Drug Crime Density in Boston

Drug Crimes in Boston with high-density areas shown using blue gradient shading. Each shade represents 1/8 of the data. Crime scenes are shown using gray dots.
Drug Crimes in Boston with high-density areas shown using blue gradient shading.
Each shade represents 1/8 of the data.
Crime scenes are shown using gray dots.

The original map can be found in my previous post, Spatial Data Visualization with R.

As a review, I use get_map function in the ggmap package to grab a map of Boston from Google Maps. The “crime report” data can be found at https://data.cityofboston.gov/. In the code below, I bring in crime data as a csv file. The data contains one observation per crime. It includes a description, crime category (drug, traffic violation, etc), and the longitude/latitude coordinates of the crime scene.

I added density areas using stat_density2d function. I feed this function a set of coordinates using the x= and y= parameters. The alpha parameter adjust transparency with 1 being perfect solid and 0 being fully transparent. I set the fill to vary with the number of points in the shaded area (..level..). I also specify bins=8, which gives us 7 shades of blue. The density areas can be interpreted as follows: all the shaded areas together contain 7/8 of drug crimes in the data. Each shade represents 1/8 of drug crimes in the data. Since all shades represent the same proportion of the data, the smaller the area of a particular shade, the higher the arrest density.

R code is given below.

//update: Thanks to Verbal for the meaningful distinction between crimes and arrest. It’s not really clear what this data actually tracks. I’m sure that this is reported crime, as called in by people. I don’t think every crime here leads to an arrest. I could be wrong.


## load data
lib='/...directory with CSV crime file.../'

## subset to drug crimes
drugs=boston[which(boston$INCIDENT_TYPE_DESCRIPTION=='DRUG CHARGES' & boston$Year=='2014'),]

## plot drug and shooting crimes
bos_plot=ggmap(get_map('Boston, Massachusetts',

## Density areas
bos_plot +
  # density areas
  stat_density2d(data = drugs,
                 aes(x = drugs$Lat, y = drugs$Long, alpha=.75,fill=..level..),
                 bins = 8,
                 geom = 'polygon')+
  #density legend
  guides(fill = guide_colorbar(barwidth = 1, barheight = 10)) +
  # crime data points
             col='gray', alpha=.5,size=1)+
  scale_alpha(guide = FALSE)+
  # Labels/Title
  ggtitle('Drug Crimes - 2014 - Boston, MA')

4 thoughts on “Drug Crime Density in Boston”

  1. Don’t forget that you’re looking at arrest data, not crime data. That is, you’re not seeing where people buy drugs, you’re seeing where they get caught.

    That could be related to the frequency of those crimes… Or they could be everywhere and police are looking harder at some places.

    1. I think that’s a great point to bring up. I would be interested to see further the buildings and street access points where more of the arrests are happening- just to see if the landscape and design of certain high arrests areas might play into more or less visible areas for crime. I wonder how one would go about obtaining the crime area? Overall though nice job with visualizing this data!

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s