An April working paper finding that the economic gains associated with attending historically black colleges and universities (HBCUs) in comparison to traditionally white institutions have shifted dramatically since the 1970s -- and not in the HBCUs’ favor -- came under heavy scrutiny Monday during a session at the National Historically Black Colleges and Universities Week Conference in Washington.

The study, conducted by Harvard University’s Roland G. Fryer and the Massachusetts Institute of Technology’s Michael Greenstone, found that graduates of HBCUs in the 1970s benefited from a 10 to 12 percent wage gain relative to those who attended traditionally white institutions. However, by the 1990s, and despite gains on measures of pre-college academic preparedness among students at black colleges, HBCU graduates had a 12 to 14 percent lower wage on average than graduates of traditionally white colleges -- accounting for a swing of roughly 20 percent.

The study quickly became controversial among HBCU leaders, and at Monday’s session, organized by the umbrella group for HBCUs, the National Association for Equal Opportunity in Higher Education, panelists raised questions about the methodology and the assumptions behind it – with David Swinton, an economist and the president of South Carolina's Benedict College, going so far as to call one of the study's major premises “racist.”

The conversation began with Lezli Baskerville, president and CEO of NAFEO, citing a number of statistics showcasing the successes of HBCUs, which disproportionately educate students from low-income backgrounds.

Among them: While HBCUs represent just 3 percent of colleges nationally, they enroll 18 percent of African Americans in higher education and graduate 30 percent of those who persist to graduation. They graduate 40 percent of African Americans who obtain degrees in the STEM (science, technology, engineering and mathematics) fields, 50 percent of those who go on to become professors and 60 percent of those who major in engineering. Valerie Rawlston Wilson, the senior resident scholar at the National Urban League Policy Institute, followed that data with some of her own, which show that after controlling for differences in institutional characteristics, family backgrounds and academic performance, HBCUs “are at least as effective in graduating African-American students comparable to traditionally white institutions, and with fewer resources.”

Following up on that, Greenstone, the study's co-author, presented the findings, which President Swinton then rebutted. Swinton cited a number of concerns about the methodology, warning against the dangers of “data mining” and indicating that it’s unclear whether the longitudinal datasets the researchers relied on were appropriately representative of HBCU graduates.

To applause and murmurs of “absolutely,” he suggested that large numbers of HBCU graduates going into the nonprofit world could help explain the wage gap, and added that prejudices on the part of employers who prefer to hire graduates from traditionally white institutions could also play a role (though he noted, too, the substantial changes in climate since the '70s and the possibility that the function of HBCUs have shifted accordingly).

Swinton described himself as unconvinced by the researchers’ attempts to control for selection bias by controlling for such factors as family income, the educational background of the students' parents, standardized test scores and grade point averages, and decried the lack of causality demonstrated by the researchers: “The real failure is that the authors have not tested a plausible theory as to why … attending an HBCU or a TWI [traditionally white institution] should result in different outcomes,” Swinton said.

Swinton also took issue with the designations of “HBCUs” and “TWIs" as categories for the study, given the heterogeneity of the colleges that fall into those two groups. The use of the constructs despite the differences and, therefore, the differential outcomes that would most likely result, was, he said, "racist."

"Rather than study how to get rid of HBCUs, society would be strengthened by focusing on improving these HBCUs," Swinton said.

In response, Greenstone said that he too was disappointed that he and Fryer couldn’t conduct an experiment to demonstrate causality, but described that concern as a “red herring” given the realities of social science research. He took offense at the remark that the work was based on a “racist” construct, calling Swinton’s assertion “unfair." He protested that he doesn’t expect the study to be the final word on the subject, but instead hopes that the findings will inspire further research. “How can information be bad?” Greenstone asked.

“Questions loom and we’re not going to resolve them in this room,” NAFEO's Baskerville said to the crowd more than a half hour after the session was scheduled to end. Warning of the danger of “misinformation” and “information that can be skewed” to empower what she described as an ongoing assault on HBCUs (in response to Greenstone’s rhetorical question about the value of information), she said that more studies providing different perspectives on the issue would be on the way shortly.