Asian Americans would not lose out under affirmative action

October 6, 2014


Last week, I reported about an updated survey jointly conducted by the National Asian American Survey (NAAS) and the Field Research Corporation that examined California voters’ attitudes towards affirmative action. That 2014 survey, led by Dr. Karthick Ramakrishnan, revealed that AAPI support for affirmative action policies have not shifted since 2012 (or the mid-nineties, under the auspices of a state referendum on affirmative action): 70% of our community’s registered voters still support affirmative action. These data corroborate similar findings from a 2001 survey conducted by a different group.

I wrote in my article that the findings of this latest 2014 study are likely to distress opponents of affirmative action. No surprise therefore that an op-ed appeared in the LA Times last week titled “Asian Americans would lose out under affirmative action“. The column is written by Yunlei Yang of the Silicon Valley Chinese Association and it is strongly critical of the 2014 NAAS survey results.

Yet, Yang’s column is also seriously flawed.

Taking Issue With Wording

Yang focuses the bulk of his criticism of the 2014 Field Research Poll on the wording of the question regarding affirmative action. He writes:

I find the poll question misleading and Ramakrishnan’s reasoning deeply flawed.

The original text of the poll question, written by a group Ramakrishnan directs, was, “Do you favor or oppose affirmative action programs designed to help blacks, women, and other minorities get better jobs and education?”

pew-attitudes-affirmative-actionThe question wording is lifted largely intact from a Pew Research Center poll on affirmative action that has been used as recently as 2007 (right). Pew is a nonpartisan survey group that conducts annual public opinion polls on a variety of subjects in America. In January of 2007, they found that 70% of survey respondents supported affirmative action, which they described as supporting “affirmative action programs designed to help blacks, women and other minorities get better jobs and education”. Your eyes do not deceive — this is exactly the same wording that Ramakrishnan used in his Field Research Poll.

As recently as two years ago, Pew surveyed attitudes towards affirmative action with the following question: “In general, do you think affirmative action programs designed to increase the number of black and minority students on college campuses are a good thing or a bad thing?” While this represents a deviation in question wording from 2007, the deviation is not really substantial; the question is more focused towards the college campus sphere (where the bulk of disagreement on affirmative action is centred) and is a little more colloquial in tone (“good thing or bad thing”), but otherwise keeps the spirit of the question intact: affirmative action programs are still defined as programs “designed” to “help” or “increase” representation for minorities.

Yang (and others) take issue with this definition of affirmative action programs as provided in Ramakrishnan’s survey question, yet this definition is not drawn from a biased liberal source; instead, this language is actually drawn the case law on affirmative action, itself. The goal of affirmative action programs arrives in the form of the 1964 Civil Rights Act, which reads:

In administering a program regarding which the recipient [of federal funding] has previously discriminated against persons on the ground of race, color, or national origin, the recipient must take affirmative action to overcome the effects of prior discrimination.

To draw upon a broad legal definition is essential; affirmative action programs are not limited just to race-conscious considerations in college admissions, but also to recruitment and retention programs.

It is only common sense, therefore, that if we wish to poll general attitudes on affirmative action programs, we must define what we mean by the phrase. Affirmative action, as defined by reams of case law and as articulated in contemporary layman’s terms, are simply programs designed to increase minority representation in the spheres of education and employment for a specific purpose — either to overcome discrimination (if one goes by the rationale of the 1964 Civil Rights Act) or to address the “compelling interest” of campus diversity (if one goes by the arguments presented in Grutter and subsequent Supreme Court decisions). Furthermore, to address the shift in rationale for affirmative action programs in the last twenty years, the 2012 NAAS survey polled respondents with multiple question wordings drawing on different aspects of affirmative action case law, and found no impact of question wording or presented rationale on survey responses. Thus, it becomes clear that this question wording of NAAS (and the more recent Field Research Poll) communicates a meaningful definition of affirmative action while it does not, itself, bias the survey outcome.

Yang also takes issue with the inclusion of “women” and “other minorities” in the NAAS survey question. Yet, affirmative action is not limited just to racial minorities: the Civil Rights Act established affirmative action programs as extending to racial and religious minorities, and subsequent case law expanded affirmative action programs to include women (via Title IX), the young and the old (via the Age Discrimination Act of 1967) and the disabled (via the 1990 Disabilities Act). To poll attitudes towards affirmative action while failing to acknowledge the full spectrum of identities to which affirmative action programs apply would be to provide an incomplete  — and even leading — definition.

Yang chooses to simply ignore this careful and deliberately unbiased methodology to assert that a different (and unwieldy) question wording — “Do you favor or oppose race-based affirmative action programs with the intention to help blacks and some other minorities (excluding Asians) to get better education, at the expenses of whites and particularly Asians, who have been historically discriminated against? (Please note that according to some studies, these affirmative action programs may actually hurt students they are intended to help.)” — would be more appropriate and would yield a different result. I agree that this question would yield a different result, but I would further argue that any result obtained by this question would be biased and therefore meaningless. It is Yang’s question — not Ramakrishnan’s — that is leading with its partisan reference to Asian exclusion; moreover, Yang’s is asking a different thing.

The questions posed by Pew and others are “principle” questions; the respondent is basically being asked whether or not they support the principle of affirmative action, irrespective of how those programs are administered. Yang’s question, on the other hand, is exactly the kind of overly specific “process” question that can — if administered carelessly — produce worthless results on this topic. Yang rightfully notes that “process” questions produce different answers on affirmative action than do “principle” questions; Pew also shows this in a 2009 survey on “racial preference”. Yet, “process” questions are fraught because in attempting to frame the process, they are readily biased by partisan interpretations of what the “process” actually is (as Yang’s own proposed question clearly demonstrates). For example, Yang cites a “more relevant” Gallup poll on “racial consideration” that shows two-thirds opposition, yet Yang fails to note that the Gallup poll asserted as part of the question that students admitted under affirmative action would also be less qualified, a highly contentious assumption.

The cited Gallup poll. Note that respondents were asked whether they supported a merit-only system, or a system where students would be admitted for the sake of diversity “even if that means admitting some minority students who otherwise would not be admitted”. This question wording posits diversity and student merit at odds with one another, and implies that the minority student must by extension be less qualified.

Gallup is, also, itself problematic: Gallup does not represent itself as a non-partisan group, and has been heavily criticized for its conservative bias. In 2012, The Guardian noted that Gallup was one of only two major polling groups to consistently show President Obama trailing behind Republican challenger Mitt Romney in the presidential elections, which should raise questions about its methodology; the group has also predicted Republican outcomes, often incorrectly, in many other races over the years. On affirmative action, Gallup is not “more relevant”, it is just biased towards more conservative views.

The experienced pollster knows to simply avoid issues of introduced partisan bias in designing his or her question by focusing  instead on a strict definition drawn from case law. That is, in fact, exactly what Pew and Ramakrishnan do: they offer a legal definition of these programs that focuses on their distinction from other admission and hiring policies by their specific goal: to address structural inaccess (to jobs and education) arising as a consequence of historic discrimination. To take issue with the presentation of this definition is to fundamentally misunderstand the legal history surrounding affirmative action, while advocating for an alternative that would produce the exact sort of cooked results Yang rails against.

Conservative groups are now attempting to capitalize on the opposition of some Asian Americans to affirmative action.
Conservative groups are now attempting to capitalize on the opposition of some Asian Americans to affirmative action.

A Hypocritical Focus on Self-Interest

When faced with a straightforward definition of affirmative action programs, Yang facetiously asks, “Who would not answer ‘yes’ to such a noble goal?”.

The answer is quite simple: one would answer “yes” to such a question if one supports affirmative action programs. And, indeed, Yang actually appears to support the principle of affirmative action, but only if those programs offer direct benefit to him or people like him. He writes (emphasis added):

[T]he Field Poll included employment, where the situation is vastly different from college admission and where Asian Americans often face discrimination and are underrepresented, especially in management and executive levels.

Later, Yang writes that he supports class-based affirmative action to correct for socioeconomic disadvantages (never mind the ample evidence showing that poor White and Asian students have greater access to privilege than wealthy Black or Latino kids, or that class-based affirmative action has been shown to be a poor substitute for race-conscious affirmative action). Thus, Yang appears to support the principle of affirmative action — programs that address the effects of systemic discrimination —  but only as long as his in-group would be treated as an aggrieved minority. Thus, he makes room for affirmative action in employment to address anti-Asian American and Pacific Islander discrimination, but does not support affirmative action in education where he argues Asian Americans are “hurt” by such programs.

Yang fails to acknowledge that if affirmative action programs have value in federal contracting and hiring  to address historic discrimination against multiple minorities (including Asians), than they must by extension also have the same value in college admissions to address historic discrimination against multiple minorities (including Asians and Pacific Islanders, but perhaps no longer including some East and South Asians). Yang further forgets that Asian Americans have historically been treated as a “preferred” group under race-conscious affirmative action programs, which have traditionally included Asian American (specifically Chinese and Japanese American) recruitment and retention programs to correct for decades-old anti-Asian exclusion.

To argue in favour of affirmative action programs in regards to employment, but to argue against affirmative action programs in college admissions — and to do so based only on the specific and contemporary circumstances of one’s in-group — is to be wildly hypocritical. It is to argue not against the morality of affirmative action, but in support of a “me-first” politic where the only rationale for one’s support of affirmative action is based entirely on whether or not it helps oneself get ahead.

Erasing the Voices of Southeast Asian Americans and Pacific Islanders

AFFIRMACT 040614Yang also makes the startling generalization that “it is an indisputable fact that Asian Americans are hurt most by race-based affirmative action in college admissions“; yet, this is a “fact” that is actually quite heavily disputed. Setting aside for a minute reports that show that the kinds of informal interactions facilitated by campus diversity benefits all students, Yang applies a gross generalization of his own political narrative to the affirmative action debate and to the associated AAPI experience. The central thesis of his column — “Asian Americans would lose out under affirmative action” — is a shocking erasure of the multiplicity of the AAPI experience: Yang uses the term “Asian American” as synonymous with the experiences of East and South Asian American students, yet the term “Asian American” includes Southeast Asian Americans such as Cambodian, Hmong, Laotian and Filipino Americans who are significantly underrepresented in the UC school system. As reported in last year, the National Commission on Asian-American and Pacific Islander Research in Education (CARE) found the following:

In the state of California, for example, Filipinos made up 25 percent of all AAPI residents in 2010, with Southeast Asians comprising 18 percent and Pacific Islanders 2 percent. But among all AAPI applicants to UC-Berkeley that year, Filipinos barely comprised 10 percent, Southeast Asians 13 percent and Pacific Islanders less than 1 percent.

Students of Southeast Asian American ethnicities are also Asian American, and our community maintains a strong alliance with those who identify as Pacific Islander; to write a column that simply erases their experiences from the “Asian American” narrative to instead buoy up a fantasy about “Asian American” exceptionalism on California’s college campuses should be alarming not just to supporters of affirmative action, but to any race activist interested in the project of inter-ethnic AAPI solidarity.


Standing Against Solidarity with Other Minorities

Speaking of solidarity, Yang disingenuously asserts that his opposition to affirmative action is actually in the best interest of African Americans and other minorities. To support this viewpoint, Yang cites the tired canard of Mismatch Theory.

Last, but not least, it’s highly questionable that affirmative action helps blacks and other minorities, which the poll takes as given. There is a famous book written by UCLA law professor Richard Sander and journalist Stuart Taylor, and the title says it all: “Mismatch: How Affirmative Action Hurts Students It’s Intended to Help, and Why Universities Won’t Admit It.”

Yang fails to offer any additional statements on Mismatch Theory other than the title of Sanders’ book (and that it is “famous”), so it’s left up to the imagination of the reader to determine which parts of Mismatch Theory Yang feels support his argument. Regardless, Yang fails to note that Sander’s book is scientifically flawed: it presents data that are largely not peer-reviewed and studies by other groups have reported the opposite to Sanders’ findings.  The only peer-reviewed paper to come from Sanders’ body of work presents data based upon what he dubs “strong racial preference” — the equivalent of a >300 point difference in SAT score — that problematically assumes an explanation of racial preference (instead of other applicant factors) for admittance of minority students with low SAT scores. It further compares (White vs. Black) applicants across a range of SAT scores that is larger than typically seen for two applicants  of these different races admitted to the same school. Even Sanders himself admits in an Intelligence Squared debate on NPR the difficulty in extrapolating from his data any predictions regarding the effect of “moderate” racial preference — a far more realistic scenario — on student performance. Critics of Sanders’ Mismatch Theory have pointed out several additional flaws in his work.

Yang goes on to cite changes in minority enrollment at the UC system, forgetting that absolute numbers do not take into account statewide changes in applicant population. He writes:

With Proposition 209 in effect since 1996, African Americans and Latinosnow account for a greater share of the University of California system’s overall admissions than when affirmative action was being practiced. In fact, Latinos’ numbers now exceed whites’ in UC freshman enrollment.

Yang simply fails to note that Latinos and Asian Americans are the fastest growing minority populations in the country, which includes the state of California; thus, consideration of enrollment numbers in the absence of this context produces disingenuous results. When one considers instead admission rate — the rate at which applicants of specific racial or ethnic backgrounds receive offer letters compared to the size of the applicant pool — we see that Proposition 209 had a devastating impact on minority admission rate, an effect that has persisted largely unchanged for the subsequent two decades.

Effect of Prop 209 on UC admission rates by race.
Effect of Prop 209 on UC admission rates by race.

Affirmative Action is a Complex Issue that Deserves Better Than This

defend-affirmative-action-tYang writes:

Race-based affirmative action is a complex and emotional issue. It requires a calm, objective and honest discussion. Biased or misleading polls and reports only serve to needlessly drive wedges between different racial and ethnic communities.

On this, Yang and I agree. Affirmative action is, indeed, a complex and emotional issue that requires calm, objective and honest discussion. Yet, the SCA-5 debate has been characterized this past year with an alarming amount of misinformation and bias, and while Yang’s column is relatively tame by comparison, it nonetheless continues to draw upon the same shoddy arguments as were first presented in the heat of the debate earlier this year, largely without consideration for the arguments’ flaws. Affirmative action is a complex topic, one that challenges Asian Americans and Pacific Islanders to deeply consider not just our position as racial minorities but more fundamentally our position in the larger project of America: will we stand on the side of equal opportunity and education access, or will we advocate a myopic “me-first” politic that dismantles any possible solidarity with not only other people of colour but also with fellow Asian Americans and Pacific Islanders?

Yang’s column steadfastly ignores these issues to parrot the same tired arguments: the same link to that New York Times article purporting anti-Asian bias at elite universities coupled with the same vague hand-waving about college mismatch and the same uncontextualized UC enrollment numbers, sprinkled with a healthy dose of the same unspoken presumptions over Asian American cultural exceptionalism (as long as “Asian American” means “Chinese American”) and minority underachievement.

If the affirmative action debate has truly divided the AAPI community as deeply as it has, then the “other side” has a duty to meet a higher standard. It has a duty to present an argument that is better conceived than this.

Did you like this content? Please consider becoming a patron of Reappropriate and get exclusive access to the brand new Reappropriate vlog!

  • Leon

    Jenn, you certainly missed the point of Yang’s article. Yang was arguing that Ramakrishnan was basically using a survey question deals with “principle” for a “process” issue thus leading to a misleading view. Ramakrishnan indicated that the survey data suggested “the opposition to SCA-5 was probably the result of selective mobilization among those Asian American voters opposed to the measure”. It is not hard to see that nothing in the survey data would lead to such a conclusion. At least, one need a better survey question, a real “process” question. SCA-5 is about “process”, not “principle”.

  • Leon, there’s literally zero evidence that anything close to a majority of Asian Americans voice disapproval of or opposition to affirmative action. Yang’s column sought to cast doubt on those findings the way climate change denialists seek to discredit climate science: it’s always about improper question formulation or some other methodology failure.

    The problem is that Yang is completely wrong – under the most basic standards for social science surveys, the question he laments is completely permissible. Yang just doesn’t like the fact that he’s in the minority on affirmative action within his community.

    Further, it’s fair to question as Jenn does the base incoherence of Yang’s argument. To imply respect for affirmative action in employment and contracting while one denies it in education is farce, and if Yang and his ilk were not so concerned with their ‘me-first’ interest group politics, they may have seen that ideology failure coming. For my part, I’m completely exhausted with Asian Americans who pretend that affirmative action is wrong because they believe it harms members of their group, as if a status quo where Black and Latino students do not experience higher education in this country in large numbers is no one’s fault and has no victims we need consider.

    Thankfully, as the survey data suggests, most Asian Americans do not buy into Yang’s crazy.

  • Rebelwerewolf

    For my part, I’m completely exhausted with Asian Americans who pretend that affirmative action is wrong because they believe it harms members of their group, as if a status quo where Black and Latino students do not experience higher education in this country in large numbers is no one’s fault and has no victims we need consider.

    Me too. I appreciate the amount of data that Jenn has gathered and her analysis, and I hope it manages to change some minds (or at least coax some folks off the fence). That said, I’m one of those people for whom it’s a moral issue where it does not matter how many are with or against me. To support affirmative action only when it directly benefits oneself is not only selfish, it is misguided and illogical, because we all live in society together.

  • Rebelwerewolf

    If you are part if an oppressed group, and you think it would be nice for anyone else to give a shit about you or stand up with you, maybe you ought to give a shit about others and stand up with them. Unless what you actually want is to fully assimilate into the group with power and join them in oppressing others.

  • Yang was arguing that Ramakrishnan was basically using a survey question deals with “principle” for a “process” issue thus leading to a misleading view.

    That’s a charitable interpretation of Yang’s writing for which I see virtually no evidence in the original text of the column. But let’s say this is what Yang was inartfully attempting to say.

    Ramakrishnan indicated that the survey data suggested “the opposition to SCA-5 was probably the result of selective mobilization among those Asian American voters opposed to the measure”. It is not hard to see that nothing in the survey data would lead to such a conclusion.

    I actually don’t have a significant issue with Ramakrishnan’s conclusion, due mainly to my issue with this statement of yours:

    SCA-5 is about “process”, not “principle”.

    This has not been clear from the anti-SCA-5 side of the argument. My experience with those arguing against SCA-5 was that it was absolutely a “principle” argument as well as a “process” argument. In addition to rhetoric equating affirmative action to Chinese Exclusion, anti-SCA-5 opponents argued against the need for affirmative action in general. They argued against campus diversity as a compelling interest. They argued against the need for affirmative action, focusing instead on cultural explanations for racial achievement gaps. To that end, the unifying message from the anti-SCA-5 debate was one that opposed not just the process of affirmative action (which anti-SCA-5 activists also seemed to fail to understand given how they spoke about holistic review), but also a unified message that stood against the principle of affirmative action. Even in Yang’s article, he argues that affirmative action “hurts” African Americans and other minorities and cites Mismatch Theory, two statements that are at direct odds with the principle of affirmative action.

    Given this messaging, it is not a leap in logic to argue that the vast majority of AAPI voters do not stand with anti-SCA-5 activists against the principle of affirmative action.

    It further remains important to ask, why is it so distressing if Ramakrishanan’s study is both scientific and accurate and SCA-5 opponents are in the minority in our community? Why is this is so vitriolic a talking point for your side? If you are in the minority, what does it matter? That should speak to the power of your side’s mobilization, which no one disputes.

  • Leon, there’s literally zero evidence that anything close to a majority of Asian Americans voice disapproval of or opposition to affirmative action.

    Agreed. The only survey presented by the anti-SCA-5 is this highly unscientific “survey”/”poll”/”petition” conducted by 80-20, which endorsed an anti-affirmative action position and then asked internet users to voluntarily indicate their agreement or disagreement. Survey respondents were invited to respond anonymously, and could theoretically answer multiple times. They also self-reported their identity, their race and their geography. A single user could submit multiple responses for their friends and family, a pollster version of straw donations, and there is evidence in the published responses that this happened (unless users were frequently congregating their entire extended family in a single place and individually signing the petition one after another by passing the laptop around). This “survey”/”poll”/”petition”/whatever reported a >90% opposition to affirmative action in the AAPI community, an outcome that is absurdly unrealistic — there is almost no poll in America that ever comes out that lopsided on any issue, let alone one as contentious as affirmative action.

    That single 80-20 survey is the total sum of the “evidence” that could ever be possibly presented showing anything close to a majority of Asian Americans disapproving of affirmative action.

    Pit that survey against findings by Ramakrishnan, by Lien and by other smaller studies, and it becomes quite clear: 60% of AAPI support affirmative action. Fact. Period.

    What the anti-SCA-5 side needs to do is adapt this fact into their talking points, not fall into anti-scientific ramblings. That they are in the political minority is not, in any way, lethal to their position (leave that to other logical inconsistencies); it becomes problematic when they take the route of science denial.

    Yang’s column sought to cast doubt on those findings the way climate change denialists seek to discredit climate science

    The parallel to climate change denialism is on point: Richard Sanders is equivalent to finding the one climate scientist who argues against climate change against the 99.9% of scientists who are desperately trying to prevent the end of the world, and propping up that one scientist with the shoddy debunked science as meaningful opposition.

  • @RWW

    I appreciate the amount of data that Jenn has gathered and her analysis, and I hope it manages to change some minds (or at least coax some folks off the fence).

    Thanks, and I hope so, although I doubt it. It’s not my first time presenting this rebuttals to this line of questioning (hence the last paragraph about the same tired link to that NY Times article and the same tired hand-waving about Mismatch), and it never really seems to affect things. I have repeatedly challenged anti-SCA-5 advocates to say something — anything — about their habit of erasing the SE Asian American and PI experience while generalizing the Chinese American experience to the term “Asian American”, to no avail. I’d like this post to be different, but pardon my cynicism.

    That said, I’m one of those people for whom it’s a moral issue where it does not matter how many are with or against me. To support affirmative action only when it directly benefits oneself is not only selfish, it is misguided and illogical, because we all live in society together.

    Agreed. As I wrote in my previous post, I’m fully prepared for Ramakrishnan to be wrong. My initial reaction to the 2012 NAAS results was skepticisim — the outcome seemed way too high and I expected at most a 50-50 split. I’ve since become convinced based on careful consideration of the methodology that the findings are more or less accurate, but had I found a problem with the survey, it wouldn’t have changed my opinion on affirmative action. My support of affirmative action is also moral, not based on popular opinion. If Yang believes his position is the moral one, than he shouldn’t care about the 2014 survey results either; instead he should focus on strengthening his argument and presenting something that is actually intellectually rigorous.

Comment Policy

Before posting, please review the following guidelines:

  • No ad hominem attacks: A person's identity, personal history, or background is not up for debate. Talk about ideas, not people.
  • Be courteous: Respect everyone else in this space.
  • Present evidence: This space endeavours to encourage academic and rational debate around identity politics. Do your best to build an argument backed not just with your own ideas, but also with science.
  • Don't be pedantic: Listen to those debating you not just for places to attack, but also where you might learn and even change your own opinion. Repeatedly arguing the same point irrespective of presented counterfacts will now be considered a violation of this site's comment policy.
  • Respect the humanity of all groups: To elevate the quality of debate, this site will no longer tolerate (racial, cultural, gender, etc.) supremacist or inferiority lines of argumentation. There are other places on the internet where nationalist arguments can be expressed; this blog is not those places.
  • Don't be an asshole: If you think your behaviour would get you punched in the face outside of the internets, don't say it on the internets.
  • Don't abuse Disqus features: Don't upvote your own comments. Don't flag other people's comments without reasonable cause. Basically, don't try to game the system. You are not being slick.

Is your comment not approved, unpublished, or deleted? Here are some common reasons why:

  • Did you sign in? You are required to register an account with Disqus or one of your social media accounts in order to comment.
  • Did your comment get caught in the spam filter? Disqus is set to automatically detect and filter out spam comments. Sometimes, its algorithm gets over-zealous, particularly if you post multiple comments in rapid succession, if your comment contains keywords often associated with spam, and/or if your comment contains multiple links. If your comment has been erroneously caught in the spam filter, contact me and I will retrieve it.
  • Did a comment get flagged? Comments will be default be published but flagged comments will be temporarily removed from view until they are reviewed by me.
  • Did you not play nice? You may have gotten banned and a bunch of your comments may have been therefore deleted. Sorry.

I monitor all comment threads, and try to address comments requiring moderation within 24-48 hours. Comments that violate this comment policy may receive a warning and removal of offensive content; overt or repeat violations are subject to deletion and/or banning of comment authors without warning.

I reserve final decision over how this comment policy will be enforced.


Play nice and don't be a jerk, and you'll do just fine.