Defining Stakeholders and Digging Deeper

We met with Sheri and Katie from the Westside Communities Alliance (WCA) on May 25th to better understand the needs of the community. We discussed the different stakeholders and possible roles in which they might use crime data and related visualizations. Sheri told us that several elders in the community blame the youth for the crimes that take place within their neighborhoods. It would be helpful if we could extract some age related information from the datasets we’ve received, to see if there is any correlation between age and crimes in the neighborhoods.

After collecting the census data, we filtered out the three NPU’s that comprise the Westside communities (K, L, and T) and broke down those NPU’s by the neighborhoods, giving us 9 neighborhoods to work with. We wanted to see what factors were correlated with crime, and more specifically, we were given a hypothesis of what the community members think about crime and wanted to test them.
First, we were told that the general consensus is that the youths commit crimes and that the seniors are the victims. To test this, we looked at the percentage of the neighborhood’s population that were under 18 and compared it to the crime rate in that neighborhood and found that the correlation coefficient was -0.09, indicating an extremely weak correlation.

Part of this may be explained by the fact that because all these neighborhoods are so close to each other, it is easy to commit crimes in an adjacent neighborhood.
However, we saw that the correlation was stronger with the percentage of the neighborhood population being seniors and the rate of crime. The correlation was about 0.45, still not very high, but much higher than the youth one.

pasted2

An extremely revealing correlation was between vacant housing units and crime, particularly rate of violent crime. Westside Communities Alliance brought up the idea that code violations or vacant housing may be related to crime, and indeed, the correlation was a striking 0.91. Our next step regarding this particular relationship would be to find out through our geocoding of the crime data whether the vacant houses are crime hotspots or if there is a lurking variable.
pasted3

Our meeting with the WCS helped us identify some key stakeholders within the community.

  • Religious leaders
  • Public safety chairs within NPU’s
  • Public safety officers
  • Civic leaders within NPU’s
  • Youth organizations
  • Members of the WCA
  • Business owners
  • Members of the city council

We spent some time brainstorming various use cases for the religious leaders, and how they might use a crime mapping tool we develop. Possibilities include

  1. View crime hotspots within the neighborhoods so religious leaders from various churches can coordinate their efforts in order to plan programs or interventions
  2. View trends in the data so they can analyze which interventions have had a positive impact on the community

We hope to develop further use cases for each of these stakeholders, so we can find commonalities and design a minimum viable product that would be useful to most, if not all of them.

We also spent time geocoding the datasets we have. Katie shared the latest dataset she received through the Freedom of Information Act, which includes close to 900 thousand records of crimes that occurred between 2008-2015. Including the previous datasets, that would mean we have about 2 million records that need to be geocoded. In an effort to optimize our efforts, we parsed the datasets for unique addresses, which reduced the number to around 200 thousand. We tried geocoding with Bing, but it flagged our machines once we went over the limit of addresses it allows us to freely geocode. We have since then shifted to ArcGIS, which so far has not put any restrictions on our geocoding process.

All of the information and data we have received about the westside has come from our partners at the WCA. It would be helpful for us to also meet with community members directly to get a first hand impression of what their concerns are. With this in mind, we plan to attend the Proctor Creek Stewardship Council Meeting on June 3rd, between 1-3pm. We also have the NPU meetings for K,L and T on our calendars, and plan to attend them later in June.