Data Visualization: Boston Homicides And Their Relation To Boston Public Schools, 2012–2014
by Maggie Quick
Content
For this data visualization, I wanted to examine the proximity of recent homicides to schools in Boston. As can be seen in this map, some schools or neighborhoods have many more homicides than others.
For example, schools in East Boston are close to the the site of four different homicides from 2012–2014, whereas across the water in Charlestown there were no homicides those years. Dorchester, Mattapan and Roxbury clearly have more homicides than other parts of the city.
Interestingly, West Roxbury and Hyde Park did not have any homicides involving guns for these three years, but did have three fatal stabbings.
Looking more closely, some homicides have taken place almost on school campuses. A February 2012 homicide was practically on the campus of the Blackstone Innovation School. Another 2012 homicide was next to UP Academy Dorchester. One 2013 homicide occurred immediately next to the Lilla G. Frederick Pilot Middle School. A March 2014 homicide occurred across the street from the Maurice Tobin K-8 School. Some news reports do not even mention when slayings occur near schools, such as this Boston.com story on Andre Matthew’s death in 2012.
Other similar circumstances can be found on the map. Perhaps most alarmingly, in 2012 67-year-old grandmother Barbara Coyne died after her attacker slit her throat while she was inside her home, which was next door to the Oliver Hazard Perry K-8 School. Coyne’s killer was sentenced to life in prison without parole in October 2015, according to multiple news sources.
Process
When coming up with this visualization, I realized I needed to pick a smaller range of years to deal with the inconsistencies in BPD’s data. I also did not want to overwhelm the map with data points. I decided to use the most recent years available because they are most relevant to students in BPS today.
I picked out the eight most relevant data points from BPD’s original spreadsheets: Case number, date, CC#, Victim, R/G, Age, Location of Incident, and Weapon. I went through and made sure all of the dates were written in the same format. I added columns for City, State, Country and Zip Code so I could map the homicides. Since this is BPD data, I could put Boston as the city for all of the incidents.
I had to plug each individual address into Google Maps to find the correct zip codes. A few of the original addresses included notes at the end like “Dor” or “Rox” to denote the neighborhood they occurred in, which I deleted. I also made sure all of the categories of Weapons matched, and combined the categories of “Knife” and “Stabbing” and “Gun” and “GSW” in my map.
It took me a few days to gather and verify all of the data to plot the school locations. I used two lists provided on Boston Public School’s website, which turned out to be slightly contradictory. Once I was confident I had the correct number of schools, I created a spreadsheet and had to enter all of their names and addresses (available on the BPS website).
I also included a column with their grade level, but that data is not uniform and I decided showing it visually would make the map too busy. Plus, you can tell the grade level from many school names, and the information is still available in each individual data point. I selected my base map for its details — I want people to be able to see street names and the outline of the schools.
