Judge disparities may really reflect prosecutor behavior rather than judicial bias

Evidence

There are still other, even more subtle, complications in the analysis of judicial decisions that may confound the estimated impact of race, ethnicity, and gender on judicial outcomes. For example, assume that prosecutors are more aggressive inside the courtroom during trials involving minority defendants, and they are milder and more accommodating against White defendants. If this is the case, the conviction and sentencing decisions of a judge may resemble favoritism toward White defendants, but this would be misleading. The disparate outcomes would be reflection of prosecutors’ actions in the courtroom. This particular potential confounding can be controlled for by including in the model (Equation 1) indicator variables for individual prosecutors who were involved in each case (prosecutor fixed effects), although in most publicly available data sets such information about prosecutors is not recorded. — Joseph Avery, Jim Sidanius, Joel Cooper, Gregory Davis, Amanda N., Bias in the Law, 739

For example, assume that prosecutors are more aggressive inside the courtroom during trials involving minority defendants, and they are milder and more accommodating against White defendants. If this is the case, the conviction and sentencing decisions of a judge may resemble favoritism toward White defendants, but this would be misleading. The disparate outcomes would be reflection of prosecutors’ actions in the courtroom. This particular potential confounding can be controlled for by including in the model (Equation 1) indicator variables for individual prosecutors who were involved in each case (prosecutor fixed effects), although in most publicly available data sets such information about prosecutors is not recorded. — Joseph Avery, Jim Sidanius, Joel Cooper, Gregory Davis, Amanda N., Bias in the Law, 738

Categorization

• DEFINE: Introduces ‘prosecutor fixed effects’ as a methodological control.
• QUESTION: Worth chasing: how many bias studies actually correct for prosecutor fixed effects, given the data limitation noted here?
• XREF: The measured-vs-actual-bias divergence echoes the reader’s core question about whether measured bias reflects the person measured or upstream institutional actors. — connects to: Judge attractiveness bias measurably reduces real-world fines and bail amounts — Anthony G. Greenwald and Mahzarin R. Banaji, Implicit Social Cognition: Attitudes,…, 92
• corroborated: 3 independent passages support this claim — Joseph Avery, Jim Sidanius, Joel Cooper, Gregory Davis, Amanda N., Bias in the Law: A Definitive Look at…, 739; Joseph Avery, Jim Sidanius, Joel Cooper, Gregory Davis, Amanda N., Bias in the Law: A Definitive Look at…, 738