Instructions. Use the dropdown menus at the bottom right of the following interface to select a type of crime to map (Violent, Property, Homicide, or All Index Crime) and one or more years to include in the analysis.

The eviction filing data, which cover the years 2016 through 2021, were obtained through a formal request to the New York State Unified Court System. Crime data were obtained for the same time period through the Rochester Police Department (RDP) Open Data Portal. Crime classifications — Violent, Property, Homicide, and All Index Crimes — follow the definitions provided in the RPD Open Data Portal.

Selecting a crime category from the relevant dropdown menu will update the right-hand-side map in the portal to show the rate of that crime per 1,000 persons. Population data come from the 2020 U.S. Census PL 94-171 Redistricting Data File. In all crime maps, “colder” colors (e.g., darker blue hues) are associated with low crime rates, while “warmer” colors (e.g., orange and red) show higher crime rates. Due to the resolution of the eviction filings data, which were provided at the ZIP Code level, all maps and charts use the ZIP Code as the unit of analysis. Hovering over any ZIP Code on either map will show precise data for that ZIP Code.

Once a crime type is specified, the scatterplot in the bottom left of the portal will graph ZIP Code-level rates of that crime type versus eviction filing rates per 1,000 persons. Each scatteplot includes a dashed trend line that summarizes the relationship between crime and eviction filings. When a trend line slopes up and to the right, the interpretation is that the two variables are positively related — when one increases, so does the other. The closer the data points lie to the trend line, the stronger is the relationship between the two variables. Scroll down below the portal for specific details on the relationships between eviction filings and the four categories of crime shown in the portal.

All data are shown as annual averages. By default, annual averages cover the entire range of data, from 2016 through 2021. Users can select individual years to explore the relationships for any given year of data; alternatively, it is possible to select multiple years (which is advisable, since any given year gives an incomplete picture), such as the most recent three years (e.g., 2019, 2020, and 2021).

Clicking on a data point in the bottom-left scatterplot will zoom to that location on the two maps. To return to the main view, simply click on the same data point in the scatterplot (again) to deselect it.

Explore the Data:

Relationships between Crime and Eviction Filings

One way to characterize long-term, general relationships between a given category of crime and eviction filings is to accept the portal’s default time horizon and explore the relationships for all six years’ worth of data. In that case, observe that all four trend lines that can be computed — one for each category of crime — are upward sloping, indicating that each of the four crime types are positively/directly related with eviction-filings. Put another way, when eviction filing rates go up, so does crime.

Two properties of each trend line can be used to add some clarity to these relationships between crime and evictions: (1) the slope, and (2) strength of association (or correlation coefficient). The following table summarizes these two parameter estimates for each trend line/crime type:

Keep in mind that the above relationships show correlation, not causation. Eviction filings and crime are related in ways that cannot be explained by chance alone; but surface-level patterns in these variables are products of deeper structural variables, such as poverty and inequality. In that sense, the foregoing maps document multiple, correlated symptoms (housing insecurity, violence, and property crime) that emerge from larger systemic problems (concentrated poverty, exploitative economic relations, racialized social and political institutions, etc.). Whereas palliative measures are necessary to treat these and other symptoms in the short term, in the long-run, transformational changes to prevailing social norms and institutions (i.e., systems changes) are needed to neutralize their root causes.