Code Monkey home page Code Monkey logo

strategic-subject-list's Introduction

Strategic Subject (SSL) Data Analysis

The Chicago Police Department recently released de-identified information from arrest data used in its Strategic Subject List (SSL) model which assigns scores from 1-500 to individuals based on their likelihood to become "parties to violence" in the future, whether as potential victims or perpetrators. The information is on the Chicago Open Data Portal. The model is said to have gone through at least 5 iterations, with this being one of the later versions.

The data description indicates that this is the output of the algorithm being applied to all CPD arrest data in that time period (August 1, 2012 through July 31, 2016), and not necessarily the list itself. However, the distinction between "high priority" individuals--often cited as being between 400 and 1,400 people--is not clear.

Additionally, according to CPD Special Order 09-11 the default view for the Strategic Subjects List dashboard used by CPD personnel lists "all subjects with an SSL score of one or higher are shown and placed in descending order of score." If all scores are taken into consideration and viewed by CPD, then there is value in looking at patterns across all individuals on the list rather than just individuals with the highest scores, especially when the cutoffs/totals aren't made clear.

Total Numbers on the List

In total, at the time this model was run for (July 31, 2016, for the period from August 1, 2012 through July 31, 2016), there were 398,684 individuals in the records of arrest data used to produce this list. There is no clear indication of how many individuals CPD considers to be "on the list", but over time numbers ranging from 426 (in the initial iteration evaluated by RAND) to to 1,400 (in a May 2016 New York Times Story), with one source referring to 30,000 individuals being included (September 2015 Chicago Sun-Times story).

List Cutoffs

In stories describing the SSL, multiple cutoffs have been mentioned for individuals to be considered "on the list", or to be on the list and given additional scrutiny. The RAND evaluation (linked above) gave the impression that the scores are created and then a semi-manual selection process happens, which resulted in 426 individuals.

However, other stories have often referred to scores of 400 and above as being subject to higher levels of scrutiny, but often aren't explicit about whether they are an actual cutoff for being on the SSL or not. A May 2016 Chicago Tribune editorial puts the cutoff for inclusion at the "upper 200s or higher", which at the time Chicago police Superintendent Eddie Johnson described as only being made up of 1,300 people. This cutoff of the upper 200s is also mentioned in a December 2016 LA Times story. However, looking at the released data, 287,404 (72.09%) of the 398,684 individuals had score of 250 or above.

Counts by Score

Score Count Individuals
> 0 398,684
>= 100 397,917
>= 150 391,046
>= 200 354,994
>= 250 287,404
>= 300 173,001
>= 350 29,212
>= 400 3,721
>= 450 709

strategic-subject-list's People

Contributors

pjsier avatar

Stargazers

 avatar  avatar  avatar  avatar

Watchers

 avatar  avatar  avatar

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    ๐Ÿ–– Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. ๐Ÿ“Š๐Ÿ“ˆ๐ŸŽ‰

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google โค๏ธ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.