Though I very much appreciated Northwestern University sociologist Carolyn Chen’s op-ed on anti-Asian discrimination in elite higher education, I found the following passage somewhat problematic:
Some educators, parents and students worry that if admissions are based purely on academic merit, selective universities will be dominated by whites and Asians and admit few blacks and Latinos, as a result of socioeconomic factors and an enduring test-score gap. We still need affirmative action for underrepresented groups, including blacks, Latinos, American Indians and Southeast Asian Americans and low-income students of all backgrounds.
But for white and Asian middle- and upper-income kids, the playing field should be equal. It is noteworthy that many high-achieving kids at selective public magnet schools are children of working-class immigrants, not well-educated professionals. Surnames like Kim, Singh and Wong should not trigger special scrutiny.
Recall the work of Caroline Hoxby and Chris Avery on the large population of high-achieving low-income students who do not apply to selective colleges and universities, which we discussed earlier this month. Hoxby and Avery find that 70 percent of these students are non-Hispanic whites. That is, if we sought to increase the representation of low-income high-achievers in elite higher education, we would disproportionately benefit non-Hispanic white students. In my view, this would be a fine outcome. Yet it raises the question of which preferences should take priority — preferences for middle- and upper-income kids from underrepresented racial groups, e.g., African Americans and Latinos, or preferences for low-income students of all backgrounds.
In Mismatch, Richard Sander and Stuart Taylor Jr. offer a coherent set of organizing principles: (1) Racial preferences should be transparent, i.e., beneficiaries should have a clear understanding of where they stand in terms of academic preparation relative to their peers. This transparency will serve as a consumer protection measure. Students benefiting from large preferences should know that they face a significant disadvantage that may well be reflected in their academic performance. (2) Racial preferences should be no larger than preferences for students from economically disadvantaged background, to emphasize the importance of having a socially balanced student body as well as ro discourage schools from making use of preferences that are so large as to raise the risks of academic mismatch. The end result of these reforms are difficult to predict: we might see Asian representation increase, but we might also see the representation of low-income non-Hispanic whites increase even more. Representation of black and Latino students might remain stable, yet the representation of low-income black and Latino students might increase relative to the number of middle- and upper-income black and Latino students.
Chen is implicitly suggesting that an increase in Asian representation can and should only come at the expense of middle- and upper-income non-Hispanic white students. But if we take the idea that the representation of low-income students of all backgrounds ought to increase, and that socioeconomic diversity should be at least as high a priority as ethnoracial diversity, the end result might look very different.