Exterior of Madison West High School
Madison West High School, in Madison, Wisc., has hosted discussions in which students and parents were segregated into groups based on their race.

In schooling, proponents of even suspect pedagogies and practices tend to insist that their preferred approach is “evidence-based.” This seems to be the case, yet again, in the debates swirling around “anti-racist” education. I’ve encountered many claims I find unconvincing, especially when they’re advanced by impassioned advocates who don’t seem to have thought all that much about what constitutes credible evidence. (For more on how I think about evidence, check out my Educational Leadership piece from earlier in the spring.)

This has been particularly noticeable when it comes to racial “affinity spaces” and the whole notion that public school systems should be comfortable separating students by race and ethnicity in order to address sensitive issues. Supporters routinely assert that there’s evidence to justify this practice, despite a startling lack of research actually supporting it (more on that in a moment).

For instance, Madison West High School, in Madison, Wisc., has hosted discussions in which students and parents were segregated into groups based on their race. This spring, after one such exercise, the local NBC affiliate published “Experts explain effects of affinity groups,” in which a district spokesperson, the high school principal, and a University of Wisconsin sociology professor all echoed the district’s contention that this is “a well established method.”

In Massachusetts, this spring, Wellesley public schools hosted a “Healing Space for Asian and Asian American students and others in the BIPOC (Black, Indigenous, People of Color) community.” The district’s email explained, “*Note: This is a safe space for our Asian/Asian American and Students of Color, *not* for students who identify only as White.” When parents expressed concerns, administrators said, “It’s important to note that affinity spaces are not discriminatory” and asserted that “hosting affinity spaces is part of a long-term, evidence-based district strategy.”

So, what’s the evidence behind this “well established,” “evidence-based” strategy?

As it turns out, there’s not much. A comprehensive search of the academic databases ProQuest and Google Scholar returns just five articles purporting to examine the benefits of “racial affinity” spaces in K–12 schooling (related articles focus on things such as video games or community centers). This is an astonishingly tiny figure, especially when compared to the thousands of studies on teacher evaluation, school choice, or math instruction—topics where the evidence is nonetheless regarded as hotly contested.

And it’s a stretch to suggest that the handful of research that does exist constitutes “evidence.” The only article that even claims to review the research literature was a 2012 “online submission” to the Education Resources Information Center by Lindsay L. Schrader and K. C. Holder, in which they make the case for “formal affinity groups” and insist that “these groups have shown again and again to be a powerful investment for students of color.” Yet, of the 17 citations, just one supposedly justifies this strong claim. And that citation, “Batiste, G. (2006),” turns out not to be a study or report at all, but a data-free National Association of Independent Schools (NAIS) PowerPoint presentation sharing the “NAIS perspective” on affinity spaces.

The rest of the results are more akin to author essays and testimonials than any kind of systematic attempt to gauge or assess the impact of such spaces. A 2016 dissertation by Cindy P. Chun employs a “Dynamic Narrative Approach” to interview 10 “diversity practitioners” at independent schools in order to identify “best practices of affinity groups,” though Chun blithely explains that it’s not her intention to provide evidence regarding the efficacy of these recommended practices. A 2017 “ethnographic case study” by Farima Pour-Khorshid, published in Teaching Education, looks at how a “racial affinity group became an important space for learning and healing for its [dozen] members” and then broadly asserts, based on the author’s judgement rather than a measurable, replicable study, that racial affinity spaces promote “personal, political, relational, and pedagogical growth.” And, in a 2019 article, Ryan Oto and Anita Chikkatur study an affinity group that a teacher created for a single class at a private high school. While they provide no systematic data on academic, social, or other outcomes, they opine that the group yielded a “curriculum that was culturally affirming for students of color by de-centering whiteness.” While that may well be true, they did not offer any verifiable or falsifiable data showing that this had any of the claimed benefits for kids.

Indeed, the fifth study, Ryan Kimmet’s 2021 University of Pennsylvania dissertation on “student perceptions of white racial affinity groups,” raised some red flags. Kimmet reported that, while students found value in discussing issues of race, they made clear “that the spaces created for affinity groups were not, in fact, safe spaces.” He observed, “There were strongly negative social repercussions for making comments that the majority of students viewed as out of line with the majority’s way of thinking.”

In short, the supposed evidence demonstrating the effectiveness of racial affinity spaces is nowhere to be found. Interviewing people who think like you do or visiting a school to write about a program you like is just fine, and can offer much of value. But it’s ludicrous to suggest that the existence of such “studies” make an approach “well established” or “evidence-based.” If I visit a program I like or interview 10 people who agree with me, and then write it up and publish it in an academic outlet, it doesn’t make my views true. And it doesn’t make them evidence. They’re simply my opinions, published in a fancier place.

At the end of the day, suggesting that racial affinity spaces are an evidence-based practice is simply irresponsible—and does grave violence to the meaning of “evidence-based.” Many have justifiably come to regard such claims with suspicion, precisely because this language is so often used to justify suspect practices. If the educators, advocates, and researchers who champion actual evidence-based educational practices would like skeptical lawmakers, parents, and pundits to look upon their efforts with more confidence, they’d do well to call out partisans when they dress preferred practices in scientific garb.

Frederick Hess is director of education policy studies at the American Enterprise Institute and an executive editor of Education Next.

This post originally appeared on Rick Hess Straight Up.

Last updated September 8, 2021