Social Mobility, College Admissions, and that Biased SAT
[please read this essay from the beginning, or at least, read Part III before this one]
My examples of elite occupations have been engineer, professor, doctor, lawyer, MBA, scientist. Besides good pay, satisfying work, and social esteem, these professions all share one feature: the requirement of, or at least the utility of, a graduate degree.
Translated: the best jobs require schooling beyond the four year college degree. Here are pertinent facts about graduate school: 1) graduate school is higher higher education, more demanding than college of all the things that college demands; 2) admission is competitive, in many cases more competitive than for elite four year colleges; 3) testing plays as central or more central a role for these graduate programs as for college admissions (here, the GRE, LSAT, MCAT or GMAT, rather than the SAT); 4) the cost of another 3-6 years of tuition, and another 3-6 years out of the workforce, raise the total financial demands of higher education far beyond the reach of a family located down in the middle of the pear.
*Engineering is the exception: you can make good money, and advance in your career, with only an undergraduate degree. But to get that undergraduate degree, you will have to take multiple math courses beyond calculus. Plus, engineering may be the one discipline where claims that tests are bogus—that math test scores are bogus—would be regarded by most as ridiculous. So I think the point holds.
**I suppose I should add banker or financier to the list of occupations. These also don’t require a graduate degree, although MBAs and law degrees are not uncommon; but in the absence of a graduate degree, you need to have attended a select group of elite colleges, or otherwise gained access to that social network, to join Wall Street. So the point about a narrowed funnel again holds.
Implications of graduate school as a hurdle
If you were prone to conspiracy theories, you might regard the imposition of graduate education, as the hurdle for getting a really good job in America, as a system diabolically designed to favor children of parents of higher socio-economic status. The cost, the emphasis on testing, the repeated emphasis on top academic performance at top schools—all conspire to weed out families of moderate means, excepting those foolish enough to pile up hundreds of thousands of dollars of debt. Which will work well enough to constrain their socio-economic rise.
It bears emphasis: a low cost undergraduate degree from [Compass direction] [State name] University is unlikely to make you a competitive applicant at the graduate program of your dreams. A degree from a flagship public institution like the University of California or University of Michigan, or preferably from an Ivy League or other elite private school, is close to the minimum expectation. And you will have to distinguish yourself further, within that elite pool, to gain entrance to medical school, or a top ranked Law or MBA program.
Test, test, test. Desk, desk, desk. Math, math, math (except for lawyers).
The requirement for graduate education tightens the school screw. The funnel grows narrower. Only the most cerebral, with the most grit for desk work, all day, every day, can pass through.
We glimpse why it is so important, among some groups, to impugn the legitimacy of tests like the SAT. It’s just another brick in the wall, one more obstacle in the way of social advancement.
To put some numbers on the pear: let the upper middle class bulge account for 15-20% of the population. Turnover each generation should be at least the 20% seen among the medieval nobility. Conclusion: there’s room for 4-5% of the population to move into the upper middle class with each generation. Relative to the 70-75% of Americans occupying the larger bulk of the pear, that translates to upward mobility for 5 to7% of these middle class families (.04/.75 to .05/.70). Do well in school, score high on tests, get into an elite college, get good grades, take more tests, get admitted to graduate school, endure graduate school, get a great job, arrive at last. The Way is narrow and strait, but it is open to more than a trickle of traffic.
Unless affirmative action intervenes.
Sorry for the two-by-four up the side of the head. This will not be a screed against affirmative action; rather, I hope for a clear-eyed view of its benefits, costs, and consequences, in the narrow context of upward social mobility via college admissions.
Returning to the numbers: in America, the bottom 5-10% of the pear overwhelmingly contains ethnic groups historically subject to oppression and injustice. That’s a socio-economic fact. For generations, marginalized groups at the bottom had little chance to enter even the middle class, much less the upper bulge of the pear.
Now assume a social intervention, deus ex machina: begin to admit the top 10% of the children of this bottom group to elite colleges. As a result, 1% of the total population gains entry to these schools (.10/.10), from groups which had not had much access before. Next, and very important: do not increase the capacity of these elite colleges. Keep the size of the entering class constant.
Result: between 14% and 20% of the middle class families who would, in the previous generation, have gained entry to these elite colleges will now be shut out (.01/.07 to .01/.05).
I submit to you: that much narrowing of the funnel will be noticeable. The affected middle class group will sense the negative change in their life chances.
And if affirmative action were to widen beyond the single most eligible ethnic group, to include all manner of historically marginalized groups—trans-gendered and gay, as well as ethnically marked—then the displacement can only be greater (see my more narrowly focused essay on displacement here).
To review: the children of current occupants of the upper middle class strata have every advantage, with parents hell-bent on reproducing their presence in the upper bulge in the pear. Good nature, good nurture, good schools, plentiful resources, all act to boost chances for the scions of current occupants. Displacement, in so far as it occurs, will fall disproportionately on the middle classes down in the body of the pear, who will find it that much harder to move decisively upward.
Social position is a zero-sum game. And social mobility must be limited, for so long as parents want their own children to live as well or better. And membership in prestigious professions must be limited if these professions are to maintain their prestige and their financial rewards.
True, the capacity of elite colleges doesn’t have to stay fixed. But all kinds of forces bend toward limiting capacity. A school can’t expand greatly, lest the caliber of its entering freshmen class go down. A school dare not increase its class size, lest it look worse on the faculty-student metric. Professors don’t want to teach more. And tenure track professors have become very, very expensive. Net, we should not expect much capacity increase at the upper levels of the academic hierarchy.
I hope this helps you understand a little better the frenzy that surrounds the process of applying to college, and why declining acceptance ratios and other elements of the admissions process cause so much hysteria. I want to emphasize that the situation has changed markedly, relative to periods as recent as the late 1990s.
- Graduate education and higher mathematics didn’t used to be so much the eye of the needle through which the socially ambitious had to pass.
- Upper middle class incomes and lifestyles didn’t used to be so elevated relative to the broad middle. That status used to mean a bigger house on the same street; now that bigger house lies in a different neighborhood miles away.
- Affirmative action didn’t used to be so thorough-going. For many years, it was limited to tokens whose numbers didn’t tip the scale
- The national population has continued to grow, even as the number of elite college slots has held constant, driving all ratios down: there are ever more at the door.
- And the broad middle class didn’t used to be under so much financial pressure
Once upon a time, a four year college degree was more than enough to secure a comfortable place in a comfortable middle class in a comfortable country where incomes were rising broadly—if you were white and already middle class. The pain of exclusion was borne mostly by the non-white and the poor. That had to change; but no one said the change had to be easy, and we do our public discourse around these sensitive matters no service if we close our eyes to mathematical facts about social position and social mobility.
The Ideological Perfect Storm
To conclude, let’s summarize the ideological necessity that drives dismissal of the SAT and standardized tests as biased. And not just biased, that neutral little word, but unfit for the purpose for which they were designed, which was to select out better from good, and identify best from among the better: to cull those most deserving of receiving a higher education from the most highly regarded schools.
The logic runs as follows:
- By definition, populations that deserve affirmative action, need affirmative action; which is to say, have a lower distribution of test scores. (If scores weren’t by and large lower, this group would not need affirmative action to correct their admission chances.)
- It’s awkward for an institution wholly committed to meritocracy to be admitting students whose “merit” is lower, as measured by some long-accepted criterion.
- Therefore, it becomes necessary to impugn the criterion: that test must be biased—badly biased, beset by bias so great as to ruin its usefulness.
Next the white middle class piles on. The clerk’s son (see Part I) is likewise not advantaged with respect to the SAT. He went to an okay school; his home environment was not unsupportive; but he is advantaged only with respect to groups at the bottom. His child’s okay SAT results aren’t going to help him advance; the clerk too has an incentive to disparage standardized testing.
Beyond the SAT
On their face, it’s difficult to distinguish the GRE, LSAT, MCAT or GMAT from the SAT. All take the form of a series of multiple choice questions, based on reading passages of prose or performing mathematical calculations. If I drew items from all five tests and mixed them together, I doubt the ordinary college graduate could reliably pick out the SAT items from the other four. All these tests are designed to fulfill the same purpose as the SAT: to distinguish better from good, and best from better, in deciding who should attend an institution of higher education, in this case graduate school, and who must be turned away from the more elite programs, or even denied admission across the board (less than half of medical school applicants get accepted anywhere).
Must not the LSAT be as biased as the SAT? And if LSAT score is one of the best predictors of performance on the bar exam, itself a series of multiple choice questions, must not the bar exam be biased too, and in the same ways against the same people as the SAT?
Injustice!
And what about that final exam in multivariate calculus—do you think there is no socio-economic bias in how well students score on it?
But how else do we allocate the limited capacity of graduate programs in Law, medicine, science, engineering and business, except by culling better from good, and best from better?
Thinkers.
Next: the allocation problem
[…] Reflections on Social Class and College Admissions, Part IV […]