The following project analyzes 311 complaint data from the NYC Open Data platform. The data represents the number of complaints issued to the NYPD in the first quarter of 2024. The data highlights patterns in complaint types and response times.
311 complaint data from the NYC Open Data platform provides access to a wealth of public datasets to improve transparency and public services in New York City. The data is accessible at NYC Open Data. 311 data is collated from residents’ reports through the 311 service. These reports can be filed through the phone, website, and mobile application for noise, sanitation, and safety concerns, among other things. The data includes complaint type, reporting date, incident location, and other essential information.
For this project, I worked with the New York Police Department (NYPD) complaints for the first quarter of 2024, analyzing response times with complaint-type severity across zip code patterns. This analysis will provide insight into the efficiency of public services in addressing community-reported issues and investigate if zip code location (neighborhood) and income influence resolution times.
Categorizing NYPD Data
NYPD 311 data was organized into buckets of similar complaint types to simplify the large dataset
NYPD 311 NYC Open Data contains hundreds of thousands of rows. Buckets were created to itemize similar complaint types and visualize them more meaningfully in Tableau. Since different complaint types have been grouped into more significant categories, or “buckets,” it is easier to see patterns in this data and trends and comparisons across categories. Consolidating data simplifies complex information, makes it easier to draw insights, and makes visuals more interpretable.
| Bucket | Explanation | Complaint Types |
|---|---|---|
| Noise-Related Complaints | Complaints related to noise disturbances in various locations (residential, public spaces, vehicles, etc.). | Noise – Commercial, Noise – House of Worship, Noise – Park, Noise – Residential, Noise – Street/Sidewalk, Noise – Vehicle |
| Traffic and Parking Issues | Complaints involving vehicles, parking violations, and traffic obstructions. | Abandoned Vehicle, Blocked Driveway, Illegal Parking, Traffic |
| Public Disorder and Safety | Complaints related to public disturbances, illegal activities, and public safety concerns. | Disorderly Youth, Drinking, Drug Activity, Encampment, Illegal Fireworks, Non-Emergency Police Matter, Urinating in Public |
| Street and Public Space Issues | Complaints involving public spaces, including graffiti, panhandling, and street performers. | Graffiti, Panhandling, Posting Advertisement, Squeegee |
| Animal-Related Complaints | Complaints involving the mistreatment or abuse of animals. | Complaints involving the mistreatment or abuse of animals. |
| Other Public Complaints | Miscellaneous complaints that do not fit into other categories, but involve public space use. | Bike/Roller/Skate |
Most Common Complaint Types
A word cloud provides a visual representation of the most common complaint types.Use the filters below for “Incident Zip” and “Incident Address” to view any resolved complaint types associated with your current address or ZIP code during January and March 2024. The larger the word, the more frequently that complaint was recorded. Looking at the most common complaint types logged into the 311 service provides additional granularity and further context on how data was consolidated into buckets.
Detailed Visualization for Resolved Complaints
Creating buckets and compressing large datasets involves trade-offs like the loss of granularity, but a heat map balances simplification by displaying detailed information. Bucket creation and compression of large data sets come at a cost: it may offer loss of granularity, oversimplification, and limited actionable insights. In this type of analysis, a balance must be found between simplification and maintaining enough detail to allow meaningful analysis. The following is a Tableau heat map that adds some granularity to the incident address, average resolution times, income level, Zip code, and complaint type. It also enables deeper analysis, which may be essential or even crucial to finding trends and patterns more powerfully.
The average resolution times were determined by subtracting each complaint’s “Created Date” from its “Closed Date” and then converting the duration to hours and minutes format (HH:MM). A second bucket named “Resolved” was created. This bucket divides resolution time by hour and day. You can view and filter these buckets with the “Resolved” filter on the right-hand side of the heat map.
Resolution Time by Borough
NYPD's 311 complaint resolution times show little variation across boroughs, despite similar population sizes.The time it takes the NYPD to resolve 311 complaints shows no significant variation between boroughs when accounting for population size. However, the Bronx recorded more complaints than Manhattan, which is relatively close in population size.
Similarly, Brooklyn and Queens are almost matched in population but have different complaint numbers. That said, there seems to be some slight disparity in the timely nature of case closures by the NYPD in Brooklyn versus those in Queens.
Complaints Resolved by Zipcode and Income
The analysis shows no significant difference in NYPD's average response times across high, mid, and low-income ZIP codes, indicating no apparent bias based on income levels.There is no significant discrepancy in average response times across ZIP codes with different income levels. Whether high, mid, or low-income, the overall averages of the response times are relatively consistent across income levels. The consistency in response times across income levels indicates that, at least from this analysis, the NYPD’s handling of 311 complaints does not seem to be influenced by a neighborhood’s socioeconomic status.
The resolution time for complaints in high-income ZIP codes is reasonably consistent with middle- and low-income areas, indicating a relatively equitable way of processing complaints. While further investigation might reveal more nuanced patterns, this data suggests no apparent bias in response times based on income.
Looking Forward
Investigating additional factors affecting response times can offer a more detailed analysis.Future research should delve into more specific factors influencing response times, such as complaint types (without using “buckets”), longitudinal data spanning at least a year, or resource allocation within each borough. While this analysis does not provide evidence of apparent bias concerning income, analysis of other variables may give further insight into the drivers of complaint resolution. This study could inspire policymakers, public agencies, and the general public to monitor and improve equitable service delivery across all neighborhoods. Additionally, expanding the scope of study to qualitative data from residents, such as satisfaction or efficiency of the resolution, might allow researchers to present a bigger picture of public service performance. Finally, conducting statistical analyses would be valuable in thoroughly examining the data and uncovering potential correlations or causal relationships.
