Little evidence of race or class bias in the criminal justice system

Hmmmm.  I am not surprised at the finding but I thought the headline was probably an overstatement.  Having read the article and the supporting links, it seems not.

From Mega-Study Finds That Minorities Don’t Receive Harsher Criminal Punishments, But That Academics Said So Anyway by Luke Rosiak.  The subheading is "Evidence for racial bias in the US criminal justice system has been consistently weak, and... scholarly narratives have too often ignored this."

As a meta-analysis, the study did not create a new dataset on criminal sentencing and race, but rather examined 51 studies conducted by others looking at the question since 2005. The numbers suggesting no or marginal racial bias in punishment were therefore collected by the existing studies, but those authors often claimed to have found racial bias in their writings, even when their numbers did not back it up.

“We express the concern that evidence for racial bias in the U.S. criminal justice system has been consistently weak, and that scholarly narratives have too often ignored this in favor of the systemic racism narrative,” Ferguson and Smith wrote.

[snip]

When combining the 51 studies, the numbers “did not reach evidentiary standards to support the hypothesis that race or class are predictive of criminal adjudication” when it comes to all crime types, violent crimes, or juvenile crimes. For drug crimes, “small disparities were found… suggesting that race/ethnicity is associated with between 1.6 to 1.8% of the variance in criminal adjudication” among blacks and Hispanics facing drug charges, versus whites–a percentage so small that policy discussions focused on it are unlikely to solve any problems.

“There was no evidence of a class effect, and Asians actually received more favorable treatment during criminal adjudication than Whites, albeit not at a level that met our evidentiary standard,” the paper said.

The absence of rigor in many studies was notable.

Among other signs of biased or poorly done science, race researchers often did not “pre-register” their studies by laying out in advance their hypothesis and what they would do to determine if it was true or not. That meant that if their initial methodology found no evidence of racism, they could simply slice and dice the data in different ways until they found some way in which they could claim vindication. (Ferguson and Smith’s meta-analysis was pre-registered.)

Such researchers also did not set out in advance what kind of numbers would be required to show that the criminal justice system is racist. “Put simply: it is helpful to know what data we’d expect to see if the theory is wrong and what the threshold for rejecting the theory might be. Without such clear guidelines, theories may persist endlessly despite having weak evidence,” Ferguson and Smith wrote.

Academia has wanted to find systemic oppression of minorities by the criminal justice system.  They were unable muster evidence that there was any bias in the system and so they simply chose to believe there was anyway.

This is the second study to find no bias.  A few more and perhaps the truth will out.