Professional School Grads From Diverse Classes Get Higher…
Text settings
Story text
SmallStandardLargeStandardWideStandardOrange
* Subscribers only
Learn more
Even before the Trump administration went to war against DEI and attempts to address historical discrimination, diversity efforts in the US were controversial. A pivotal moment came in 2023, when the Supreme Court ruled that race-based affirmative action programs violated the Constitution. The decision partly rested on universities’ inability to clearly measure the benefits of diverse student bodies and the lack of defined standards to determine when equity had been achieved and such programs should end.
A new paper highlights the uncertainty. “Learning theory argues that racial diversity promotes student learning, which should increase salaries,” its authors write. “However, well-documented racial wage discrimination indicates that higher racial diversity should decrease salaries.”
But the authors—Debanjan Mitra, Peter Golder, and Mariya Topchy—have developed a metric suggesting that graduates benefit financially if they graduate with a diverse peer group. The researchers argue that this evidence should be sufficient to prompt courts to reconsider earlier rulings.
Diversity and salaries
Doing this sort of research is challenging, largely because there are no clear metrics. Outcomes also vary widely based on factors school quality, baseline diversity, and the economic conditions at graduation, which can overshadow potential benefits. So while some research has suggested advantages to more diverse cohorts, the evidence remains limited.
The new paper responds to these challenges by both narrowing and expanding its focus. It narrows the analysis to business and law schools, tracking only a single outcome: starting salary. At the same time, the researchers broaden the research, drawing on decades of data from nearly 350 schools, including nearly 3,000 business school grads and even more from law schools, spanning over 20 years of graduating classes.
The data doesn’t include every graduate of these programs, typically covering about 75 percent of each class. But Mitra, Golder, and Topchy assess diversity by analyzing the available student data and examining the overall diversity of the school’s admitted classes.
The authors took a deceptively simple approach, examining the correlation between racial diversity in a school’s cohort and graduates’ starting salaries. In business schools, high-diversity cohorts earned starting salaries that were a standard deviation or more above the median 966 times out of 3,964 cohorts. For low-diversity cohorts, that number was just 534. For relatively low starting salaries, high-diversity cohorts showed up 531 times, while low-diversity ones appeared 933 times, largely reversing the numbers.
The pattern held for law schools. High-diversity groups saw high salaries in 1,128 of 3,386 opportunities, compared with 490 for low-diversity cohorts. The same was true for both types of graduate programs when the authors measured diversity using data for the entire entering class rather than only the students being analyzed.
Plenty of confounding factors could still explain the results, so Mitra, Golder, and Topchy implemented a few controls. More than a few, actually—it’s not often you see the word “thirteenth” in a paper’s list of potential caveats.
The alternative explanations
The authors provided a long list of possible confounders. For example, they removed the top and bottom 5 percent of starting salaries to rule out outliers (which didn’t affect the results). They gave each school a separate time trend to see if there were local economic factors, but the results stayed largely the same. They also tested various measures of diversity, examined different diversity thresholds, and controlled for university prestige, size, and urban settings. None of those changed the trends.
They also plotted each school’s cohorts individually and found that the diversity/salary correlation was positive and significant 40 times and negative and significant 19 times. For law schools, the numbers were 64 positive and 28 negative. Switching from median starting salary to mean starting salary had no effect.
Only one of the 13 added any nuance to the big picture. In that case, the trend was stronger for students entering the public sector or joining large companies. Otherwise, there was little evidence of factors that might be throwing off the results.
Is there a way these two factors are linked that isn’t causal and wasn’t considered by Mitra, Golder, and Topchy? Possibly. But the effect appears robust and seems to show up no matter how the analysis is done.
The big unanswered question is how this effect arises. It’s unly to be driven simply by employers’ preference for schools with historically diverse student bodies or by the fact that prestigious schools tend to have more diverse classes, as both are accounted for in the analysis. It’s entirely possible that people who experience more diverse perspectives present as more impressive during the hiring process, but it’s not clear how.
Advocating change
It’s obvious that the study was motivated by the Supreme Court ruling blocking affirmative action. The authors note that the court’s decision rested on three points: that the benefits of diversity were difficult to quantify, that they weren’t directly connected to the goals of education institutions, and that there was no clear standard for determining when an affirmative action program had accomplished enough to be ended.
The authors argue that they’ve cleared those objections by providing a measurable goal that serves as a valid endpoint for professional education. They also note that the measurement itself provides an indication of when diversity is sufficient that all entering classes benefit from it. On that basis, they argue that the ruling ending these programs should be reconsidered.
“Courts should revisit affirmative action on the basis of our evidence and also because previous courts, business executives, military leaders and university leaders have consistently affirmed its benefits, which only the most recent Supreme Court has rejected,” they wrote.
While the current Supreme Court has proven willing to revisit many past precedents, it seems unly it would do so for one of its own decisions.
Nature, 2026. DOI: 10.1038/s41586-026-10425-7 (About DOIs).
John Timmer Senior Science Editor
John is Ars Technica’s science editor. He has a Bachelor of Arts in Biochemistry from Columbia University, and a Ph.D. in Molecular and Cell Biology from the University of California, Berkeley. When physically separated from his keyboard, he tends to seek out a bicycle, or a scenic location for communing with his hiking boots.
Sumber Artikel:
Arstechnica.com