ACS–NYCHA Census-Tract Analytical Dataset
Description
This dataset and accompanying materials support the study “The Unequal Housing Geography Beneath New York City’s Diversity,” which examines how racial and ethnic diversity is associated with housing conditions, neighborhood socioeconomic context, and institutional public housing presence across New York City census tracts. The study draws primarily on 2020–2024 American Community Survey (ACS) five-year estimates, New York City Housing Authority (NYCHA) development information, and 2020 census-tract geography. The principal analytical dataset contains 2,185 census tracts across the Bronx, Brooklyn, Manhattan, Queens, and Staten Island. The repository includes tract-level ACS analytical datasets; ACS–NYCHA integrated analytical data; cleaned NYCHA development-level data; tract-level NYCHA spatial-allocation data; New York City census-tract geometry; R scripts for data preparation and spatial integration; recovered supporting analysis code; and study-specific replication documentation, including a data dictionary, methodological transparency statement, computational-environment record, file inventory, and replication instructions. The repository is designed to support transparent inspection, verification, and reproducibility of the documented research workflow. The preserved analysis code reflects the computational materials recovered from the research process and should not be interpreted as a complete historical reproduction script for every model reported in the associated paper. Generated statistical outputs—including regression tables, exported model summaries, descriptive-output files, figure images, screenshots, and other generated results—are intentionally excluded. The associated paper remains the authoritative source for reported findings, tables, figures, limitations, and substantive interpretation.
Files
Steps to reproduce
Download the repository and retain the folder structure (01_data, 02_code, and 03_documentation). Review README_URBANA_Replication_Guide.docx, URBANA_Replication_Instructions.docx, and URBANA_Data_Dictionary.xlsx for the repository structure, file descriptions, variables, and reproduction scope. Use acs_tract_analysis_expanded.csv as the core ACS tract-level analytical dataset. The principal analytical sample contains 2,185 retained New York City census tracts. To inspect the NYCHA data-preparation workflow, run 01_clean_nycha_data.R. This script cleans and standardizes the NYCHA development-level data used in subsequent spatial processing. Run 02_join_nycha_to_tracts.R to reproduce the documented spatial-integration workflow linking NYCHA development geography to 2020 census tracts and constructing tract-level NYCHA measures. Local file paths may need to be adjusted to the user's directory structure. The resulting tract-level NYCHA measures can be compared with nycha_tract_spatial_allocation.csv and the integrated analytical datasets acs_tract_analysis_expanded_plus_nycha.csv and acs_tract_analysis_expanded_rich_nycha.csv. 02_analysis_correlations_regression_extracted.R contains recovered supporting analysis code and is retained as part of the computational provenance. It should not be interpreted as a complete historical reproduction script for every regression model reported in the associated paper. Generated outputs are intentionally not included in the repository. Researchers should generate analytical outputs independently from the deposited data and code within the documented reproduction scope. Reported statistical results, tables, figures, and their interpretation should be compared with the associated paper. The preserved scripts are written in R. See URBANA_Computational_Environment.docx for verified computational requirements and URBANA_Methodological_Transparency_Statement.docx for methodological and reproducibility boundaries.