Profs use 'flawed' Harvard study to push 'implicit,' 'explicit' racism theory

Two Texas A&M University professors published a study claiming that “explicit” and “implicit” racial attitudes caused higher COVID-19 death rates among African-Americans, partially relying on data characterized by one conservative organization as “flawed.” 

Professors George Cunningham and Lisa Wigfall examined “the potential moderating effects of explicit racial attitudes and implicit racial attitudes on the relationship between percent of Black county residents and COVID-19 cases and deaths.” 

They found that the percentage of African-American residents was positively associated with deaths from the disease, then pinned explicit and implicit racism as an important correlating factor.

Working from findings that racial minorities have been disproportionately affected by the virus, the professors sought to consider “the role of explicit and implicit racial attitudes.”

[RELATED: American Medical Association declares racism a ‘public health threat’]

They defined explicit racial attitudes as “those of which people are aware and that they consciously, deliberately maintain” and implicit racial attitudes as reflecting “automatic responses that manifest when an external stimulus corresponds with an individual’s association set that link the stimulus and particular characteristics.”

The researchers then used county-level data for COVID-19 cases and deaths, the number of black residents, and racial attitudes. They measured both implicit and explicit racism using Harvard University’s Project Implicit, which collects “bias and prejudice data” from people around the world.

A 2017 Heritage Foundation report from statistician Althea Nagai explained that the project’s “Implicit Association Test,” which the professors utilized in their study, is “flawed” in spite of extensive mainstream media coverage. 

In fact, “there are many scientific critics of this test, and it is far from settled science.”

Nagai points out that an individual’s test score may vary from trial to trial, pointing to its lack of reliability. Room for error is introduced when test data is measured in milliseconds. There are a “wide variety of other explanations” for differences when measured in milliseconds.

Additionally, “the IAT could tap into a fear of being called a racist instead of being an unconscious racist” for some test takers.

[RELATED: Pitt med students add vow against ‘systemic racism’ to Hippocratic Oath]

Nevertheless, Cunningham and Wigfall concluded that “aggregate measures of explicit racial attitudes and implicit racial bias provided additional explanatory evidence.” Specifically, “the relationship between the percent of Black county residents and COVID-19 cases was stronger when explicit racial attitudes and implicit racial attitudes, respectively, were high relative to when they were low.”

The professors’ discussions of limitations did not include concerns about Project Implicit, as raised by the Heritage Foundation.

Campus Reform reached out to Cunningham and Wigfall for comment and will update this article accordingly.

Follow the author of this article on Twitter: @BenZeisloft