IAT as an example of zombie ideological thinking

From Anomalies in implicit attitudes research by Edouard Machery, Center for Philosophy of Science, University of Pittsburgh.  

The Conclusion:

We do not know what indirect measures measure; indirect measures are unreliable at the individual level, and people’s scores vary from occasion to occasion; indirect measures predict behavior poorly, and we do not know in which contexts they could be more predictive; in any case, the hope of measuring broad traits is not fulfilled by the development of indirect measures; and there is still no reason to believe that they measure anything that makes a causal difference. These issues would not be too concerning for a budding science; they are anomalies for a thirty-year old research tradition that has been extremely successful at selling itself to policy makers and the public at large. So, should social psychologists pack up and move to other research topics or should they stubbornly try to address the anomalies pointed out in this article? It is unwise to predict the future of science, and the issues presented here could well be resolved by the many psychologists working on indirect measures, but it would also be unwise to dismiss them as mere challenges to be addressed in the course of normal science.

A little more broadly:

Since the 1990s, social psychologists have developed several unobtrusive or indirect measures of attitudes, the most famous of which is the Implicit Association Test. Psychologists, and following them philosophers and policy makers, often take these measures to tap into a new kind of attitudes, distinct from the attitudes that older direct measures assess: implicit attitudes. Implicit attitudes have been blamed for many of the enduring social ills, and they justify costly training programs in the corporate world, universities, and police departments.

In recent years, however, there have been increasing concerns about the quality of indirect measurements. This article reviews the most important issues related to that question:

There is little evidence that direct and indirect measures measure distinct things.

Indirect measures are unreliable: Your score today does not predict well your score tomorrow.

Indirect measures predict behavior poorly.

There is no causal evidence that whatever it is indirect measures tap into affects behavior.

These issues have been around for now decades, but they have barely been addressed despite their basic nature. But then why do so many believe in implicit attitudes at all, as a sui generis kind of attitudes?

One more zombie ideological indulgence.  All the evidence points away from Implicit Attitude Tests as being in any way meaningful but policy makers keep relying on them to shape their policies.