BY STEVEN LUBET

This essay was originally published on SAGE’s online network for social scientists, Social Science Space, and is republished with their permission.

The right-wing Heritage Foundation has accused university Diversity, Equity, and Inclusion (DEI) offices of spreading anti-Semitism on campuses, but its recently issued report does not back up the claim. Although Heritage touts the study, titled Inclusion Delusion: The Antisemitism of Diversity, Equity, and Inclusion Staff at Universities, as revealing that “an overwhelming number of DEI hires are spewing antisemitic views about Israel on social media,” it is actually an instance of selective data presentation in support of a conservative cause.

The study’s authors claim that DEI offices not only fail to protect the interests of Jewish student, but instead “foment” anti-Semitism and “may be contributing to an increase in anti-Jewish hatred.” While certain DEI advocates have been inexcusably dismissive of Jewish students and faculty, sometimes veering into negative stereotypes and even overt hostility, the study’s methodology does not begin to establish its overheated assertion that “political activism of DEI staff may help explain the rising frequency of antisemitic incidents” on campuses.

The Heritage study began by identifying 2,933 DEI personnel at 65 major US universities, of whom 741 had publicly accessible Twitter accounts. Those accounts were then searched for tweets, retweets, and likes that evinced, as the report put it, “anti-Israel attitudes that are so out of proportion and imbalanced as to constitute antisemitism.”

The researchers ultimately found 605 tweets with extraordinarily harsh statements about Israel, a phenomenon that may indeed coincide with anti-Semitism. Many such tweets threw down a new twist on an ancient blood libel, as in “israel (sic) has a particular loathing for children. they target them with violence specifically and intentionally every single day.”

Another charged,

Y’all love to add the word liberal in front of the most evil things and it’s unhingedddd. Wtf is a liberal Zionist? What’s next? Liberal Nazi? Liberal colonizer? Liberal murderer? Liberal imperialist? Liberal fascist?

And a third threatened, “‘from the river to the sea’ means that we will decolonize every block and every grain of sand in palestine (sic).”

In contrast to all that condemnation, there were only 28 tweets favorable to Israel, some of them only mildly so. But the imbalance does not tell the whole story, because the most important information is strangely missing from the report.

In a study of anti-Semitism among DEI personal, the most significant figure would be the number or proportion of biased staffers, not the ratio of nasty tweets. And yet, as I have explained in The Forward, the Heritage report never tells us whether the 605 troubling tweets were generated by a single person tweeting almost 600 times, by 30 DEI officers tweeting 20 times apiece, or by 300 personnel tweeting only twice each.

Although the Heritage website claims to have documented an “overwhelming pattern” of anti—a conclusion repeated in media reports—there may well be no pattern at all. If it turns out that only a handful of DEI personnel were responsible for nearly all of the Israel bashing—which can hardly be ruled out, given the profound enmity toward Israel of some progressives—the modal number of angry tweets per staffer could easily be zero.

The authors of the Heritage Report—Jay Greene, a Harvard PhD, and James Paul, a doctoral fellow at the University of Arkansas—have not been forthcoming about the glaring omission in their report. They did not respond to my repeated requests for the number of unique accounts responsible for the anti-Israel tweets identified in their study. Nor have they posted their underlying data, which would have been routine for a published academic paper.

In a recent webinar, Dr. Greene was asked twice about whether the hostile tweets might have been concentrated in a few accounts. He was evasive both times, saying first that the question indicated a “failure to grasp sampling techniques,” comparing his study to a public opinion poll, and later claiming that the number of tweeters was not relevant, as opposed to the “percentage of tweets.”

Neither explanation holds up. Samples in public opinion polls are carefully randomized and controlled for demographics. The number of individuals in each category is not only relevant but essential.

It is obvious that Greene and Paul could have reported the actual number of hostile tweeters if it had indeed been “overwhelming.” Greene acknowledged at the webinar that they had collected the names and titles of every tweeter, as well as their university affiliations, so it would have taken nothing more than counting to provide the needed figure.

In social science, it is called “data dredging”—considered methodologically improper—when researchers publish only favorable results and suppress everything else. In litigation, an “adverse inference” can be drawn against a party who fails to produce material evidence that is under its control.

Absent a credible explanation for the conspicuous omission—and there is no explanation at all in the report—the most logical conclusion is that a full account of the findings would undermine or refute the Heritage Foundation’s accusations against DEI programs.

The report does make one interesting comparison, contrasting tweets about China and Israel. It turns out that China was named in only about one third as many tweets as Israel during the study’s time frame, which is not specified in the report. Most of those comments (133 out of 216) were positive toward China, notwithstanding the anti-democracy crackdown in Hong Kong and ongoing repression of Uighurs, among other depredations. This disparity was offered as “evidence of a double-standard with respect to the Jewish state.”

It is a huge leap, however, to infer antipathy toward Israel, let alone anti-Semitism, from silence on China. If the study occurred in the shadow of the 2020 presidential campaign, when Donald Trump was regularly inveighing against the “China virus,” such forbearance might indicate only an intention to avoid stoking then-rampant anti-Asian prejudice.

I have never hesitated to call out anti-Semitism, often masquerading as anti-Zionism, among progressives. It is equally important, however, to avoid unsupported accusations, especially when made in pursuit of unrelated political objectives.

The Heritage Foundation is bluntly engaged in an all-out campaign to defund university DEI programs, calling them a “Diversity-Industrial Complex” driven by “watered down Marxism,” as is exemplified by the Twitter study’s conclusion: “It is clear that DEI staff at universities actually function as political activists, articulating and enforcing a narrow and radical ideological agenda.”

Perhaps that passes for clarity in some quarters, but the anti-Semitism angle just muddies it up.

Steven Lubet is Williams Memorial Professor at the Northwestern University Pritzker School of Law.