Talk:Broken windows theory

Direct contradictory natural experiment evidence

MISDEMEANOR PROSECUTION, Amanda Y. Agan, et al. By using a natural experiment between DA departments that assign misdemeanor cases randomly among prosecutors, and looking at re-offense rates among prosecutors who are relatively more or less aggressive in pursuing non-violent misdemeanor cases, the authors find that every(non-violent, misdemeanor) case fewer recommended for prosecution reduces the number of future criminal complaints against individuals from that prosecutor's pool by 2.1. The most significant reduction for future reports are violent crimes.

That's right. More than one. This holds regardless of variance in conviction rates. The authors postulate that being within the criminal system, whether held in jail, in various hearings, or just the disruption of being formally accused of a crime on present and future work opportunities, all amount to social stressors that take people on the margins of society and remove most of their safety net and social resources, putting them into more desperate situations.

The effect is larger and clearer in, but not restricted to, first time defendants.

Overall, this gives the clear impression that Broken windows policing has the inverse effect ascribed, when examined at the prosecution level. ikanreed ^{🐐Bleat at me} 20:01, 30 March 2021 (UTC)