« "Hell on earth" and "jackbooted authoritarianism" | Main | Protesting sport »

Monday, April 28, 2014

Affirmative Action and Top Ten Percent Policies

With the Supreme Court’s blow to affirmative action last week, state universities may increasingly turn to the Texas model of automatic acceptance for applicants at the top of their high school class rank. When colleges draw from the top five or ten percent at all high schools, they may be able to recruit an entering class that mirrors the ethnic and racial diversity of high school graduates. While top class rank policies raise a number of concerns and their impact on diversity is mixed, there is a potentially more important benefit from a tweak of the policies.

Suppose that instead of looking just at GPA, admissions offices looked at a range of measures, including test scores, artistic talent, and athletic skills, and admitted the top students from each high school. Suppose further that all selective colleges—public and private—employed a top ten (or one) percent admissions policy. By whatever measures admissions offices used to rank applicants, the colleges would admit the top applicants from all high schools (of a minimum size). 

Parents would recognize that their children would do better in the application process by attending Urban High than by attending Suburban or Private High. Instead of concentrating their children in the highest performing schools, parents of means would have a strong incentive to spread their children across the full range of schools. 

There would be two important effects for cities. First, their property tax bases would grow, as more families chose to live in the cities than in the suburbs. Cities would be in a better position to invest in infrastructure and finance public services. Second, school quality would improve. Once their children were attending one of the lower performing schools, parents would push for improvements in the quality of the school, whether by seeking more public dollars or by raising more supplemental funding. The upper and middle socioeconomic classes might still focus their attention on the schools that their children attend, but the number of such schools would have increased. The gap in quality between the top schools and the bottom schools should narrow, and school quality should become more uniformly high. Rural areas also should benefit from top ten policies.

Would parents really send their children to lower performing schools to take advantage of high class rank admissions policies? They already have in Texas.

After that state’s ten percent policy was adopted, a number of parents moved their children to schools with lower levels of achievement by the student body. The effects would be even greater if Ivy League and other elite universities followed the Texas model. The ten percent policy also has had a substantial impact on property values as families have moved into neighborhoods with lower-performing schools. Because parents adjust their choices of schools in response to high class rank policies, academic standards at selective universities needn't suffer (except perhaps in the short-term).

We've known for a long time that we do better by the disadvantaged when we link their fortunes to the fortunes of the advantaged in society. Top class rank policies can provide that linkage. And they can do so without having to pass any laws. (For more discussion of this idea, see here).

[cross-posted at orentlicher.tumblr.com]

Posted by David Orentlicher on April 28, 2014 at 10:39 AM | Permalink


I'm sure you discuss this in the article, but my understanding was that the evidence is pretty mixed as to whether the Top Ten Percent plan is beneficial to minorities. From what I recall, there is some evidence that white students gaming the system (usually by going to the lower performing neighborhood school instead of a more competitive magnet school) were essentially crowding out high achieving minorities from those schools, and the people really bearing the cost of the policy were middle class minorities in more integrated schools. And that's not to mention that class sizes at the most prestigious Texas public schools (UT and A&M) have ballooned during that time -- a time in which the Texas economy has grown disproportionately rapidly. It's not clear that public institutions in states with a more representative growth rate could absorb that sort of increase.

Posted by: My $0.02 | Apr 28, 2014 11:36:52 AM

Very interesting. This does seem like a potentially clever way to improve ailing high schools without intrusive (eg busing), expensive, or otherwise politically unrealistic interventions. I wonder, however, whether the policy justification for favoring the top ten percent of each school would make sense anymore if it became clear that a significant number of the beneficiaries were "parents of means" who chose lower-ranked high schools in order to improve their children's chances of getting into good colleges.

Posted by: AF | Apr 28, 2014 11:38:34 AM

There are various comprehensive ways to deal with overall social problems that would be ideal but one reason why more limited race conscious measures were used was that there was not enough support for them.

The first comment suggests the specific concept -- which various people also argue is race conscious in certain respects (did not Ginsburg do it in one of her opinions? I think so) -- is a complicated matter. I welcome serious debate and attempts to try it out. I'm unsure about doing so only because an amendment like the one upheld last week was in place.

Still, race is itself a problem in this society. Just dealing with class or whatever will not solve it. Sometimes, race conscious (or sex conscious, or religion conscious etc.) means probably are necessary. Those, to speak generally, who say "I think we should use class based measures," e.g., to me are probably a bit naive if that is all they got.

Posted by: Joe | Apr 28, 2014 12:35:42 PM

I also wonder *when* middle class students attend underperforming schools. The rational thing to do is to send your child to an over performing school, and then (when things get tough at the top) transfer to an underperforming school for the last year or so (e.g., by renting in that school district). In that case, not only do minority and poor students get crowded out, but the economically disadvantaged students get no benefit from having middle class students work along side them, or contribute cash to the school over a long period of time, or all the other benefits of having middle class folks in school. In fact, the message is quite the opposite: it's "we're using your inability to perform to help our already advantaged kids get an additional advantage." I'm not saying this is exactly what happens; but it seems to me the economically rational thing to do.

Posted by: Eric Miller | Apr 28, 2014 12:47:32 PM

Eric Miller:
Here in NYC I have heard of some strategic renting but of the opposite variety -- a family will rent a tiny apartment in a desirable school district, enroll their youngest child, and then after a year leave for cheaper, more specious accommodations. Not only is the oldest child grandfathered in, but his younger siblings are too.

It's hard to see how the strategy as you outlined it would work in a family with multiple children. Although there's always the related scams involving a child nominally living with a relative.

Posted by: brad | Apr 28, 2014 3:45:53 PM

Brad, don't be unfair: "the economically rational thing to do"-line of thinking rarely intersects with real life.

Posted by: new prof | Apr 28, 2014 4:03:26 PM

I echo what others have written concerning "Ten Percent Policies." As many scholars have argued (including Justice Ginsburg), while the policies are racially neutral on their face, they are indeed racially conscious remedies.

What is frustrating about this post and other attempts to promote these policies is that they rest on an assumption that if we somehow make black and Latino schools better, then black and Latino students will become "qualified" and gain admission into selective institutions of higher learning.

This is partially true. But it is a dangerous hypothesis when overstated and not placed into a larger context of racism in university admissions. We need to stop operating on the assumption that racism does not exist in the current admissions process, and that the current process is race-neutral. It is not. University admissions rely on racially biased testing and grading. These racial inequalities keep a number of highly qualified minority students out of selective admissions. A ten-percent policy does nothing to address this real problem.

Until we deal with the systemic problem of racism and its deep roots at multiple levels of society--particularly in the ways in which we assess students--we will continue to chase our tail wondering why blacks and Latinos can't improve, and using race-sconcious (yet facially neutral) policies in order to reach some arbitrary end point for affirmative action.

Posted by: Sheldon Bernard Lyke | Apr 28, 2014 5:05:25 PM

There is a larger question of what exactly the merit that universities are supposed to be using as a criteria. Forget the bias in the test -- what is it they are trying to measure in the first place, and what role does that (whatever it is) play in admissions?

It is at least plausible, though I think somewhat dubious, to consider athletic prowess a relevant form of merit, but legacy? How can we even begin to talk about merit with a a straight face in a system that includes pervasive advantages to legacies?

If there are any tax scholars here, here's an interesting question:
If a taxpayer makes a contribution to a university with the expectation that said donation will lead to admission for his child, is that contribution a tax deductible donation or payment for value (like the cost of dinner at a charitable event)?

Posted by: brad | Apr 28, 2014 5:57:46 PM

Perhaps I'm backwards, but it seems to me that policies should be helping minorities and poorer performing children get to better schools, not mobilizing the better (usually more white) students to worse schools. Good students in bad schools will not make the teachers any better, and thus minorities stuck in those school districts will not be any better served with the mobilization invoked by this model.

Posted by: Andrew Kartchner | Apr 28, 2014 8:21:13 PM

@Sheldon - Yes. There is such an aggressive push on the Court and elsewhere to embrace this post-racial fantasy land, and I think it needs to be aggressively resisted. I think scholars can play an important role in describing and developing a vocabulary for contemporary racism, as well as other forms of subordination. The sense that we are at "the end of history" in terms of race needs to be actively and constantly refuted.

Also, my understanding is that Top 10% policies only "work" to produce racial diversity when you have rather pervasive residential segregation. This is related to but a bit separate from the points above. So it is a bit of a deal with the devil to rely on these policies to produce educational diversity.

Posted by: Susannah Pollvogt | Apr 29, 2014 11:42:31 AM

Something no one here has mentioned yet -- the bias against Asian-American students in higher ed admissions is so well-documented that Asian-American applicants are coached to hide their racial identity on their applications.

Seems that nobody on the right or the left wishes to implement an admissions system based on "pure merit" because those on the right are afraid Asian-American students will win around 75% of the available slots, like the mostly-Asian-American student body at Stuyvesant High School in NYC, where admission is granted only to the highest test scorers. Those on the left are afraid the other minorities will continue to be left out (see the NYPS admissions test lawsuit). I wonder how this has played out in the context of Top 10% policies.

Posted by: hush | Apr 30, 2014 4:49:42 PM


A number of scholars have discussed the admissions bias against Asian-Americans. In the 90's Claire Jean Kim discussed the Racial Triangulation of Asian-Americans, and Jerry Kang at UCLA Law has discussed how Asian-American's face "negative action" in admission. You state that those on the left are afraid the "other minorities will continue to be left out", but work by Bill Kidder shows that Asian-American's do not experience "negative action" because other racial minorities "take" Asian-American "slots". Whites are the beneficiaries when Asian-American numbers are held low, yet the narrative pits Asian-Americans, Blacks and Latinos in a zero-sum game "people of color fight club" for resources.

In addition, I am happy that you put "pure merit" in quotes. High test scores and high grades are not synonymous with merit in a world where racism can affect one's grades/scores.

Posted by: Sheldon Bernard Lyke | May 1, 2014 12:28:26 PM

On merit--the ideas of "merit" and "meritocracy" are wonderful, but they are just that--ideas. As an aspirational goal, we should strive to create conditions that maximize the expression of individual merit. But in taking account of current social reality, GPAs and standardized test scores cannot somehow be isolated from prior dynamics of social, economic and educational marginalization--they cannot be "purified" to exclude the effects of social stratification.

But this is not to say that I see hush as advocating a particular idea of merit--just a response to the concept, generally.

Also, there is a lot more going on in terms of success than simply gaining entrance to a university. For example, there is some evidence of racial selectivity in how faculty approach mentoring students:


Posted by: Susannah Pollvogt | May 1, 2014 3:10:58 PM

These comments are very helpful. Thank you all for your contributions. While top of the class policies do not provide a comprehensive solution to the problem of discrimination in our educational system, they can do much to level the playing field. The more we move toward a single-tiered system of education, the better off will be the disadvantaged, just as the disadvantaged are better off under a Medicare program that serves all seniors than under a health care system for non-seniors that has different programs for the poor than for the prosperous. It would be great if we could improve low performing schools for the sake of their current students, but we're far more likely to see improvement of the schools if we give them a stronger constituency.

On the gaming question, that's more of a concern with test scores, athletic ability, or artistic talent than with GPA. Students moving from higher-performing schools to lower-performing schools will take their grades with them. Hence, the more years they have in the lower-performing schools, the better off they are in terms of class rank.

Posted by: David Orentlicher | May 1, 2014 3:35:30 PM

Post a comment