How do we know the extent of racial bias in policing? New research suggests administrative records often mask it. Police officers, when choosing whom to investigate, may show racial discrimination that isn't captured by standard data analysis methods. This gap means past estimates could be seriously off.

* The Problem: Studies relying on administrative records miss cases where police observe but don’t act against civilians.

* New Findings: Racial disparities in investigation rates themselves can hide deeper biases, even among those investigated.

* Traditional Limits: Current methods for estimating bias using existing data are flawed and often severely underestimate the true level of discrimination or completely miss it.

Using causal mediation analysis with principal stratification, researchers show exactly how these standard approaches lead to bias. A new correction method helps fix this issue, giving better estimates even when assumptions aren't perfect. They also provide sharp bounds for understanding race effects on police behavior.

This work is crucial because accurate measurement of racial discrimination in policing matters greatly to scholars and policymakers seeking solutions.