis of merit is, of course, in and of itself going to knock down a “99% white male” student proportion, simply because many minority students have merit. But AA is another matter entirely. AA admits people due to race. I challenge you to provide proof of significant numbers of Asians being granted admission in the 1900s on the basis of race, *not merit.
*
Finally, even before WWII, there were Japanese students attending universities in America.
You seem to be ignoring that up until the 1960s when immigration restrictions on Asians were restrictions, the number of Asian-Americans was quite miniscule. In the 1940s and 1950s Chinese and Japanese Americans were such a minuscule percentage of the population I’m not sure it’s clear they were underrepresented.
As for say Vietnamese-Americans and Korean-Americans there were virtually none living in the US prior to AA(which was implemented in the 1960s) so it’s quite silly to suggest that prior to AA they were denied access to elite schools.
Beyond that, I’m not sure how many Vietnamese and Korean immigrants would appreciate being lumped in with people who are of Chinese and especially those of Japanese descent.
I’ve answered that, more than once. Are you simply repeating your question because you know your reasoning has no leg to stand on?
If you’re using that kind of vagueness to argue, why not just say AA does nothing because we can’t be sure if racial increases/decreases are the result of outside factors? YOU want to say AA causes harm to these Asians by limiting their opportunities, and now you want to say AA didn’t help to increase them in the first place? Try to have it both ways and you’ll fail.
Pick one: AA doesn’t help Asians because it doesn’t affect racial quotas or AA does affect racial quotas and helped Asians get into these schools
You seem to want to confuse the argument. Are you talking about the early 1900’s before significant minority participation in higher learning or late 1900’s after AA’s been around for a while. But I suspect you know that. I’ll answer your question in its entirety, including the things you’ve left out because of the biased questioning:
Allowing minorities the opportunities to test for these schools increases racial balance from majority white to plurality white. Yet there are still problems. Without AA, schools would change either too slowly or not at all, and would remain disproportionately white, way past when such things are seen as negatives (ie. decades after it stopped being cool to discriminate in these areas). Therefore, AA was created to further balance out minority enrollment. It admits people due to race because racism hadn’t gone away, because even with less restrictions on minorities, they were still being discriminated against. With AA, the systematic racism of enrollment discrimination could no longer stand because quotas were needed. Therefore, it further reduced racism and is a good thing.
There is still pervasive racism today against minorities. Therefore, AA is still necessary to lower the dominance of one racial group in schools and increase the diversity for other, less represented minorities.
By your line of questioning, it seems you want to just come out and say that the reason why there was less minorities in schools was because there was, physically, less minorities in the US, and the lifting of restrictions on immigration meant that the factors that held back Asian enrollment were lifted as well (and therefore they don’t need AA). I assure you that I didn’t ignore it, I just find it irrelevant.
More than Asians were minorities in the US. AA helps all minorities, not just blacks or Asians. Less Asians in schools due to immigration barriers does not mean there was simply no racism or no discrimination towards minorities. If that was true, there’d been more blacks in schools, but there wasn’t, not proportional to their population at least.
You can think that immigration restrictions were the main factor, and that might be true, but I’m not debating that. I’m debating whether AA is fair and should continue. I think it is. The controversy has always been that AA artificially lessens the numbers of some races while boosting others that maybe didn’t test so great. I don’t find that a problem at all. If only 1% of Harvard is black and 50% white, I want more blacks to be enrolled and less whites. Right now, there is a large percentage of Asians in Harvard and less so with blacks and latinos. Therefore, Harvard should be rejecting more Asians in favor of blacks and latinos. That it hurts Asians is not a problem for me that needs fixing, it is a feature that should continue