This data file contains the tract-level data used to generate an updated version of the Emergency Rental Assistance Prioritization (ERAP) Index 2.0. The index includes the 10 indicators, standardized percentile scores for the three subindices into which those indicators are grouped, and standardized percentile scores for the overall index value. It also includes data on the number of extremely low-income (ELI) renters that were used to exclude tracts with zero ELI renters from the index. The data dictionary provides specific definitions for each of the fields in the provided tract-level CSV and geoJSON.
The research team constructed the index using data from two primary sources: American Community Survey (ACS) five-year estimates and the US Department of Housing and Urban Development’s Comprehensive Housing Affordability Strategy (CHAS) dataset. We are using the most recent data available as of this update: the 2017–2021 estimates for the ACS and the 2015–2019 CHAS data.
The total index is made up of three subindices: Housing, Household Characteristics, and Income. Each subindex is made up of two to five indicators, pulled from the two data sources. The indicators were each standardized into z-scores, which were indexed to the state level. The index was constructed with weighted averages of the three subindex values. Weights were estimated empirically using tract-level eviction filing rate numbers from the Eviction Lab, which we consider to be a proxy for a tract's level of need for rental assistance. The feature maps and displays the percentile (instead of raw values) that a tract falls into, both for the full index and each subindex, compared with all other census tracts in its state.
The updated feature for this index is Mapping Neighborhoods with the Highest Risk of Housing Instability and Homelessness. The original feature in which the index was first published is Where to Prioritize Emergency Rental Assistance to Keep Renters in Their Homes.
Data and Resources
|Data Dictionary Files
|Original Data Source
|Data Quality or Limitations
Tract-level analysis using ACS data faces uncertainty due to sampling error because the survey only reaches a subset of Americans. Neighborhood-level estimates are more granular, but because they are based on fewer people, they are inherently less certain. There may be other omitted variables that contribute to a census tract’s need for rental assistance, including legal or policy conditions.
The ERAP Index 2.0 should be used as one input to informing community decisionmaking, along with local sources of quantitative and qualitative data and substantive input from residents with lived experience of housing instability.
Urban Institute. 2023. “Rental Assistance Priority Index 2.0.” https://datacatalog.urban.org/dataset/rental-assistance-priority-index-20. Data originally sourced from the 2017-2021 American Community Survey and the 2015–2019 US Department of Housing and Urban Developments Comprehensive Housing Affordability Strategy data, developed at the Urban Institute, and made available under the ODC-BY 1.0 Attribution License.