A popular quiz on Harvard University’s website was designed in 1998 by psychologists to determine a person’s level of subconscious racism.  Although the test has been thoroughly debunked by researchers since its introduction, it has remained a fixture of progressive activism.

The “implicit bias” test, known formally as the “Implicit Association test,” is a test designed by psychologists Anthony Greenwald, Debbie McGhee, and Jordan Schwartz to determine a person’s subconscious racism. Mahzarin Banaji, who has served as the chair of the psychology department at Harvard University, also contributed to the project. The test has received much fanfare from progressives since its introduction in 1998.

The test, which can now be taken online, asks participants to respond to a series of photos of human faces. As words flash on the screen, participants are then asked to categorize them as “positive” or “negative.” The test analyzes the participant’s responses and tells them whether or not they carry subconscious biases. “Project Implicit,” the organization that is in charge of maintaining the online test, has partnered with Harvard University to expand its reach. The test is currently hosted on Harvard’s website.

However, there is a problem with the test. Researchers across the political spectrum have questioned its accuracy. One study conducted of the “implicit association test” revealed that it had a test-retest reliability of 0.60, meaning that individual participants would likely receive a different test result after taking the test for a second time.

VOX Senior Correspondent German Lopez published a column in March 2017 entitled “For years, this popular test measured anyone’s racial bias. But it might not work after all.” After Lopez first took the test, it told him that he carried no preference for white or black people. But Lopez took the test again to confirm its consistency. To Lopez’s surprise, he received a different result each additional time he took the test.

I took the IAT again a few days later. This time, I wasn’t so happy with my results: It turns out I had a slight automatic preference for white people. According to this, I was a little racist at the subconscious level — against black people.

Then I took the test again later on. This time, my results genuinely surprised me: It found once again that I had a slight automatic preference — only now it was in favor of black people. I was racist, but against white people, according to the test.

New York Magazine published a column detailing the trials and tribulations of the “implicit association” test.

“The IAT, it turns out, has serious issues on both the reliability and validity fronts, which is surprising given its popularity and the very exciting claims that have been made about its potential to address racism,” columnist Jesse Singal wrote. “That’s what the research says, at least, and it raises serious questions about how the IAT became such a social-science darling in the first place.”

Breitbart News reported in July that the researchers behind the “implicit association test” still defend the project despite concerns about its accuracy.