← All projects

LAPD Crime Data Analysis

Tools: R (dplyr, ggplot2, lubridate), RStudio

Period covered: 2020–2023 · Tools: R (dplyr, ggplot2, lubridate), RStudio

Overview

Using publicly available LAPD crime data, I isolated a 188% spike in vehicle burglaries in Central LA and identified the specific time windows driving it — 6PM to midnight, concentrated in parking areas — to support public safety resource planning.

Approach

Data was obtained through LAcity.org and refined to exclude incomplete 2024 records. The dataset was cross-referenced against LAPD crime and MO codes to provide descriptive context for reported incidents, while geographic identities for each LAPD area were established using publicly available LAPD reference information.

Findings

  • 188% increase in vehicle burglaries in Central LA over the period analyzed.
  • Incidents concentrated in the 6PM–midnight window.
  • Parking areas were the highest-risk location type.

alt textalt text

Impact

By addressing resource allocation, infrastructure improvements, and increasing public awareness, actionable steps can be taken to combat an emergent trend of vehicle break-ins within Central LA. The data suggests these strategies would not only help reduce crime but also contribute to a safer environment for the public at large.

Code

View the analysis on Kaggle