Methodology

How we source, process, and present housing data across OpenHousing.

Data Sources

SourceDataUpdate Frequency
Census ACS (1-Year)Income, rent, demographics, tenureAnnual (September)
Census ACS (5-Year)ZIP-level estimates, small geographiesAnnual (December)
HUD CHASCost burden by income, race, household type~2 year lag
HUD Fair Market RentsFMR by county and metroAnnual (October)
FFIEC HMDAMortgage application dataAnnual (March)
Zillow ZHVIHome values by ZIP, metro, stateMonthly
Zillow ZORIObserved rent indexMonthly
FRED (Federal Reserve)Mortgage rates, HPI, economic indicatorsWeekly/Monthly
BLSEmployment, wages by occupationAnnual/Quarterly
HUD PIT/HICHomelessness countsAnnual (June)
Census Building PermitsNew construction permitsMonthly
Eviction LabEviction filing dataWeekly

Affordability Score

OpenHousing's affordability score (0–100) is a composite index that measures how affordable housing is in a given geography relative to local incomes. A score of 100 represents perfect affordability; 0 represents extreme unaffordability.

Components

  • Rent-to-Income Ratio (30% weight): Median gross rent as a percentage of median household income. Benchmarked against the 30% affordability standard.
  • Price-to-Income Ratio (25% weight): Median home value divided by median household income. Compared to the historical norm of 3.0x.
  • Cost Burden Rate (25% weight): Share of renter households paying 30%+ of income on housing. Lower is better.
  • Income-to-Mortgage Qualification (20% weight): Whether the median household income qualifies for a mortgage on the median-priced home at current rates (28% front-end DTI, 10% down, 30-year fixed).

Score Ranges

  • 70–100: Affordable — Housing costs are manageable for most households
  • 40–69: Moderate — Significant share of households face cost burden
  • 0–39: Crisis — Majority of lower-income households are severely burdened

Update Frequency

OpenHousing data is updated on the following schedule:

  • Monthly: Zillow home values and rents, mortgage rates, building permits
  • Quarterly: Affordability scores, rankings, metro comparisons
  • Annually: Census ACS data, HMDA mortgage data, HUD CHAS and homelessness data, BLS wage data

Limitations

  • Census ACS data has a 1–2 year lag. Our most recent ACS data is from 2023.
  • HUD CHAS data has a ~3 year lag and uses 5-year ACS estimates.
  • PIT counts are acknowledged to undercount the homeless population, particularly unsheltered individuals.
  • Zillow indices may not capture the full range of housing costs, particularly at the low end of the market.
  • Our affordability score is a simplified model — individual circumstances vary widely.

Questions?

For questions about our methodology, data sources, or calculations, contact us at info@thedataproject.ai.