A team analyzing data from almost all PhD graduates in the US over the past 30 years find that underrepresented groups are more likely to publish innovative research — and yet, they are less likely to earn academic positions and their innovations are more often overlooked.

The diversity paradox has been discussed beforehand in different contexts. It’s expressed in slightly different ways, but the main idea is this: diversity breeds innovation, yet the groups that bring diversity tend to have less successful careers.

A new study wanted to see whether the diversity paradox also holds for scientists — spoiler alert, it does.

A team led by Bas Hofstra at Stanford University analyzed 1.2 million US doctoral recipients, following their careers into publishing and faculty positions. They used machine learning to predict people’s gender and race based on their name. This wasn’t exactly perfect and was particularly challenging for nonbinary gender, but overall, researchers expect the accuracy to be extremely high (based on a record of names, 95% of the names in the study were distinctive).

The participants were split into three racial groups: white, Asian, and underrepresented (which gathered minorities such as Hispanics, African Americans, Native Americans, and any other category not in the first three).

Researchers tried to quantify the innovation, researchers looked at 3 things:

• general novelty (the number of new ideas brought in);
• impactful novelty (how many mentions, not citations, the papers received in the future);
• distal novelty (linking existing ideas and combining them in new ways).

Researchers found those novel contributions by gender and racial minorities are less likely to be mentioned, even when they are equally impactful. Furthermore, equally impactful contributions of gender and racial minorities are less likely to result in successful scientific careers than for majority groups.

“These results suggest there may be unwarranted reproduction of stratification in academic careers that discounts diversity’s role in innovation and partly explains the underrepresentation of some groups in academia,” the researchers note.

The authors also report that minorities produced more distal innovation than their peers, connecting ideas in new ways — which may explain a part of this effect, as these new ways are harder to accept and understand for others in science. These are the big breakthroughs, the frameworks that pull ideas from different fields together and create new theories. It is concerning that distal novelty in general, is inversely related to impactful novelty, as the study found.

In a sense, this means that for scientists it can be dangerous to be truly innovative, and minorities are less afraid — and paradoxically, they’re more likely to be punished for it.

Overall, this is signaling that minorities play an important and underappreciated role in science, the researchers conclude.

“These results suggest that the scientific careers of underrepresented groups end prematurely despite their crucial role in generating novel conceptual discoveries and innovation. Which trailblazers has science missed out on as a consequence?”

The study was published in PNAS.