“Science Is Real.” If you’ve driven or walked through an affluent liberal neighborhood in the last couple of years, you’ve likely seen this simple, three-word declaration on one of those “In This House, We Believe:” signs, alongside “Black Lives Matter” and “No Human Is Illegal.” They, it is implied—those mouth-breathing conservatives—don’t believe in science.

For a long time, I was in this camp. It baffled me that conservatives could be so cut off from reality. Why couldn’t they understand that the earth is billions of years old, that there are mountains of evidence supporting the theory of evolution, that climate change is real and human-caused? Why did they cling so tightly to their ignorance?

But in the course of researching and writing The Quick Fix: Why Fad Psychology Can’t Cure Our Social Ills, I’ve come to realize that conservatives don’t have a monopoly on magical thinking. The fact is that many liberals also believe a lot of silly things that aren’t backed up by research, and it isn’t an accident which ideas catch on. The successful ones, enthusiastically promoted by impressive-seeming experts on the TED Talks stage, tend to buttress the stories we want to be true.

The more profoundly a given half-baked psychological idea ties into our individual political identity, the more apt we are to throw science out the window. Take one of my favorite examples from the book, the Implicit Association Test (I.A.T.)—a brief computer exercise that, theoretically, can reveal one’s level of implicit (that is, unconscious) bias against Black people, or members of other marginalized groups, by comparing the reaction times to various stimuli, say, positive words paired with Black faces versus negative words paired with white faces.

Conservatives don’t have a monopoly on magical thinking. The fact is that many liberals believe a lot of silly things about psychology that aren’t backed up by research.

The I.A.T. has been a diversity-training and educational darling since it was introduced, in 1998. It was featured in a recent Today-show segment that absolutely gushed over its revolutionary potential to ameliorate racism—one of countless such credulous treatments to have been aired or published in the last two-plus decades. But it barely measures anything. Or at least that was the conclusion of a group of experts who wrote in 2015 that the statistical weaknesses of I.A.T.’s “render them problematic to use to classify persons as likely to engage in discrimination.” The experts also noted that “attempts to diagnostically use such measures for individuals risk undesirably high rates of erroneous classifications.” Among the authors of that paper? The founders of the test itself.

Yet the I.A.T. remains a cultural juggernaut, not because there is strong evidence behind it—there isn’t—but because it tells an appealing story: We’re all a little bit racist, but we can address that racism with a few minutes of computer-aided introspection. No matter that the test appears to be fundamentally flawed, and that no one has proved that implicit bias (as compared with explicit bias, structural racism, or other factors) is a major driver of racial disparities in the first place. Why trouble yourself with the complex reality of the Black-white wealth gap—a problem caused by myriad interconnected forces, some with historical roots tracing back centuries—when you can play around with cute computer games instead? There’s something in our nature that wants to be wowed, to have the critical parts of our brains turned off for a bit.

Over and over and over, I found examples of liberals all too willing to shirk the demands of rigorous science and evidence as soon as they came across something shiny and sexy. Those “In This House, We Believe:” signs should instead read: “Science Is Real (When It Conforms to Our Worldview).”

Jesse Singal’s The Quick Fix: Why Fad Psychology Can’t Cure Our Social Ills will be published on April 6 by Farrar, Straus & Giroux