Carolina Data Challenge
Does composition of Chapel Hill residents influence the public satisfaction with the performance of Chapel Hill police?
By Sydney Hansen, Katerina Wu, Gabby Ruehle, and Merve Bayraktar
The Chapel Hill crime data set can lend some interesting insight as to the attitudes that shape public satisfaction with the Chapel Hill police department. We explored some of the demographic patterns that may influence this public perception of the work that the Chapel Hill police department does. Our findings are summarized below.
About the Data
Points that stood out from the exploratory graphs were as follows. More men than women are arrested in the Chapel Hill area, and the average age of those arrested varies only slightly over time of day. The number of arrests peaks during the months of August and declines in the month of September, coinciding with the time that many students arrive on campus for the beginning of the school year. The average age of those arrested has steadily increased over the years from 2014 to 2019. Unsurprisingly, the number of arrests peaks during the weekends, and is happens most often when alcohol is present. These findings lead to interesting implications about the demographics that interact with the Chapel Hill police force.
Findings
According to our analyses, the distribution of satisfaction of the Chapel Hill police seems to be distributed evenly around the town of Chapel Hill. The measure of satisfaction, determined by survey responses to various measures of police competency, is distributed roughly the same across categories. There are more responses in the sections of the map that have an older median age demographic, which may mean that older populations are more likely to answer the survey about satisfaction. This poses some potential challenges for determining the true satisfaction of Chapel Hill residents in the police, if all populations are not providing their responses.
Ideas for Further Analysis
Time of arrest is an interesting factor to consider in relation to the public image of Chapel Hill. As shown above, there are certain times in the first half of the day where the number of arrests spikes around 2 AM, and then gradually ebbs down at 6 AM. On the other hand, there seems to be a more even distribution of arrests being made during the afternoon that slowly rises to a peak at 11 PM. Beyond that simple breakdown, we can see that it’s typically safest during work hours. Furthermore the most common arrest is made during 2 AM for impared driving. Originally we analyzed which race was most affected by arrests by the CHPD, but we found this to be unhelpful as Chapel Hill is a predominantly white area and thus all arrests heavily skewed to white people being arrested. In future analysis it would also be interesting to compare whether the arrest percentage of black people is proportionate to the population percentage of black people living in Chapel Hill.
Applications for Future Regression Analysis
Future analysis could also focus on combining the satisfaction scores of the police department with the arrests and analyzing those factors by their location. This would provide insight into how people’s responses to the survey data possibly changed if more arrests were more prevalent around them. For example, the latitude and longitude indicators could pinpoint the location of the respondent, and regression could be used to determine the significance of the location of the respondent on their perception of the Chapel Hill police force.