The criminal justice system is overburdened and expensive. What if we could harness advances in social science and math to predict which criminals are most likely to re-offend? What if we had a better way to sentence criminals efficiently and appropriately, for both criminals and society as a whole?

That’s the idea behind risk assessment algorithms like COMPAS. And while the theory is excellent, we’ve hit a few stumbling blocks with accuracy and fairness. The data collection includes questions about an offender’s education, work history, family, friends, and attitudes toward society. We know that these elements correlate with anti-social behavior, so why can’t a complex algorithm using 137 different data points give us an accurate picture of who’s most dangerous?

The problem might be that it’s actually too complex -- which is why random groups of internet volunteers yield almost identical predictive results when given only a few simple pieces of information. Researchers have also concluded that a handful of basic questions are as predictive as the black box algorithm that made the Supreme Court shrug.

Is there a way to fine-tune these algorithms to be better than collective human judgment? Can math help to safeguard fairness in the sentencing process and improve outcomes in criminal justice? And if we did develop an accurate math-based model to predict recidivism, how ethical is it to blame current criminals for potential future crimes?

Can human behavior become an equation?

*** ADDITIONAL READING ***

Sample COMPAS Risk Assessment: [ Ссылка ]

COMPAS-R Updated Risk Assessment: [ Ссылка ]

“The accuracy, fairness, and limits of predicting recidivism.” Julia Dressel. [ Ссылка ]

“Understanding risk assessment instruments in criminal justice,” Brookings Institution: [ Ссылка ]

“Machine Bias,” Julia Angwin, Jeff Larson, Surya Mattu and Lauren Kirchner, ProPublica: [ Ссылка ]

“The limits of human predictions of recidivism,” Lin, Jung, Goel and Skeem: [ Ссылка ]

“Even Imperfect Algorithms Can Improve the Criminal Justice System,” New York Times: [ Ссылка ]

Equivant’s response to criticism: [ Ссылка ]

“A Popular Algorithm Is No Better at Predicting Crimes Than Random People,” Ed Yong: [ Ссылка ]

“The Age of Secrecy and Unfairness in Recidivism Prediction,” Rudin, Wang, and Coker: [ Ссылка ]

“Practitioner’s Guide to COMPAS Core,” [ Ссылка ]

State v. Loomis summary: [ Ссылка ]

*** LINKS ***

Vsauce2:
TikTok: [ Ссылка ]
Twitter: [ Ссылка ]
Facebook: [ Ссылка ]
Talk Vsauce2 in The Create Unknown Discord: [ Ссылка ]
Vsauce2 on Reddit: [ Ссылка ]

Hosted and Produced by Kevin Lieber
Instagram: [ Ссылка ]
Twitter: [ Ссылка ]
Podcast: [ Ссылка ]

Research and Writing by Matthew Tabor
[ Ссылка ]

Editing by John Swan
[ Ссылка ]

Police Sketches by Art Melt
Twitter: [ Ссылка ]
IG: [ Ссылка ]

Huge Thanks To Paula Lieber
[ Ссылка ]

Vsauce's Curiosity Box: [ Ссылка ]

#education #vsauce #crime

свернуть