What if college admissions were based on random lotteries? Fewer black, low-income, male students would be accepted, new study finds

register here for the daily 74’s newsletter. Donate here to support independent journalism of the 74.

Over 50 years ago, public intellectuals played with a radical idea: What if colleges used random lotteries to admit students?

It is a notion that first emerged in the 1960s, when middle-class families sought greater access to higher education and the paths to prestigious schools became increasingly competitive. . Since then, the lotteries have been praised like a possible solution when the admissions process came under scrutiny – most recently, after the “Varsity Blues” corruption scandal and a federal lawsuit alleging Harvard discriminates against Asian American students.


Pondiscio: Everything in education can be played by the privileged. Eliminating merit in admissions will only boost their advantage

But new research suggests the shift to a lottery-based system could create significant new challenges for schools seeking to promote racial, ethnic, and socio-economic diversity. The study, published today in the journal of the American Educational Research Association, finds that low-income Hispanic and African American students would be less likely to be accepted into colleges in a multitude of lottery scenarios; in some cases even the percentage of male students admitted would be sharply reduced.

Dominique Baker, co-author of the study and professor of educational policy at Southern Methodist University, said she was drawn to studying the issue because of the wide range of advocates from different sides of the political spectrum. American who supported the randomization of university admissions. for various reasons.

Stay informed.
Invest in independent journalism. And help The 74 make an impact.

Donate now and help us reach our NewsMatch goal.

Former assistant dean of admissions at the University of Virginia, Baker noted that lotteries are too often seen as a silver bullet to what plagues higher education. The mechanisms for admission to college “already have an incredible amount of luck,” she said. “We just don’t see it because it’s an opaque process. “

Baker and his co-author, University of Michigan professor Michael Bastedo, used two US Department of Education data sets to simulate what college admissions numbers would look like in various lottery setups. One, the Longitudinal Study of Education, followed a group of high school students for several years starting in 2002; the other, the High School Longitudinal Study, followed a different group of first year high school students from 2009. Both datasets included academic and demographic information – including grade point average as well as results. tests – for large samples of students as they progress from high school to college.

They then identified all the students in each sample who were above several minimum thresholds for academic performance: those who were at or above the 25th and 50th percentiles for the SAT, GPA, and weighted GPA scores. These, they explained, would be the eligible students to participate in a national notional admissions lottery.

Then, they generated 1,000 random simulations to project which students in the sample might be accepted into at least moderately selective colleges, using GPA minimum eligibility thresholds, SAT scores, and a combination of the two. Finally, they compared these results to the sample of students who attend very or moderately selective institutions under the current system.

Their conclusions were striking. Black, Hispanic, and low-income applicants generally declined as a proportion of students admitted in most lottery scenarios. Consistently, lotteries that have established a minimum GPA threshold for eligibility have resulted in significantly fewer male students accepted than what prevails today – typically between 36 and 40 percent. When lotteries used a combined threshold of both SAT and weighted GPA scores, the eligible pool of African American and Latino students fell below the proportions currently enrolled in selective colleges.

Baker said the trends seen in lottery simulations indicated large disparities in the performance of students from different demographic groups on different eligibility metrics. Black and Latino students tend to score lower than White and Asian American students on college entrance exams like the ACT and SAT, while girls generally score higher GPAs. higher than boys.

“None of the thresholds we looked at showed evidence of creating a more equitable class,” Baker said. “And you can see differences because different inequalities are built into surrogacy compared to test scores – that’s why you see the difference for men with surrogacy that you don’t see for test scores.”

In terms of racial groups, virtually every lottery lineup has produced higher percentages of white and Asian American students than under current admissions procedures. In some lotteries, the proportions of low-income students and students of color fall below 2 percent of the admitted class. This is due in part to the random fluctuations in selection that accompany annual lotteries; over sufficient years and draws, the characteristics of admitted students should resemble those of the total pool of eligible participants. But that might not be the case in any given year, Baker observed.

“Many years of admitted lottery classes – 50 years, 100 years – would look a bit like the general population,” she said. “But an individual lottery is not guaranteed to be demographically similar to the larger population.”

It should be noted that supporters cited many reasons why a move to lotteries would be superior to the current system, many (alleviating pressure on potential applicants or removing low-level corruption from admissions exchanging donations. against letters of acceptance) unrelated to fairness concerns. For those who complain that elite colleges have quietly capped Asian American parts of their student bodies, the increased number of those students admitted under the study’s lottery screenings could even serve as an endorsement.


What Harvard Affirmative Action Victory Means for Students Facing “Endemic Inequality” in K-12 Schools

Baker said she and Bastedo were “shocked” at the magnitude “of some of the changes detected in the lottery simulations, but said the results generally reflected what existing evidence has shown regarding disparities in performance. in high school grades and university entrance exams.

“It’s well anchored in the research literature we have. If you choose different types of thresholds, whatever demographic trends we see for this metric, we should expect to see them in the lottery as well. “


Subscribe to the 74 newsletter

Send letter to editor

Source link

Comments are closed.