How ‘Undermatching’ Harms Smart Low-Income Students

radford.jpgMost readers of Minding the Campus are well aware of the phenomenon of “mismatching” in colleges first brought to national attention in regard to African American students by Cornell economics professor Thomas Sowell in the late 1960s and the early 1970s. Sowell showed that many of the black students at Cornell, who often had scores on national exams like the SAT that would have placed them in the middle of the pack at hundreds of respectable middle-selectivity colleges and universities, were upwardly thrust into the competitive hothouse of an Ivy League institution like Cornell where they were simply overwhelmed by the academic demands of the college curriculum. Many of these mismatched blacks, Sowell found, were on academic probation; some lashed out in frustration at the Cornell administration in the famous “guns on campus” episode.

Shunning the Best Schools

Since the 1970s we have learned a good deal about mismatch and in particular the harmful effect it has for undergraduates seeking a degree in a STEM subject (science, technology, engineering, math) and for those in law school who hope to graduate and pass the bar exam. (Readers who want to learn more about the harm of mismatch might begin with Richard Sander and Stuart Taylor’s Mismatch: How Affirmative Action Hurts Students It’s Intended to Help, and Why Universities Won’t Admit It).  In recent years, however, there has been a fair amount of research on a different kind of mismatch than the “overmatching” that Sowell and others have written about.  Some students, including many poor whites and students from rural and small town backgrounds, have SAT scores and a general high-school preparation that would enable them to gain access to highly competitive colleges and universities purely on their own academic merit without “affirmative action” preferences of any kind.  Yet they often fail to apply to such schools and wind up attending lower- level institutions where most of the students are considerably beneath them academically.  Instead of being “overmatched” they are “undermatched,” and researchers have sought to determine why this situation occurs and what might be done to correct it.

Over the past year two important studies have appeared addressing this undermatching phenomenon, which have received considerable coverage in the popular press.  Both see undermatched students as missing out on what more highly selective colleges and universities have to offer.  The first is by Stanford economist Carolyn Hoxby and Harvard public policy professor Christopher Avery. It appeared as a National Bureau of Economic Research Working Paper (“The Missing ‘One-Offs’: The Hidden Supply of High-Achieving, Low-Income Students”).  Hoxby and Avery had access to enormous data bases that enabled them to look at all students from lower- income backgrounds (bottom quartile of all U.S. households containing high school seniors) who scored at the 90th percentile or above on either the SAT (with scores higher than or equal to 1300) or the ACT (with scores higher than or equal to 29).  Since only about 40 percent of all secondary school students take one of these two college aptitude tests (the other 60 percent are usually not college-bound or intend to enroll in a non-selective college), this top 10 percent of SAT and ACT test takers represents the top 4 percent of all high school seniors — an elite group to be sure.

A Striking Finding

What Hoxby and Avery found was striking.  More than half (53 percent) of these “low-income, high achievers” do not apply to even a single selective college or university, defined as one whose median SAT or ACT score is within 15 percentiles of their own high score.  And some who do apply to high-selectivity schools often apply to just one, not realizing what a poor strategy this is.  “It is not unusual,” Hoxby and Avery write, “to see [low-income, high-achieving] students who apply to only a local non-selective college and one extremely selective and well-known college — Harvard, for instance.  No expert would advise such a strategy because the probability of getting into an extremely selective, well-known college is low if a student applies to just one — even if the student’s test scores and grades are typical of the college’s students.”

Those low-income high achievers who do apply and enroll in very competitive colleges are usually those who come from large central cities “where they are likely to attend selective, magnet, or other high schools with a critical mass of high achievers.”  Such schools, Hoxby and Avery report, typically have not only many students applying to higher-selectivity institutions, but have recent graduates who have enrolled in such institutions and guidance counselors who are savvy about the prospects of their better students getting admitted to such schools.  They are also usually aware that the more elite colleges and universities, while they have high “sticker prices” in terms of tuition, room, and board, also have much more money than lower-ranking institutions to give out in scholarship aid.  These urban institutions are disproportionately black, Latino, and Asian.

Call on the Alumni

Many of the low-income, high-achieving students who do not currently apply to very competitive colleges are white and live in areas far from the major urban centers, often in small to medium-sized towns or rural areas.  These are the “low-hanging fruit” whose number Hoxby and Avery estimate, for the high school graduating class of 2008, to be between 25,000 and 35,000. Rural and small-town high schools often have few students who score at the level of the typical student in an elite college, and those who do have few peers like themselves and little knowledge about either their prospects of gaining access to an elite institution or the scholarship advantages that acceptance to such an institution might provide. If the most competitive colleges and universities want to increase income diversity among their students, these are the obvious targets of recruitment.  Hoxby and Avery recommend that they do this by employing more of their alumni in the recruitment and information-providing process since it would be prohibitively expensive to have a permanent admissions staff large enough to send representatives to areas of the country thinly populated with high-achieving students.  Recruiting such students would clearly increase income diversity on many college campuses, but Hoxby and Avery acknowledge that it would not bring many more blacks and Latinos to campus, a prime goal of most institutions.

The other important undermatch study is by Alexandria Walton Radford titled Top Student, Top School? How Social Class Shapes Where Valedictorians Go to College.  Radford is a Princeton-trained sociologist who co-authored with Thomas Espenshade the influential study of affirmative action preferences and college-admissions practices, No Longer Separate, Not Yet Equal.  Radford’s most recent work is an in-depth study of approximately 900 high-school valedictorians all of whom graduated between 2003 and 2006. The valedictorians were selected from local newspaper accounts in five states, and each filled out an extensive questionnaire about their family background, their college admissions behavior, the colleges and universities they got accepted to, and the institutions they actually enrolled in.  Respondents were separated into three socioeconomic categories, high- medium-, and low-SES. A subset consisting of 55 of these students were selected for extensive personal interviews.

Fear of Overmatching

Much of what Radford found coincides with the findings of the Hoxby/Avery study.  Low-SES valedictorians, for instance, were much less likely to apply to a top-rated college than those from wealthier families, though their academic credentials were often equal to these higher-SES students.  While four in five of the valedictorians from high-SES families applied to at least one of the top 72 colleges and universities listed in the U.S. News and World Report ratings, only half of the low-SES valedictorians did so.  The low-SES valedictorians usually got much less encouragement from home to apply to elite colleges, often because their parents had little knowledge of such institutions, did not know about their generous financial aid, and encouraged their offspring to attend colleges close to home.  The low-SES students were often without friends, neighbors, or classmates who had ever attended an elite college, and some thought they would be overmatched intellectually at such institutions even though their own high school achievement and scores on standardized tests indicated otherwise.

It is for these reasons, Radford says, that the top colleges and universities have so few students who do not hail from middle or upper-middle-class backgrounds.  “Without affecting the quality of matriculants enrolled,” Radford writes, “top institutions could increase the proportion of less-affluent students in their student bodies” by a considerable margin.  Like many other commentators, Radford thinks greater socioeconomic diversity would be a very good thing for the top colleges and universities in America, and that those from lower-SES circumstances who have the academic qualifications to attend such institutions miss out on a great deal by attending lower-selectivity schools.  “America’s best students,” she writes, “should all be bound for America’s best colleges and universities, regardless of their socioeconomic status” (emphasis in original).  The benefits of attending a top-rated school, she says, include more generous financial aid packages, greater chances of graduating, and greater chances of going on to graduate or professional school.   To encourage more high-achieving, low-SES students to apply to elite colleges and universities Radford recommends better outreach to these students by college-admissions officers, better high school counseling, and better dissemination of information about the generous financial aid often available at these institutions.

Big Fish, Small Pond

I have three comments to make about the recent concern over these undermatched students.  First, the concern is well-placed in so far as many high-achieving students who come from lower-SES backgrounds and who are often the first in their family to attend college aren’t well informed about their college options.  Many, if better informed, would wind up applying to, and attending, much more selective institutions than they attend now.  The institutions would surely benefit from what affirmative action policies today clearly do not deliver — i.e. greater demographic diversity without compromising academic standards.  Whether the students themselves would benefit is a difficult question to answer, since there’s something to be said for being a big fish in a little pond (i.e., a high-end student in a middle-level school) rather than one medium-sized fish in a large pond of many equals and superiors.  And contrary to what many think, the independent “school effect” of attending a more selective college, once motivation and background factors are weighed in, is not very great — most of the income and other career advantages of graduates of elite colleges can be attributed to the greater talent, motivation, and drive of the students who get admitted to those institutions rather than to the institutions themselves.

My second comment concerns the misplaced hope of some that a better search for low- income, high-achieving students will greatly offset the middle-to-upper-middle-class bias found on most elite college campuses.  Without lowering standards, modest changes can certainly be affected in the socio-economic makeup of college campuses by reaching out to low-income, high achievers, and such changes would certainly be welcome.  But there is no getting around the hard logic that Richard Herrnstein explained more than 40 years ago.  Parents with high academic ability are much more likely to have children with similar academic gifts than parents with lower abilities (academic ability is partially heritable), and high academic ability in America, when combined with perseverance and hard work, usually translates into higher-paying jobs and higher socio-economic status. The children of highly educated professionals from Scarsdale will always be more likely to possess the intellectual and other talents needed to gain admission to elite institutions than the children of blue-collar workers from Brooklyn, and they will also have many other advantages that translate into a class skewing of the student bodies at the most selective institutions.  There is no getting around this fact.

My final concern is with the “brain drain” and its aftermath that would inevitably follow if top-achieving students universally spurned middle-level colleges. Having myself taught at four such colleges (Moravian College, Valparaiso University, College of New Rochelle, Rider College), I know how important such students can be to elevating the intellectual level of the colleges they attend.  Almost all sizable middle-level colleges will contain at least a few students who would do well at Ivy League institutions, but their loss to the institutions they now attend would be substantial.  I don’t know whether this loss should be of less concern than the diversity-enhancement gain their presence in the Ivies and other elite institutions would represent. I’m genuinely conflicted on this question.  Others may have a different weighing of this tradeoff, but it is important to realize that there is a tradeoff here and that high-end students who fail to attend high-end institutions do not necessarily do themselves or others a disservice.  They are, in fact, an invaluable resource to the lesser-selectivity institutions they attend.


  • Russell K. Nieli

    Russell K. Nieli is a Senior Preceptor in Princeton University's James Madison Program in American Ideals and Institutions, and a Lecturer in Princeton's Politics Department. He is the author of "Wounds That Will Not Heal: Affirmative Action and Our Continuing Racial Divide."

5 thoughts on “How ‘Undermatching’ Harms Smart Low-Income Students

  1. Perhaps another reason poor but talented students avoid the elite schools has to do with the chasm between the wealthy students who have endless prerogatives for vacation and week-ends, and the limitations for the poor, imposed by a paucity of resources. Two economic worlds rubbing elbows in class and living very different lives otherwise is uncomfortable for the have-nots.

  2. I do have one comment, much of anyone’s success come from networking, and the top schools have the best networking. One of the problems of social status is of course how you form networks. Still it is much easier to network when working within your own culture rather than another.

  3. Hoxby and Avery think “it would be prohibitively expensive to have a permanent admissions staff large enough to send representatives to areas of the country thinly populated with high-achieving students”. This notion should be contrasted with what Jacques Steinberg reported in “The Gatekeepers”, where the Wesleyan admissions officer pursued an elusive Native American student in a way that brings to mind a 19th century naturalist stalking a rare bird species. IIRC, that student did not survive in the school past his freshman year. The schools will devote considerable resources to find and admit some students who meet particular diversity criteria that is not simply related to geography.

  4. Students don’t necessarily get any better an education at the “elite” institutions than they would at a less prestigious school. Often, the reverse is true.

  5. Thank God these students are eschewing these “elite” schools. Whether they realize it or not they are simply avoiding the social and political inbreeding that is practiced at these “elite” schools. They are we will be a better country for their choices.

Leave a Reply

Your email address will not be published. Required fields are marked *