Skip to contents

County-level homeless counts with missing values imputed, plus selected American Community Survey (ACS) demographic covariates. Counts are produced by disaggregating HUD Continuum of Care (CoC) Point-in-Time counts to counties using population-density weighting and imputing counties with no CoC via a spatial Poisson model, following Almquist, Helwig and You (2020) doi:10.1080/08898480.2019.1636574 . See homeless_na for the pre-imputation version.

Usage

homeless

Format

A data frame with 3143 rows (counties and county equivalents in the 50 states and DC) and the following variables:

state

Two-digit state FIPS code.

fips

Five-digit county FIPS code.

count07

Estimated homeless count, 2007 (imputed where missing).

count08

Estimated homeless count, 2008.

count09

Estimated homeless count, 2009.

count10

Estimated homeless count, 2010.

count11

Estimated homeless count, 2011.

count12

Estimated homeless count, 2012.

count13

Estimated homeless count, 2013.

count14

Estimated homeless count, 2014.

count15

Estimated homeless count, 2015.

count16

Estimated homeless count, 2016.

count17

Estimated homeless count, 2017.

count18

Estimated homeless count, 2018.

count19

Estimated homeless count, 2019.

count20

Estimated homeless count, 2020.

count21

Estimated homeless count, 2021.

count22

Estimated homeless count, 2022.

count23

Estimated homeless count, 2023.

count24

Estimated homeless count, 2024.

count25

Estimated homeless count, 2025.

population

County population (ACS 5-year estimate).

density

Population density: 2010 Census population per square km.

pctblack

Percent Black population.

pctageunder18

Percent of population under 18.

pctfoodstamp

Percent of housing units receiving SNAP/food stamps.

pctvacanthousing

Percent of housing units vacant.

pctmedhousingcost

Median annual housing cost as a percent of median household income.

medhousingval

Median housing value, in $1,000.

pctmarried

Percent of population now married.

avg

Mean homeless count across 2007–2025.

state_name

State name.

Source

Homeless counts: HUD Point-in-Time data (see hud2007) disaggregated to counties. Covariates: U.S. Census Bureau American Community Survey 5-year estimates, via the tidycensus package.

Details

Estimates are bounded by a plausibility cap of 2,000 homeless per 100,000 residents (2 percent of population), or 3,500 per 100,000 for the dense major-metro cores (New York City boroughs and Los Angeles County). The cap is applied by redistributing any excess within the same CoC, so a CoC's Point-in-Time total is preserved across its counties. HUD CoC counts (hud2007) are never altered (none exceed the cap in any year).

References

Almquist, Z. W., Helwig, N. E. and You, Y. (2020). Connecting Continuum of Care point-in-time homeless counts to United States Census areal units. Mathematical Population Studies, 27(1), 46–58.