Equity in College Admissions

The Wall Street Journal calls the University of California system’s recent decision to eliminate the SAT and ACT in admissions “a historic blow to excellence in higher education” and declares that the regents have “put racial politics above merit.”

The reader comments are even harsher. “What next?” writes one reader. “Will Olympic time trials be replaced by coaches’ recommendations, essays, and interviews?”

Now that most highly selective schools have abandoned the SAT and ACT, at least temporarily, we need to ask: How can they admit students fairly -- and withstand legal scrutiny -- in the absence of a standardized test?

A recent article in EdSurge, and co-published with Slate, reports that the technology sector is offering institutions admissions algorithms that will help them predict college success -- and avoid discrimination lawsuits.

Even before the shift away from college entrance exams, colleges and universities had already embraced data-driven enrollment management and financial aid allocation, a subject explored by Cathy O’Neil in her influential study, "Weapons of Math Destruction." Algorithms offered a seemingly objective and scientific way to balance competing institutional interests: in maximizing tuition revenue and ensuring ethnic, racial and socioeconomic diversity and gender balance, while sustaining or raising the institution’s U.S. News & World Report rankings.

A major goal, then, was to quantify the likelihood that a particular student would enroll, and the level of need- or merit-based financial aid that would seal the deal.

Now, the challenge is even greater: Given high school grade inflation and the absence of board scores, how will selective institutions identify the most talented students?

The answer hinges on what kinds of student a college or university wants:

• A student with polish and strong high school preparation.
• A student with a special talent, whether athletic, musical or artistic, or scientific and technological.
• A student who contributes to a campus’ diversity, which might be geographical, socioeconomic, ideological or in some other form.
• A student who is likely to contribute to the institution’s economic well-being, either now or in the future.

What variables, in short, should the algorithms favor?

• Grade point average, therefore favoring students who take lots of Advanced Placement or IB classes.
• Advanced courses (like calculus) taken and completed successfully.
• A focused talent (like musicianship) or, conversely, breadth.
• Evidence of leadership.
• Evidence of grit and other socioemotional and noncognitive factors, like persistence, determination and an ability to overcome obstacles.
• Demonstrated accomplishments.

In each instance, it is possible to identify measurable indicators that can serve as proxies for the desired trait. But privileging any one variable can lead to new forms of bias. For example, should students from historically strong high schools receive preference? To what extent should variables that correlate with family income -- like the number of AP courses taken -- be valued?

For public institutions, another possibility is to follow the Texas example, and enroll a certain percentage of students from each of the state’s high schools. Currently, the Texas plan admits roughly 7 percent of the top-performing public high school students into the University of Texas at Austin, based on class rank. These students make up about three-fourths of the entering class, with the other quarter admitted using a holistic evaluation.

Weirdly, but perhaps not surprisingly, the 10 Percent Plan produces more racial and ethnic diversity than from the holistic admissions process.

The effect of the 10 Percent Plan has been to dramatically increase enrollment not only of students from underrepresented groups, but rural students -- which is why the plan has succeeded politically.

The plan has also had the positive side effect of boosting the qualifications of students at Texas’s other colleges and universities.

Note that the Texas plan is coupled with a cap on the proportion of out-of-state students.

I find the main criticisms of the plan -- that it rests (and may actually reinforce) on racial segregation or incentivizes ambitious students to enroll in underresourced schools -- disingenuous. It strikes me as a sensible way to balance merit, access, diversity and inclusion.

Another critique is that the plan overemphasizes GPA and fails to take sufficient account of other indicators of aptitude and achievement, making it more difficult to craft an entering class with a diversity of talents and breadth of interests. But since a quarter of the class is admitted on a basis other than class rank, admissions officers still have a great deal of flexibility in shaping the student body.

The most powerful objections are that the 10 Percent Plan failed to expand the number of high schools that consistently send students to UT and didn’t significantly increase the number of African American students or to increase the number of Latinx students proportional to their share of the college-aged population.

The explanation, I suspect, is twofold, involving money and campus culture.

Without aggressive recruitment, which also includes making prospective students aware of the availability of financial aid, even a plan like Texas’s will only have a limited impact.

Supplementing the 10 Percent Plan are tiered and scaled layers of support, including the Texas Interdisciplinary Plan, which identifies the quarter of the class least likely to graduate and provides those students with peer mentoring, learning communities and supplemental instruction; the University Leadership Network, a scholarship and professional development program that serves 500 economically disadvantaged students; and a Freshman Research Initiative in the natural sciences that serves 1,000 students in 30 different research streams.

Each of these programs is designed to foster a sense of belonging and a growth and success-oriented mind-set.

Whatever its weaknesses and limitations, the 10 Percent Plan demonstrates that talent is widespread and that gifted students flourish when given opportunity and appropriate support. Precisely because the plan rests on a widely recognized measure of student achievement, it has succeeded in gaining a level of legitimacy that more subjective approaches to admissions do not.

However, an attempt to implement something like the 10 Percent Plan for admission into New York City’s elite high schools -- which currently rests on a single standardized test -- produced such an outcry that the mayor backed away from the plan. Instead, the de Blasio administration increased the number of seats to 20 percent that are reserved for students from low-income families that scored just below the admissions cut-off and attend a summer prep program.

However, just 35 percent of those in the summer prep program are African American and Latinx, in a K-12 system in which they make up almost 70 percent of the student population. Still, the mayor found a less controversial and more broadly acceptable way to achieve his policy goal of promoting greater diversity.

In New York City, admission to the specialized high schools was regarded, rightly or wrongly, as a zero-sum game, given the city’s failure to increase the number of selective high schools.

Achieving equity and inclusion in a diverse society hinges on admissions measures that are widely regarded as legitimate. These also depend on having sufficient numbers of spaces at institutions that are regarded as roughly equivalent.

Part of the furor over the Varsity Blues scandal and controversy over affirmative action within the Ivy League is that the elite private schools have artificially limited the number of seats and have run an opaque, highly unpredictable admissions process. Although the number of highly selective, well-endowed institutions has increased, and enrollment at state flagships has risen somewhat, the number of talented applicants far outstrips supply.

There are several paths forward. In recent years, many broad-access regional and urban campuses have instituted honors programs to provide gifted students with an experience that combines the mentoring, seminars and co-curricular activities of a liberal arts college with the resources of a large university.

Another is to identify a niche and identity in which a campus can achieve a reputation for distinction (for example, the University of Minnesota at Rochester’s collaboration with the Mayo Clinic School of Health Sciences or UT Dallas’s growing reputation in STEM).

Among the tragedies likely to grow out of the pandemic is that an important option will fade. Many small institutions that offer an education of the highest caliber will face financial pressures that will diminish their quality and appeal for years to come.

Steven Mintz is professor of history at the University of Texas at Austin.