Skip to main content

jbrnbrg

NYPD Article 221 Arrests and Summonses

Today’s post builds upon some of the work from a previous post to include historic and year-to-date Arrest and Summonses counts by NYPD precinct. The aim is to use visualization (plotly, primarily) to explore the impact of policy changes on racial equity over time.

Data

Historic and Year-to-Date information from NYC’s OpenData API for NYPD Arrests and Summonses related to Article 221 of the New York Penal Law (PL221) from 2009-07-01 through 2019-06-30 (10-years of data). Note that my data only includes PL221 arrests and summonses that come from NYPD patrol (e.g. transit police are not included).

I have excluded the code used within this post to keep the focus on the exploration but you can find applicable instructions to obtain data and create visuals from my previous posts.

Here’s a preview of the Arrest and Summons data after downloading and cleaning:

## Observations: 129,360
## Variables: 6
## $ yearmonth  <chr> "2009-07", "2009-07", "2009-07", "2009-07", "2009-0...
## $ source_dat <chr> "Arrests", "Arrests", "Arrests", "Arrests", "Arrest...
## $ pct        <int> 1, 10, 100, 101, 102, 103, 104, 105, 106, 107, 108,...
## $ susp_race  <chr> "AmIAN", "AmIAN", "AmIAN", "AmIAN", "AmIAN", "AmIAN...
## $ qty        <dbl> 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, ...
## $ event_date <date> 2009-07-01, 2009-07-01, 2009-07-01, 2009-07-01, 20...

PL221 Over Time

In the past 10 years, America’s largest police force has come under scrutiny several times based on the topic of racial equity and that appears to have coincided with a significant downturn in the volume of PL221 arrests. Hover over the red circles in the below plot to see notes providing context for a given month - links for all notes can be found at the end of this post.


The above plot depicts a 96% drop in the number of monthly arrests from a peak of 5,451 in March 2011 to just 218 June 2019 (the latest available data at the time of this posting).

PL221 Equity Over Time

Volume

When looking at the racial breakdown of monthly arrests and summonses, the volumes of each racial group have also dropped significantly. Click or double-click the legend entries in the below plot view races in isolation.


We can also see that the summons data only started including a breakdown by race late in 2016.

Proportion

When reviewing counts on the basis of the proportion of total PL221 Arrests per month among racial groups, the values seem persistent except for the last 12 months. The most-recent 12-month period shows a near-10-year-low percentage of white arrests (2019-06: 3.7%) while the percentage of black arrests is at a near-10-year-high (2019-06: 61%).


Government surveys show that black and white people use the drug at roughly the same rate (Table 1.25C) but the racial breakdown among arrests and summons tells a different story.

Although we don’t get an as-complete picture with the Summons data, we can see a similar proportional breakdown among races emerging in the last 12 months.

Closing Remarks + Shiny App

The good news is that policy changes appear to coincide with a reduction in the the overall number of PL221 arrests in the past 10 years. The bad news is that racial disparity among the recipients of PL221 charges have remained steady and actually worsened in recent months. Furthermore, substituting arrests for summonses still means entry into the criminal justice system - something that vulnerable populations can have difficulty navigating.

I’m interested in keeping an eye on this data as it updates so I’ve created this shiny app to do that:

The app is connected directly to the NYC OpenData API for NYPD historic and year-to-date arrests and summonses related to PL221 so as soon as new data is relaeased, the plots will update. Look out for an additional shiny app I am developing that’ll use this data along with plotly’s animation capabilities - here’s a preview:

Animated GIF of plotly’s animation feature

2019-09-19