Wednesday, February 22, 2012

Beyond Race in Affirmative Action

The NYTimes asked me to comment on Fisher v. Texas, the case involving affirmative action in higher education that the Supreme Court has agreed to hear.

Needless to say I anticipate being viciously attacked. I invite you to visit the Times comment section and engage my detractors ;-)

Rely on Merit, Not Race

In considering Fisher v. University of Texas, let’s acknowledge a key factual point about affirmative action: We have good tools for predicting college success, and those tools work about equally well across all ethnic groups and even for rich legacy candidates.

... Race-based preference produces a population of students whose average intellectual strength varies strongly according to race. Surely this is opposite to the meritocratic ideal and highly corrosive to the atmosphere on campus. Furthermore, the evidence is strong that students of weaker ability who are admitted via preference do not close the gap during college. For these reasons, the Supreme Court would be wise to end the practice of race-based preference in college admissions.


RKU1 said...

I'm a little surprised the NYT didn't bother including Gregory Rodriguez, since for the last dozen years or more he's probably been America's finest journalist/public intellectual on race/ethnicity/immigration type issues.  His absence also means they didn't include a "representative" of America's largest non-white racial minority group, which really seems a bit odd.

I also noticed another huge "ethnic absence" as well, but that one didn't surprise me in the slightest...

Guy_Brodude said...

But, aren't the SATs culturally biased? I mean, they once had a question about [i]regattas[/i], for heaven's sake!

A good article, but unfortunately I don't think it will change any minds. This whole issue is such an emotional powderkeg, and the innumeracy of the general public (as well as many "intellectuals") is a factor as well: the knee-jerk response will be to name some prominent black person of high intelligence and ask you to "refute this."

Robert Buttons said...

 [on campus] meritocratic ideal .............

Robert Sykes said...

This ought to be a simple matter. The 14th Amendment, Article 1, states "No State shall ...deny to any person within its jurisdiction equal protection of the laws." This is the source of the separation and church and state. It essentially inverted the meaning of the establishment clause (1st Amendment) by applying it to the states. It is also the source of the requirement that state (but not private) colleges and universities respect the freedom of speech of their faculty, staff and students. So, it seems obvious that affirmative action, which creates privileged classes, must be unconstitutional.

Bobdisqus said...

You are right on the policy, and brave to put your neck on the block.
I do have one quibble about the language though. Can we please substitute demonstrated ability for merit.  Merit has a connotation of virtue or earned reward from personal effort.  To the extent that IQ is a result of genetic factors outside of the control of the individual (the assumption of your BGI work) I think merit is not the right word to use.  We all have a duty to make the best use of the abilities we are given but we don’t all start with the same blank slate.  To get to the right tail of that bell curve requires luck in the genetic lottery as well as those individual efforts associated with merit.  The individual with a 1275 SAT may have far surpassed the individual with a 1600 SAT on all of those qualities we judge as virtuous but just did not have the advantages awarded in the genetic lottery.
Definition of MERIT
a obsolete : reward or punishment due
b : the qualities or actions that constitute the basis of one's deserts
c : a praiseworthy quality : virtue
d : character or conduct deserving reward, honor, or esteem; also : achievement
: spiritual credit held to be earned by performance of righteous acts and to ensure future benefits
a plural : the substance of a legal case apart from matters of jurisdiction, procedure, or form
b : individual significance or justification

Soren Kay said...

Are there any hard statistics out there looking at how academic mismatch changed when a university was forced to go without affirmative action?

See this blog post.

As U.S. Supreme Court reconsiders affirmative action, Michigan offers example of impact of race-blind college admissions

steve hsu said...

IIRC, after Prop 209 the number of underrepresented minorities (URMs) at UC decreased substantially, but their graduation rates went up significantly.

But you don't need such dramatic evidence. The Duke study (and all other similar studies) find that higher SAT score predict better grades or class rank (and the opposite with lower score). That means URMs admitted with higher scores typically do better at college.

All you need to know about mismatch is that if you regress class rank or GPA on SAT you don't need an additional ethnicity or race variable to improve the fit. The predictors work about equally well for all groups.

Richard Seiter said...

Steve, even if the analysis validates your POV I think finding statistics like Soren mentions could be of benefit.  I believe many of the people who disagree with you would be more responsive to concrete examples than they are to an abstract analysis (which some may not even understand).

steve hsu said...

Good point. I suppose it's most effective just to point out that URM graduation rates went up after UC went to race blind admissions.

Soren Kay said...

I don't need dramatic evidence(I'm 100% on your side), but I think those talking about academic mismatch need to do a better job quantifying the variance in "intellectual ability" between different ethnic groups on various campuses. Campuses need to judged(and ranked) against one another by how cross-racially equitable they are with regard to "intellectual ability". It would be nice if policies(university, government, court ordered, whatever) were imposed to keep the variance in check.

tractal said...

Yes, arguments about merit really aren't quite right if they are put in moral terms. Merit policies don't make sense because they are moral, they make sense because they promote efficiency and other desirable things. To me this suggests the whole issue is somewhat more complicated than Steve wants to make it. Meritocratic admissions policies tend to produce certain kinds of goods, but non meritocratic policies may produce different kinds of good. Policies which enhance URM life prospects at the expense of efficiency may make sense depending on how you weight those two goods. Is American society really better when doctors are slightly more competent but almost none of them are black? 

Appealing to general normative rules artificially makes complex issues seem simple. Moreover, we don't even know what is really at stake in terms of efficiency. Smart people are still going on to college, only some of them go to lesser institutions. According to the kind of strong IQ theories which we all hold, that doesn't necessarily make too much difference in terms of ultimate value creation. In any case, the harm caused by anti meritocratic policies depends on context. Its probably not too severe in Undergraduate admissions but probably unacceptably high in something like combat officer selection. At some point, anti-meritocratic policies will cause high inefficiency because the second best candidate simply can't do the job, as in math professorships for instance. Fortunately, we tend to see AA anti-meritocracy drop off in such IQ threshold fields. 

Richard Seiter said...

Are the UC URM graduation rates you reference absolute numbers or percentage of URM admittees?  If the former, that argument is a slam dunk.  If the latter, I think you would get some push back (although it's still a pretty good argument IMHO).  I think it would be beneficial to redirect the conversation into maximizing the success URMs have in the education system as a whole (back to your point about maybe a less elite college would be "better", with an emphasis on later life outcomes) rather than maximizing admitted URMs at any given university.  Also addressing the idea that classroom diversity has benefits for non-URMs as well (including the idea that diversity is about far more than race as has been discussed here, and can be constructive and/or destructive).

What kind of followup process do college admissions committees do to refine their models?  Are students who fail to graduate consulted about what might have contributed to (or could have prevented) that?  Or their feelings about whether a less rigorous program would have been a good thing for them?

steve hsu said...

Unfortunately I was under time pressure when preparing the op-ed, otherwise I could have argued from this perspective. Apparently, it's a slam dunk:

"4) Graduation trends. URM graduation rates have improved sharply since Prop 209 went into effect. For the six cohorts of black freshmen who started at UC campuses before Prop 209 went into effect (the matriculating years of 1992 through 1997), the average 4-year graduation rate was only 22.2%. For the years since 1998 (matriculating years 1998 through 2005), the black 4-year graduation rate across the UC system is 39.4% -- a near doubling. For Hispanics the 4-year graduation numbers are 27.2% for 1992-97, and 41.8% for 1998-2005. Six year graduation rates have risen as well, though less dramatically. Combined with the matriculation trends described above, the number of black and Hispanics graduating from the UC system has been rising steadily and remarkably. For example, the number of black students who matriculated at UC campuses in 2005 and graduated in 2009 was over two-and-a-half times higher than the number of blacks who earned 4-year degrees annually in the early 1990s."

Gabriel_Betteredge said...

Steve, the thing is, you state that the meritocracy is the ideal...but you never provide any reasons why this is so. From a justice standpoint, the people you suggest merit admission into the best schools in most cases don't deserve that position: ability is a function of genes, environment, and work ethic/ambition. The first two are outside of an individual's control. Arguably work ethic/ambition is also largely genetic and environmental. So from a justice standpoint the people who are poised to succeed most are often no more deserving of admission than those with lousy qualifications. Of course, if you don't care about justice and simply want to produce the highest number of smart, talented people who can contribute to the economy then you're right on the money. 

botti said...

Some comments seem to suggest that SAT is just a reflection of SES, so I've tried to post some comments in response referring to the research below. Hopefully they get published :)

steve hsu said...

Maybe you can post the URM graduation rates since prop 209 (see my other comment). I think phrasing the argument in terms of SAT as predictor of college performance is too confusing for most readers -- just directly stating the mismatch hypothesis (URMs get better outcomes when matched in ability to the rest of the campus) and supporting data would be better.

Richard Seiter said...

Wow.  Thanks for the link.  It seems like those results should change the conversation a bit ;-)

steve hsu said...

Don't hold your breath.

botti said...

Ok, will do.

Robert Buttons said...

SES is irrelevant wrt the validity of the SAT. Some SAT critics just cannot comprehend the concept of validity.

 But here is evidence showing other factors in play besides SES. B

RKU1 said...

Well, since no one followed up on my remark (which was actually half-facetious), I suppose I must...

The obvious reason that Gregory wasn't including in the discussion mix was that he's been a strong public critic of racial AA programs in the past, and presumably the NYT people felt that such criticism, especially coming from the "representative" of America's largest non-white racial minority group, might tend to "confuse" people.

So, why didn't they just pick some other Hispanic writer instead?  Well it would have been just like the NBA putting an Asian on the All-Star team, but not selecting Jeremy Lin---they would have become a total laughingstock...

Guy_Brodude said...

Not to heap too much scorn on my home country...but where else on this Earth except America would rhetoric to the effect of "people only get high scores on standardized tests because they study more" be considered a valid argument against a meritocratic model?

As if all (or even most) URM applicants are underprivileged. As if prep courses actually result in huge gains for the average student. As if Prof. Hsu didn't post data showing the predictive power of the SAT in the article! As if, as if, as if...

Miley Cyrax said...

Any discussion on affirmative action that neglects the Espenshade and Chung 2005 paper is incomplete, the one that observed that Asians were at the equivalent of an automatic 50 point disadvantage on the SATs vs. whites in college admissions, and 280 points vs. blacks. Also, in the absence of affirmative action, 4/5 of spots occupied by African Americans at top schools would have been supplanted by Asian Americans.
With this in mind, affirmative action is woefully unfair to Asians in favor of blacks. 

botti said...

Is Ian Haney-Lopez a Hispanic writer? He is coming out with some remarkable bullsh1t in his comments thread (Roger Clegg actually pops up there to debate him too):

"Discrimination has two meanings: differentiation and mistreatment. Affirmative action certainly involves differentiation on the basis of race. But it hardly constitutes mistreatment, at least of the sort that has traditionally merited constitutional protection.
In any democracy, there are people benefited and harmed by state action; indeed, we're all typically both benefited and harmed, sometimes by the same state act. This harm does not constitute mistreatment of the sort typically regarded as discriminatory.
Rather, the sort of mistreatment that has traditionally warranted constitutional protection has been the oppression of historically subordinated, socially and politically ostracized groups: think racial and religious minorities as well as women, and more recently, homosexuality.Frankly, the idea that affirmative action constitutes "discrimination" against whites in this sense of oppression and ostracism is laughably absurd."

RKU1 said...

Never heard of him.  Presumably, he's some sort of (half-?)Hispanic-activist type.  Looks like he got an endowment from that black guy on Wall Street who's being sued by the Lousiana pension fund for defrauding them of hundreds of millions of dollars I think.

Frankly, I really can't imagine suffering through a comment-thread debate about racial AA.  I've been closely following the issue for something like thirty-five years, and I don't think I've seen more than two or three original points made during that entire period.

BTW, that's really funny that they DID add a Hispanic guy after all.  When I'd previously checked the NYT page several times earlier, he hadn't been there, hence my original comment.  Guess they found someone, thereby rendering portions of my own remarks "inoperative"...

botti said...

***Frankly, I really can't imagine suffering through a comment-thread debate about racial AA.  I've been closely following the issue for something like thirty-five years, and I don't think I've seen more than two or three original points made during that entire period.***

I agree, it is a painful exercise. I'm impressed Haney-Lopez has ventured into his own comments thread.

I'm trying to imagine what the comments would be like if they invited Richard Lynn or Arthur Jensen to provide their take on the subject (Lynn actually had an article published in the UK Daily Mail a couple of years ago explaining a relative lack of women in science).

RKU1 said...

Can't really see what difference Lynn or Jensen or Francis Galton himself would make.  Anyone who has two brain cells knows perfectly well that racial AA admits large numbers of less academically-able (i.e. "smart") students.  That's the whole *reason* for racial AA.  And anyone honest would admit that.

Obviously, the requirement of brain cells and honesty probably excludes most of the participants in the NYT debate, so I'd guess it's more like professional wrestling, or maybe a debate between rival PR hirelings.

Taking a closer look at the page, it seems like everyone except Steve is a law professor, so presumably the legal issues are the central point of dispute.  But that's just as silly, since racial AA was *obviously* unconstitutional when it was first implemented forty-five or so years ago, and is obviously still unconstitutional today.

Or maybe not.  After all, these days the president claims he has the right to summarily execute anyone, American citizen or not, anywhere in the world whom he decides is a "bad person."  Compared to that, what's really so unconstitutional about justifying racial AA on the grounds of the emanations of the penumbras of the special diversity implications of the Bill of Rights.

Presumably, lots of different political blocs and factions are lobbying the Supremes and their law clerks right now, and maybe AA will be trimmed back a little or something.  But I tend to doubt that anything really dramatic will happen.

dwbudd said...

Gabriel, the problem with your argument is that "justice" is fraught with peril.  Often, political types who cannot make a logical, objective case for their argument reach for the reductionist hammer of "justice" or "fairness."  Right now, our President is campaigning on a record that no one can honestly point to as being terrifically successful, and on proposals that are as likely to harm the economy as to improve it, so he talks about economic "fairness."

What is "fairness?"  "Justice?"  Is it "justice" that two applicants to Stanford who grew up next door to each other in Alamo, Calif, who attended the same schools, and whose parents work in the same industry in the same capacity, but one happens to be black or Latino, and the other Asian should be treated differently?  That one, with superior class rank and SAT scores would be passed over in favour of the other because of some vague idea that Stanford wants to increase "diversity?"

The problem with arguing about what is "justice" is that everybody gets to decide for himself what he thinks is just and what is fair, and thus the discussion becomes irreducible.

Yan Shen said...

 Is there an analogous situation in New Zealand when it comes to Asians and academic discrimination?

botti said...

I'm at work so can't view those youtube clips. Not as far as I'm aware for asians who are NZ citizens. I know that there are higher entry criteria for international students (who generally come from Asia). Otherwise admissions are simply on the basis of grades (with quotas for Maori, Pacific Island, and some mature applicants).

Off the top of my head the only place that uses interviews to screen applicants is Auckland med school. I know a few people who had excellent grades but didn't get in following the interview. They were all european. One had a rather un-pc view of the relevance of the Treaty of Waitangi to medical practice.

Plenck said...

Most recent New Zealand PISA scores.The order is science, reading, maths.White: 555, 541, 537Asian: 530, 522, 529

Yan Shen said...

 Interesting. One possible explanation for the over-representation of Asians in New Zealand at the elite levels of academics (see the first YouTube video linked to above) and the low average Asian scores is that the Asian category lumps together East Asians with other racial groups. Does the Asian distribution have a higher SD than the white distribution?

Plenck said...

I am almost certain Asian means south, east, or south-east Asian. It certainly doesn’t include Pacific Islanders as it does in the US. 

Here is a 2006 Asian ethnicity breakdown:

Chinese: 147,570
Indian (really all Indian sub-continentals): 104,583
Korean: 30,792
Filipino: 16,938
Japanese: 11,907
Sri Lankan: 8,313
Combodian: 6,915
Thai: 6,057
Other: 25,179

Who knows how well this correlates with 15 year olds in 2009 (with which the PISA results are concerned)

The NZ Asian results may be low relative to NZ whites but they compare well with Japan’s results: 539, 520, 529. 

I couldn’t easily find information on the SD I might try again later if I have time.

Matthew Carnegie said...

This is not the 2009 data, but the Asians have higher variation in the 2006 data, slightly more so in terms of overrepresentation at the low end.

That may be from immigration effects undersampling the middle range of the parent Asian populations though (and/or oversampling the high end), rather than from ethnic structure (not sure how plausible this is based on NZ's immigration policy and demographic characteristics of immigrants, but you would need very slight effects I think).

Blog Archive