The other day I summarized a Pew study of what factors Americans thought should be prioritized in making college-admissions decisions. High-school grades and standardized-test results were by far the most important criteria, with extracurricular activities a ways behind and race or ethnicity far behind, with only 7% of Americans seeing it as a major factor in college admissions, 19% as a minor factor, and 74% as “not a factor”. (Remember, this is how people think things should be, not how they are.) Other factors were more important than ethnicity/race, including being the first in one’s family to go to college, while two were less important (gender and whether a relative attended the school).
This ordering of criteria held for all groups surveyed, including ethnic groups, Republicans and Democrats. The importance of criteria varied, with race/ethnicity—the subject of debates over affirmative action—being rated as a more important criterion for Asians, blacks, and Hispanics than for whites; but in no ethnic group did fewer than 59% of respondents see race/ethnicity as “not important” as a factor.
Today we have a related study from Inside Higher Ed (click on screenshot below), which gives nearly identical results from a different sample (and using different questions). The study was conducted by by The Washington Post and the Schar School of Policy and Government at George Mason University. 1,238 adults were surveyed (Pew used 11,687). The results are given in the headline:
There’s not much text, but the article gives two bar charts. If we compare the “support Supreme Court ban on affirmative action” data below with data from the Pew study (using “race not important” as the equivalent of “support Supreme Court ban”, we get comparable but not identical results. The comparison: of all US results, 63% here as compared to 74% in Pew, among whites we get 66% here compared to 79% in the Pew study, among blacks we get 47% here as opposed to 59% in the Pew study, among Hispanics we get 50% here as compared to 68% in the Pew data, and for Asians (plus Pacific Islanders) we get 65% below compared to 63% for Asians from Pew.
In all but one group, support for Supreme Court ban on affirmative action in runs lower than the data for those who don’t think race should be an important factor in admission (Pew)—almost surely because people coul be opposed to seeing their opinions made into law by the Supreme Court. And in fact more than half of blacks oppose a Supreme Court banning of affirmative action.
However, every group polled thinks that programs designed to increase racial diversity of college students are a “good thing”. This is not expected from the data above, in which most groups show support for banning affirmative action. This is somewhat of a paradox because affirmative action is one of the only programs that can increase diversity among college students.
The conclusion: Americans are to some extent confused. Most support the Supreme Court making laws banning affirmative action yet most think diversity is a good thing. And most don’t think race should be an important factor in admission. Perhaps the conflict here is the coercion imposed by a Court ruling, or perhaps respondents saw other ways to increase the good of racial diversity on campus beyond a Supreme Court ruling. I suppose there are many explanations for this disparity, including simply different samples, or that, in the poll reported here, the respondents the way they thought the pollsters wanted to hear. Readers are welcome to comment.
I think racial diversity is a good thing, but my dilemma has always been that to increase it, one needs to reduce the “meritocratic” standards usually used for admissions. I’m coming around to the view that affirmative action is tenable but on a socioeconomic rather than a racial basis. That kind of affirmative action will certainly increase racial diversity, but will do so without violating the upcoming Supreme Court decision that will ban race as a criterion for admission. (It’s almost a given that this decision will occur.)
Isn’t the simplist solution to this “contradiction” simply that many people don’t think race should be used for admissions, but aspire to a society where diversity has been accomplished before students enter university? Obviously that’s (at present) naive and profoundly aspirational, but I think it’s a good description of how many view this issue and apparent conflict.
Yes, but realizing the aspiration is something people better think about.
May I respectfully ask why? I don’t mean the cynical view that having something as an aspirational goal means precisely that we intend not to think about it. Rather, I ask why it is important or desirable that diversity be achieved for its own sake, that makes it something we “[had] better think about.” You seem to be foretelling grave consequences if diversity is not achieved and I’m curious as to what you think those might be.
Cosmetic diversity in Canada (and to a useful extent viewpoint diversity also) is a by-product of vigorous immigration, which we need to prevent demographic stagnation. Since almost all our immigrants come from majority non-white parts of the world, immigration leads to diversity whether we want it or not. But in the States, you’re all American: diversity just means differently coloured Americans. I’m not really clear about what that gains you.
Leslie, you have me thinking about the difference between proportionately representative diversity and disproportionate diversity. 13% of the US population is African American; two of the nine (22%) US Supreme Court Justices are; that means the demographic is overrepresented, and yet it was hailed as being a success of diversity.
I’m not a fan of affirmative action, but I want things limited to proportionately representative diversity. Don’t go out of your way to push the percentage of college students above 51% female, 13% African American, 18% Hispanic, etc., or whatever the precise numbers are for any given age cohort of students.
Of course, once you factor in the global population – after all, humanity does not end at arbitrary borders – all of those statistics are out the window. I guess we could apply “proportionately representative diversity” policy to citizens only, without any such policies for non-citizens.
That is the expensive option but the best option. Make sure the public schools in the underrepresented communities are at the same level (or better) in the underserved communities and give the extra assistance needed to help those students benefit from that school. The problem with this solution is it takes money and time but will give the best result.
This experiment is on going in New York state. It is not the only place and it has been going on for some time. The results suggest that differences in spending are not critical factors for success in education.
Although there are differences in spending between wealthy school districts and poorer ones in NY, the differences are small (see figure 2 here) and even the poorest NY school district is spending nearly double the national average per pupil,. Still, despite all that spending, NY students are doing the same or worse than their peers (see link in my comment below).
The experiences in New York and elsewhere are saying (loudly, IMO) that increasing per pupil spending is not the answer to the problem and that focusing on that option takes focus away from where answers (if they exist) may lie.
An alternative explanation is cultural. Families differ in how they raise kids, and the differences affect lots of life outcomes. Family income and school funding seem to matter less than whether there are two parents in the home, whether they read to their kids from a young age, and whether the parents value and enforce norms about studying and achievement. Those things are hard to engineer through a school bond or a tax change. And people are sensitive about this bc it has a blame-the-victim vibe. But it does seem to be mostly true.
Diversity would also increase if disparities in performance were reduced. Most Americans probably accept the level playing field idea and steps to improve performance by minorities (and probably white, as well).
Despite decades of debate and numerous programs to reduce disparities, the disparities remain. I think the core reason is because it is too hard and political suicide to do what it really takes to level the playing field. In the face of limited budgets and historically segregated housing, what it would really take is movement of tax $$ from affluent districts –> poorer districts. That won’t happen because when people believe that something is being taken away from them, they get very angry very quickly.
And yet, after lots of attempts to solve the problem by spending money, lots of studies show that money spent is barely a significant factor.
Case in point; New York state, with the highest per pupil spending (nearly double the national average) has lower or no difference in outcome, depending on grade level. For pre-covid results; https://files.eric.ed.gov/fulltext/ED604331.pdf
It isn’t just money.
Very true, but few wish to admit it. Spending in the poorly-performing schools in Baltimore is more than $20,000 per student, while M Zuckerberg granted Newark’s schools one hundred million dollars some years ago with no improvement whatsoever, nor are those clamoring simply for more funding likely familiar with the famous failed experiment in social engineering, through the schools, in Kansas City during the 1990s.
If you think that tax money is not now being transferred from high-income earners to low-income districts and their denizens, your tax bracket must be too low for you to be aware of it. (Most redistribution money comes from individuals in the top quintile. “Districts” themselves aren’t taxpaying entities.). Where do you think money for policing, fire protection, schools, healthcare, public health, snow removal, public transit, parks, basketball courts, and welfare to poor neighbourhoods comes from now? They certainly aren’t self-financing. How much more of other people’s wealth do you propose to confiscate in the name of eliminating disparities? What evidence do you have that this would work in the sense of producing socially desirable results? More desirable than what current redistribution achieves, I mean.
I think Affirmative Action had a place in society many years ago, and we have made much progress over the years. However, I think today, Affirmative Action has outlived its usefulness, and is now causing more harm than good. It is now taking racism in the opposite direction, and all the good that was achieved is now being reversed. Racism is just as bad from either direction, You can’t fight racism is with more racism. Defining anyone by simply the color of their skin, and not from who they are as a person, or from their accomplishments is wrong. Martin Luther King, Jr. said it best in his “I Have A Dream” speech – we need to judge men NOT by the color of their skin, but by the content of their character. We need to eliminate the RACE questions on all applications, college and work. Let’s go back to meritocracy, and award people who are the best at what they do rather than by a label we have pinned to them.
Using only meritocracy for admissions applications will guarantee that under-represented minorities that are disproportionally in lower income school districts will continue to be under-represented in colleges. It will decrease diversity at most colleges, rather than maintain it.
Yes, but you yourself say “lower income school districts”. The disadvantage is being poor, not the colour of your skin. It’s true that black people are more likely to be poor than white people but let’s target the real handicaps for academic achievement rather than an imperfect proxy.
Wealthy blacks still do worse in school than poor whites. So focusing on income disparities won’t increase black achievement. Quotas, on the other hand, are sure to succeed.
Do they? I’d like to see the evidence for that.
Which of the two claims do you dispute?
https://www.nytimes.com/interactive/2018/03/19/upshot/race-class-white-and-black-men.html Warning: school performance was not assessed in this anonymized census-based study.
Black boys from wealthy black families slide down the socioeconomic ladder as they become adults. Black girls do not. Poor white boys are much more likely to rise up out of poverty than black boys. The exception was as follows:
**The authors, [of the Equality of Opportunity Project, now based at Harvard] including the Stanford economist Raj Chetty and two census researchers, Maggie R. Jones and Sonya R. Porter, tried to identify neighborhoods where poor black boys do well, and as well as whites.
“The problem,” Mr. Chetty said, “is that there are essentially no such neighborhoods in America.”
The few neighborhoods that met this standard were in areas that showed less discrimination in surveys and tests of racial bias. They mostly had low poverty rates. And, intriguingly, these pockets — including parts of the Maryland suburbs of Washington, and corners of Queens and the Bronx — were the places where many lower-income black children had fathers at home. Poor black boys did well in such places, whether their own fathers were present or not.**
The anecdotes in the NY Times article suggest in that these neighbourhoods, “black” people are more likely to be immigrants from modern-day African countries, not Blacks descended from slaves as the term is usually used. It must confound people like Ibram Kendi (quoted in the NY Times article) that people of colour clamour to immigrate to a country where they will surely according to critical race theory be oppressed by whites. but thrive nonetheless.
Black men raised in the top 1 percent of the income distribution were as likely to be incarcerated on any given day — about 2.1% — as white men raised in households earning about $36,000. Black men and women were much less likely to be married at any income percentile than white men and women.
RE increasing racial diversity by using economic disadvantage as an admission factor (note that “affirmative action is not an issue that directly affects most college students, because the majority attend schools that are not at all selective”):
Susan Dynarski: At Elite Colleges, Racial Diversity Requires Affirmative Action
ungated: https://archive.ph/4e3tG
Susan Dynarski is a professor of education, public policy and economics at the University of Michigan; since July 2021 professor of education at the Harvard Graduate School of Education
A quote from the article:
“A close look at the numbers shows that the only effective way to increase racial diversity at elite colleges is by considering race when deciding who gets in.
There are proposals to get around the affirmative action controversy by ignoring race and instead paying attention to economic disadvantage. Give poor applicants a boost and greater racial diversity will follow, so this argument goes.
But this approach can’t do the job of race-based affirmative action for a very simple reason: Most poor people are white. Putting a thumb on the scale for low-income students will help far more white students than black or Hispanic students.”
This may be of interest to PCC(E) and readers here:
Peter Schuck: One nation undecided: Clear thinking about five hard issues that divide us. Princeton Univ Press, 2017
Ch.5 Affirmative action, pp.273-314
Also interesting:
John Ellis: A More Diverse America Turns Against Racial Preferences
Wall Street Journal, Eastern edition; 15 Oct 2022: A.15
ungated: https://archive.ph/gAocj
Mr. Ellis is a professor emeritus of German literature at the University of California, Santa Cruz, and author of “The Breakdown of Higher Education: How It Happened, the Damage It Does, and What Can Be Done” (Encounter Books, 2020).
Quote from the article:
But as the public attempted to slam the door shut on racial preferences, the universities were busy trying to open it wide. The stealthy end-runs around the law gave way to support for “equity”: the desire for racial proportionality in all things — never mind that the Supreme Court has held that quotas in college admissions are unlawful. Accordingly, many colleges have begun to abandon the use of test scores in applications.
In line with this hardening of campus attitudes, increasingly powerful diversity, equity and inclusion bureaucracies arose to achieve these aims …
The most visible sign of DEI’s clout is its gradually seizing control of faculty appointments. At the University of California, Santa Cruz, DEI personnel prescreen applicants for all faculty positions and can throw out applications whose mandatory statements of commitment to diversity they don’t like. At Berkeley about 75% of applicants for a teaching position in life sciences were rejected in this stage during the 2018-19 academic year. The prescreening resulted in Hispanics representing 59% of the finalists, despite comprising only 14% of applicants. White applicants made up 14% of the final pool, down from their original 54%.
Not only is this practice an illegal political test for faculty employment, it’s also a stunning reversal of the policy that once made our universities great. For decades, the hiring of faculty was driven by the judgment of competent professionals, not by ideological administrators.
After reading Ellis’ WSJ article. I read his book: “The Breakdown of Higher Education: How It Happened, the Damage It Does, and What Can Be Done” (Encounter Books, 2020).
I found it very interesting and recommend it to PCC(E) and readers here.
Peter thanks that’s good stuff. Quotes like this make me despair:
“Most poor people are white. Putting a thumb on the scale for low-income students will help far more white students than black or Hispanic students.”
In 2021 ~8% of members of white families in the US lived in poverty vs. ~18% of members of black families (for all people not just families the difference is similar ~10% vs. ~20%). Favouring poor student applicants helps a much larger proportion of black families (or individuals) than white families.
The only way an economist like Dynarski can say this with a straight face is by not specifying that “putting a thumb on the scale…will help far more *poor* white students than poor black or Hispanic students.” This is sadly what the Davos demographic means by affirmative action: only poor black people are worthy of a hand up; poor white people deserve what they get.
Maybe I’m being pedantic, but the 53:47 split here is not significantly different from 50:50. With an overall sample of 1238 people, and assuming that 13% of them are black, there would only be 160 black respondents, and the error bar on that is about 10%, so the 53 really means anything from about 43 to about 63. Indeed, in the above charts, none of the racial groups differ significantly from the figure for the overall population.
Not pedantic at all. A very important observation. In elections, a 53:47 result is a decisive victory because there is no error bar around the result. Polls, especially subgroups within polls, are subject to error estimates for the reasons you point out. Thanks.
Or perhaps most people are lazily assuming that what was true in the past is true today: college admissions officers tend to look down on or discount applications from minority students. Because race shouldn’t be an important factor for who gets accepted, this discriminatory practice should be stopped, and this will allow more diversity on campus.
As evidence, they’re using numerous popular movies and tv shows about plucky black people overcoming racist barriers in the 1920’s – 60s. Yes. We need more of this sort of thing.
Thought experiment: Assume that over the next ten years we could solve the pipeline problem and double the number of racial minorities who were academically prepared for college. Next, assume that our colleges successfully recruited, taught, mentored, and graduated these students. Success in diversity, right? Now assume that virtually none of these students had been sufficiently prepared for Ivy League-level attendance; all had attended state flagships or, even, second- and third-tier institutions. Success?
Imagine a college admissions officer from an institution known for rigor and a student body with median SAT scores of 1500 coming into contact with talented minority students who have respectable scores of 1200 – 1300. Would that officer, and his or her host institution, consider helping to place these students in a different institution better matched to the students’ preparation, or would they admit some of them, despite a likelihood that the students would struggle among more academically advanced peers? (One could ask the same about a white, first-generation student from a low-income, working-class community.)
I ask this to help tease out our objectives in pursuing diversity. Are we principally concerned with individual minority students and helping each succeed in life? Is the chief concern the broad, social value of access to opportunity? Or is diversity more an institutional goal, the desire of each individual college to claim that it is “diverse” and enjoy whatever educational and social value comes from being such? And if it is the latter, does that sometimes work at cross purposes to the other goals?
You assume that a student with an SAT score of 1300 would struggle in a school with a median SAT score of 1500. First, if the median score is 1500, half the students would be below that number, many in the 1300 range. Second, it is far from necessarily true that a student with a 1300 score would struggle in such a school. A 1300 score does indicate a talented student, who with hard work can easily compete with students that scored a little higher on the SAT. You give too much weight to SAT scores as a predictor of academic performance, especially when there is not a great difference in scores. Per the article linked below, a student with a 1300 score is in the 92nd percentile. I don’t consider this a significant difference compared to a student in the 99th percentile.
https://blog.collegevine.com/new-sat-vs-old-sat-score-conversion-chart/
I appreciate the comments, Historian. To clarify: I said that a 1300 is respectable, but in a student subpopulation with a median score of 1500, that 1300 student is no longer at the 92nd percentile; she is near the bottom of the class of most such schools. I do not assume that such a student would necessarily struggle. I said there is a “likelihood”, and I still contend it is probable. Of course, it is not unusual for the SAT scores of disadvantaged students to understate their potential: lack of test prep, mentoring, rigorous course availability, etc. Nor is it unheard of, I suppose, for a college to insufficiently challenge its median student. In each case, a lower-scoring student could succeed in a population of “better” students. To reverse the scenario, it is also possible that a 1500-scoring student would not be bored in a curriculum geared toward those at the 1300 level. But there is also a good chance that such a student would, in fact, be somewhat bored and not be challenged sufficiently or perform to potential.
The “elite” schools are in a bind, aren’t they? Their reputations hinge on exclusivity, yet they beat the drums of inclusion and diversity thus the more loudly. You suggest that test scores are not so relevant, and I am open to that, to a degree; many of our state flagships, after all, educate students across a broader range of academic abilities and preparation than do many of their private elite counterparts. Here is a proposal. The Ivies and like institutions can form a consortium, set a minimum combination of SAT / GPA / class rank that they deem predictive of success in their curricula, and all agree to admit by lottery. I propose a minimum SAT of 1180; after all, the 80th percentile is respectable (right?), and it is still well above that of many colleges.
Diversity and inclusion problem for the “elites” quite nearly solved—at least as best as one might solve it given the inherent tension between these goals and the desired reputations (and need?) for exclusivity. I do wonder, however, how long their faculty would stick around if they had to teach a representative student population of the caliber anywhere near that of the average college in America.
I’m glad to hear that people don’t want institutional preference for certain demographics. I hope that people of all backgrounds (citizenship, visa status, etc.) are also treated equally in college admissions. It’s always sad to see people think that discrimination on the basis of citizenship is more acceptable than discrimination on the basis of ethnic heritage.
Now if only we could convince the governments of the world that all humans should have equal access, with no bureaucratic hurdles. The visa process is rough.
At great Canadian universities (McGill, Toronto, UBC), 30-40% of the operating budget is paid for by the provincial government (vs. ~10% at great land grant universities in the US). Favouring Canadian citizens in admission makes sense because citizens paid much of the cost to make those folks better educated (and employable). At my small good (not great) university, equal access to all humans would mean admitting <100 Canadians (population ~38 million) each year plus thousands of folks from China, India, and Africa (population ~4 billion). That's not a bad thing on its face, but it's not what Canadian citizens paid taxes to do.
I’ve seen some governments basically using a voucher system to fund higher education. Assume the basic unsubsidized cost per year is $10,000; instead of handing money to universities for each in-state or domestic student, hand the same amount to the students so that they can study anywhere they want in the world. 90% will still study at home, and the other 10% will bring back something worthwhile. That means that local taxpayers are not subsidizing foreign students, but do pay for some of the local kids to go abroad.
It’s just one policy of many. There are lots of options if you start looking at solutions around the world. Oh, the solution above is one I have seen most in governments practicing petrosocialism.
I agree with John McWhorter on this. Affirmative action for higher education should be based on socio-economic status and not race. However, as non-Asian people of color are disproportionately of lower SES, then this will also have the effect of increasing racial diversity as well.
So a win-win it seems…addresses the major problem without alienating millions of poor and lower class white people.
I see no contradiction in favoring diversity while not favoring affirmative action as a means to diversity. IMO, however (and I’m under the impression that our host agrees with this), the goal should not be diversity at all, but rather equality of opportunity. And to attempt to determine whether equality of opportunity is in place by looking at diversity is clearly circular. Once equality of opportunity is in place, diversity or lack of diversity becomes irrelevant.
I don’t see a contradiction, as the two issues overlap, but are not mutually exclusive. I think people like the general idea of diversity, as in the old coke commercial with all the happy young people teaching us all to sing.
Liking that in abstract does not mean that they approve of the idea of qualified and motivated students being turned away because of their race.
The second question is a generalization. The respondent might be thinking of programs to offer tutoring to primary school kids in poor areas.
I think the difference between the two questions derives from the fact that one was strictly for college admission, and the other for affirmative action in general. Most of affirmative action, as it impacts average black people, doesn’t have anything to do with college admissions, it’s preferential treatment for government jobs or preferences for minority contracting. After the closing down of factories/replacement of poor blacks and whites by people from Latin America in low-skilled factory jobs, government jobs are an important path to middle class and suburbia for black people (afaik).
I expect everyone will have moved on from this post, but popping up on Wilfred Reilly’s Twitter feed just now (he’s well worth following by the way), is a graph that purports to show that there is relatively little difference in US public spending on schools according to racial group, and that, indeed, slightly *more* is spent on blacks than on whites. Here’s the Tweet.