Junk Science

There is a lot of junk in psychology today, Jesse Singal tells us in this article in the WSJ. The title in tomorrow’s print edition is “The False Promise of Quick-Fix Psychology Why We Keep Falling For Fad Psychology.”

Some of this junk impacts criminal justice and law enforcement issues.

“Implicit bias” is often blamed for “disparities” in criminal justice outcomes. That is, Group A is X percent of people getting a given impact but only Y percent of the population. But are claims to measure implicit bias really valid?

But it’s the implicit association test, a brief computerized exercise anyone can complete on Harvard’s Project Implicit website, where the gap between hype and reality is most staggering. The test’s creators, Drs. Banaji and Greenwald, have long claimed that implicit bias may help to explain persistent racial disparities in American society, especially given the well-documented decline of explicit bias. This is a tidily inspiring story: There is a mental flaw many of us are carrying around without even realizing it, but here comes a new tool that can shine a light on the problem and help us to overcome it. That’s likely why the IAT narrative was so quickly accepted by corporations and institutions nationwide.

Experts have long understood, however, that modern racial inequalities are the result of complicated social systems that can reproduce themselves even in the absence of ongoing discrimination. Urban-planning decisions made in the 1960s can continue to resonate to this day, even if contemporary urban planners are unbiased. Intergenerational poverty is a notoriously difficult problem to overcome for reasons that go well beyond simple discrimination, whether implicit or explicit.

Implicit bias could certainly play a role in exacerbating these problems. Perhaps some white real-estate agents who view themselves as unbiased unconsciously favor white buyers over Black ones, contributing to housing discrimination. But no one has come close to proving that implicit bias is so significant that it deserves to dominate American racial-justice and diversity-training efforts.

Outsized estimates of implicit bias feed the myth that bias is the cause of problems that are mainly caused by deeper issues, more difficult to fix. People want to believe that there is a solution readily at hand. As H. L. Mencken said, “There is always a well-known solution to every human problem—neat, plausible, and wrong.” If we are misled into focusing on what is really a secondary cause, then we will not focus on seeking the difficult solutions for the primary cause. That is where junk social science causes deep and lasting harm.