Green Space Accessibility for Affordable Housing
Interactive dashboard analyzing disparities in access to high-quality public parks near affordable housing buildings in Chicago
Overview
Green Space Accessibility for Affordable Housing is a data science project that analyzes disparities in access to high-quality public parks near affordable housing buildings in Chicago. By combining housing data, census tract information, and spatial/ratings data on city green spaces, we developed an Accessibility Index that quantifies access based on park ratings, size, and proximity.
What It Does
The project includes:
- Accessibility Index: A quantitative measure combining park ratings, size, and proximity to affordable housing
- Interactive Dashboard: Built with Python Dash to visualize the Accessibility Index and enable exploration of green space access patterns
- Spatial Analysis: Integration of OpenStreetMap park data, Yelp/Google Places API reviews, and U.S. Census Bureau ACS data
- Inequity Identification: Reveals patterns of inequity in green space access across Chicago neighborhoods
Why I Made It
This project was developed for CAPP 30122: Data Science for Public Policy at the University of Chicago. Public parks and green spaces are vital to community well-being, providing spaces for recreation, social connection, and mental health. Ensuring equitable access to high-quality green spaces is a key responsibility for urban planners and policymakers—especially for residents of affordable housing.
Technologies & Skills
- Python for data processing and analysis (Pandas, GeoPandas)
- OSMnx API for extracting OpenStreetMap data on Chicago parks
- Yelp & Google Places APIs for park ratings and reviews
- U.S. Census Bureau ACS API for census tract data
- Python Dash for interactive dashboard development
- Spatial Analysis using GeoPandas for proximity calculations and spatial joins
- Data Pipeline Design integrating multiple data sources (housing, parks, reviews, census)
Real-World Impact
The interactive dashboard empowers planners and decision-makers with a data-driven approach to:
- Identify areas with limited access to quality parks near affordable housing
- Promote more equitable urban green space planning through evidence-based insights
- Support policy decisions on park development and resource allocation
Team & My Role
Collaboration: Begum Akkas, Andrés Camacho, Evan Fantozzi, Grace Kluender
My contributions:
- Designed and developed the end-to-end data pipeline architecture integrating multiple data sources
- Created the interactive Python Dash dashboard with accessibility index visualizations
- Developed Kepler.gl maps for spatial visualization of green space accessibility patterns
- Implemented OpenStreetMap data extraction workflows using OSMnx API
- Built Google Places API integration for park ratings and reviews data collection
- Developed comprehensive pytest test suite for key dashboard functionalities and data pipeline components
- Processed and cleaned multi-source data (housing developments, park locations, reviews, census tracts)
Source Code
Repository: uchicago-2025-capp30122/30122-project-treehuggers
Data Sources
- OpenStreetMap Chicago Parks Data - Extracted via OSMnx API for Python
- Affordable Rental Housing Developments - City of Chicago data portal
- Yelp Business Search API - Park ratings and reviews
- Google Places Nearby Search API - Additional park ratings data
- U.S. Census Bureau ACS - Census tract demographic and economic data