A couple of months ago, an important paper by Althea Nagai — research fellow at the Center for Equal Opportunity — was published by the Heritage Foundation and released at a special event there. At the event the paper, “The Implicit Association Test: Flawed Science Tricks Americans into Believing They Are Unconscious Racists,” were discussed by Dr. Nagai, myself, and Heather Mac Donald of the Manhattan Institute. You can listen to the panel here and read the paper here. The following week, the Heritage Foundation’s Hans von Spakovsky and I wrote a column on the paper for National Review Online, which follows:
Are the vast majority of us infested with implicit racial bias? That is the claim made by people such as Eric Holder, Barack Obama, and a host of media pundits and academics. But this is a farfetched claim not supported by good scientific evidence. It is a myth that has been pushed for years by those who profit politically from dividing Americans and promoting racial preferences.
Those pushing this flimsy claim have relied, in part, on the Implicit Association Test (IAT), created by some University of Washington psychologists in 1998. As outlined in a new Heritage Foundation study by Althea Nagai, a research fellow at the Center for Equal Opportunity, it measures the speed with which you associate “pleasant or unpleasant words such as ‘joy,’ ‘crime,’ or ‘work’ with categories [such as] ‘black’ or ‘white,’ ‘male’ or ‘female.’” The differences are measured in milliseconds, and they supposedly show whether you are biased or not.
(Full disclosure: One of us is a Heritage fellow; the other is president and general counsel of the Center for Equal Opportunity.)
As Nagai spells out, the IAT suffers from serious scientific weaknesses. To start with, the test is not reliable, meaning someone who takes it multiple times is likely to get widely varying scores. This was demonstrated by one of our colleagues, who took the test twice in the same afternoon. The first time it scored her as having a “slight preference” for blacks, the second as having a “slight preference” for whites. Yet under this test, “one-tenth of a second can lead to highly charged accusations of racism.”
The IAT’s supporters also fail to take into account that the test’s headline-grabbing result — that whites take slightly longer to pair pleasant words with black faces, relative to white faces — could stem from a variety of factors other than unconscious racism, including things as mundane as differences in cognitive quickness and flexibility.
Other factors that could drive the results range from the test-taker’s familiarity with certain pairings of words and pictures to what is called a “figure/ground” effect, where white faces and names “are the familiar and fall into the background, while black faces and names are more distinctive, thus becoming more prominent.” An IAT testing people’s responses to familiar versus nonsense words produces the same result that the black-versus-white IAT does. Fear of being classified as a racist could also affect the results, as can repeatedly taking the test, knowing its purpose, or just being aware of stereotypes (whether or not you hold them).
Further, the IAT has not been shown to “significantly predict discriminatory behavior”; the results do not “closely relate to any other measures of discrimination,” and the proportion of false positives — in effect, false accusations of racism — is “substantial,” with estimates “ranging from 60 percent to 90 percent.” Even the test’s creators have admitted “there is substantial risk for both falsely identifying people as eventual discriminators and failing to identify people who will discriminate.”
So why is this dubious test being used and promoted in the media, academia, the government, and the corporate world? As Nagai explains, major media outlets such as the New York Times and the Washington Post have positively profiled the IAT, as have CNN and PBS. She says that even medical schools (which should know better) such as those at Stanford, Ohio State, and Johns Hopkins are encouraging their faculty and students to take the IAT to root out their unconscious racism, claiming falsely that the test is “both reliable and valid.” Duke University has incorporated it into a “second-year medical school course.” And during the Obama administration, Sally Yates, the acting attorney general, ordered 28,000 employees of the U.S. Justice Department to undergo training to “recognize and address implicit bias.” The need for this training was allegedly “based on the latest social science research.”
The IAT and the claim of unconscious bias are being pushed to justify racial preferences in employment, college admissions, government contracting, and numerous other areas of our economy and culture. Obviously, the type of quotas and racial discrimination that progressives want will in their minds be justified if they convince America that unconscious racism exists everywhere. Nagai quotes activists who say that race and gender quotas and preferences will become unnecessary only “when the nation’s implicit bias against those social categories goes to zero.”
The bottom line here, according to Nagai, is that the IAT “fails to prove that we are a nation of unconscious racists.” This claim is defamatory, and it is dangerous. As Nagai says, by “emphasizing the finding that unconscious racism is spread throughout society . . . the race IAT research may sadly serve to make race relations worse.” Public cynicism is fed by “false accusations of racism” and “true instances of racism lose their salience.”
She couldn’t be more right.
