Second project | Mass shooting in America: A serious and recurrent problem
For my second project I wanted to visualize the mass shooting problem in the United States. I thought that using an animated map that would show on a timeline each of the occurrences along with the number of people injured and killed was a good way of making an impact about how this has become a widespread phenomenon in the United States over these last couple of years.
For this project the best data I was able to find comes the Gun Violence Archive project that has compiled all the mass shooting in the United States since November 2016. For methodology purposes I wanted to make sure if there was an official definition for “mass shootings”. According to a July 2015 Congressional Research Service report a mass shooting happens only when 4 or more victims were murdered using firearms. So I proceeded to filter my dataset that originally had 500 incidents to include only the shootings when this was true.
I then proceeded to geocode all the cities I had in my data for placing them in a map. I used a service called GPS Visualizer that allows to input the name of a city and state and then get the corresponding latitude and longitude by using Bing Maps API. For visualizing my findings At the beginning I thought of using Carto which is a very good tool for making animated maps. There was, however, something missing : I wanted to represent the magnitude of each shooting by circle size, something that Carto doesn’t allow.
So I tested for the first time the Map time series template available in Flourish which gave me almost everything that I wanted: size by magnitude, a counter of victims and a very nice graphical (and responsive) interface. If I wanted to be picky, I would love to be able to hover on the circles when the map is in motion for having more information on each shooting and probably being able to see the date on the screen when the circle appears for a more accurate depiction. None of this was possible on the tool.
For completing my visualization I added a second chart that summarizes the 5 largest mass shooting found in my database. This is interesting because it allowed me to really grasp the magnitude of the Texas mass shooting that happened last year.
Things that I would like to do if I had more time
1. Find a larger dataset with more historical data. Having data for two years is nice, but for being able to draw a deeper understanding and the evolution of this phenomenon it would be great to have more data.
2. Doing more charts to contextualize about this phenomenon with other data sources. For example, gun ownership rate per 1000 inhabitants, average price of a gun in USA etc.