← Back to the write-up
Independent analysis · built on a real published study

Where the things that predict crime actually change across Chicago

Most crime maps assume one relationship holds true everywhere in a city. This one doesn't: it works out a separate answer for every neighborhood, so you can see exactly where each real factor matters a lot, matters a little, or points in the opposite direction than you'd expect. Built on real Chicago data from 2023–2025 (777 neighborhoods), using the same approach as a real 2025 study, Cai (2025), PLOS One.

Where each factor matters more, less, or the opposite of what you'd expect

Pick a crime type, then pick what to look at. "How well it fits" shows how much of that crime type these four factors actually explain in each part of the city. Each factor's own map shows where more of it lines up with more crime (green) or less crime (red) — not just whether it matters somewhere in Chicago, but exactly where and how strongly.

Does crime cluster into real patterns, or is it scattered randomly?

All three crime types cluster into clear, real patterns across the city rather than being scattered at random. That's what justifies breaking this down neighborhood by neighborhood instead of settling for one citywide average. ("Moran's I" below is the statistical test that confirms this — further from 0 means stronger real clustering, and the tiny p-values mean it isn't a coincidence.)

What else is connected, even outside the main model

Two more real things were checked against each crime type, but kept separate from the main model, the same way the original study handled them: how dense the street network is, and how many vacant or abandoned buildings are nearby. Positive numbers mean more of that thing tends to go along with more of that crime type; negative numbers mean the opposite. Vacant buildings turn out to matter a lot for criminal damage and battery, but barely at all for theft.

Sources

  1. Boeing, G. (2017). OSMnx: New methods for acquiring, constructing, analyzing, and visualizing complex street networks. Computers, Environment and Urban Systems, 65, 126–139. https://doi.org/10.1016/j.compenvurbsys.2017.05.004
  2. Cai, Y. (2025). Advancing urban management: Integrating GIS, LLMs, and media narratives into environmental and socio-economic analyses for enhanced urban crime analysis. PLOS ONE. https://doi.org/10.1371/journal.pone.0331788
  3. City of Chicago. (2026). 311 service requests [Data set]. Chicago Data Portal. https://data.cityofchicago.org/Service-Requests/311-Service-Requests/v6vf-nfxy
  4. City of Chicago. (2026). Boundaries – city [Data set]. Chicago Data Portal. https://data.cityofchicago.org/Facilities-Geographic-Boundaries/Boundaries-City/qqq8-j68g
  5. City of Chicago. (2026). Crimes – 2001 to present [Data set]. Chicago Data Portal. https://data.cityofchicago.org/Public-Safety/Crimes-2001-to-Present/ijzp-q8t2
  6. OpenStreetMap contributors. (2026). Chicago street network and points of interest [Data set]. OpenStreetMap. https://www.openstreetmap.org
  7. Oshan, T. M., Li, Z., Kang, W., Wolf, L. J., & Fotheringham, A. S. (2019). mgwr: A Python implementation of multiscale geographically weighted regression for investigating process spatial heterogeneity and scale. ISPRS International Journal of Geo-Information, 8(6), Article 269. https://doi.org/10.3390/ijgi8060269
  8. U.S. Census Bureau. (2023). American Community Survey 5-year estimates, 2019–2023 [Data set]. https://www.census.gov/programs-surveys/acs
  9. U.S. Census Bureau. (2023). 2023 cartographic boundary files: Census tracts [Data set]. https://www.census.gov/geographies/mapping-files/time-series/geo/carto-boundary-file.html
  10. U.S. Environmental Protection Agency. (2021). Smart Location Database, version 3.0 [Data set]. https://www.epa.gov/smartgrowth/smart-location-mapping