sfbrigade/datasci_justice-by-public-defender

NLP and inferential stats to identify which public defenders more consistently seek plea bargains for their defendants, independent of case merits.

Justice by Public Defender

A Social Justice Project by the Data Science Working Group at Code for San Francisco

This project aims to use natural language processing and inferential statistics to identify which public defenders, from case records, reliably seek plea bargains for their defendants, independent of case merits. As well, this project will employ regression analyses to look for relationships between kinds of cases, defendants, and outcomes, assuming the demographic data can be cross-referenced from somewhere.

Team:

• Jude Calvillo (project lead and contact)
• And YOU? (seeking teammates)

Known complications:

• Access to case records is siloed, not standardized, and often difficult to gain.
• Case record formats can differ by county/state/court.
• Given the above, we'll focus on gaining the records of a single county or of the State of California.

Status, as of Dec. 4, 2016:

• Looking for a community partner. The Innocence Project seems ideal.
• Looking for an efficient, reliable way to secure all criminal case records, of some range of years, from one or more California counties.

Keywords:

social justice, criminal justice, state of california, justice system, public defenders, public defense