Digital Storytelling


Is Oakland Safe? An Investigation into Pittsburgh Arrest Data

By Sarah Danielle

January 30, 2019

            For this Pittsburgh Data Set Project, I chose to look at the Pittsburgh Police Arrest Records and was specifically interested in arrests in Oakland. My idea for what to do with that data actually came to me as I was writing my previous blog post introducing myself. Since I went to a small, private Christian school growing up, I was a pretty big anomaly when I chose to attend a large, secular school—in a city no less! I started thinking about how when high school students tour campuses (I toured over a dozen, so I consider myself an expert in this subject), a large selling point for many is campus safety and the tools that the university provides to keep their students safe. Not only is this to show to students that they will feel comfortable on campus, it is also a big play to the parents. Many parents of soon-to-be college students are most concerned not about their child’s prospective college location or size, but of how safe the campus is. I realized that I could use the data provided by the Pittsburgh Police on arrests to see how safe Oakland is relative to the rest of the city.

            When I looked at the data, I decided to limit the arrest dates to months and days when Pitt’s campus was open, because that is what would pertain to students the most. After finding Oakland’s percentages, I realized that comparing them to other neighborhoods would be the most effective way to prove that Oakland’s number of arrests is surprisingly proportional. At first, I considered a bar graph, which I did make and use, but I realized that it doesn’t quite portray the gravity of the difference in arrest percentages as compared to population and area. The next step I took was to create a heat map. This took a lot of playing with to get an image that portrayed the data that I wanted, especially because I was working with a monster spreadsheet that had over 17,000 rows of data. What I came up with gives the reader a better understanding of the concentration of arrests over the neighborhoods and the color gradient shows Oakland as a much less concentrated area than others. I attempted looking into getting a WMS for the map underlay, but it proved to be a little too complicated for my skills.

            Given more time and after gaining some more experience, the implications of this research are really interesting. I could further this investigation by analyzing the breakdown of different kinds of crimes in Oakland and the surrounding neighborhoods to find what is most prevalent and translate that into easily understandable data visualizations. Additionally, I could compare this data with other city schools in the state to see how the different universities compare to one another. Overall, this was a really interesting undertaking. The data visualization process was frustrating at times and I definitely am still not quite confident with the coding aspects of all of this, but it was very satisfying to get the information written down and in theory, out there for others to view and understand. I am excited to learn more about using my powers of digital literacy for good!