Data.gov Federal Housing Datasets Catalog

The Data.gov Federal Housing Datasets Catalog sits on Datadory's single-family residential REITs shelf as the widest front door into American housing data: one index spanning 552,271 government listings from federal agencies, states, counties, cities, tribes and universities, filtered to the records that matter for single-family rental work. Useful housing queries surface HUD, Census Bureau, FHFA and SEC feeds alongside county registers and city permit files, and every listing arrives with its complete DCAT dictionary: title, publishing organization and type, bureau and program codes, publisher-declared cadence, temporal span, reuse designation, contact point and the formats attached.

Two behaviors shape how the slice works in practice. First, default relevance ranking interleaves trending popular listings regardless of topic, so a naive housing query returns disability tables and vehicle registrations before it reaches anything rentable - disciplined querying with quoted phrases, format facets and organization filters is the difference between noise and a pipeline. Second, every record declares its own cadence and temporal span, so the catalog doubles as a directory of which housing series are monitoring-grade versus baseline-only before anyone commits a pipeline to the wrong feed.

Get a sample of this dataset and see live rows from the housing slice - not a screenshot of rows.

Three listings pulled straight from the catalog, kept in their raw shape - including the pair that proves the relevance-ranking quirk:

# data.gov federal housing datasets catalog - three records as returned
title:                FY 2025 Disability Compensation Recipients by County
organization:         Department of Veterans Affairs
organization_type:    Federal
resource_formats:     json, xml, csv
dataset_last_updated: June 19, 2026

title:                Electric Vehicle Population Data
organization:         State of Washington
organization_type:    State
resource_formats:     json, xml, csv, kml, html

title:                Form N-PORT Data Sets
organization:         U.S. Securities and Exchange Commission
accrual_periodicity:  R/P3M (quarterly)
distributions:        27 ZIP files, 2019Q4 - 2026Q2

The first two rows are the cautionary tale: a veterans' compensation table and Washington State's EV registry outrank anything housing-specific under default ordering, which is exactly why the slice is curated rather than taken raw. The third row is the payoff - the SEC's Form N-PORT data sets, twenty-seven quarterly ZIP distributions running 2019Q4 through 2026Q2, the institutional-disclosure record that matters when you track who holds single-family exposure. A sample returns rows in exactly this shape, scoped to the publishers and geographies you name.

Ten named fields ride on every listing, and the dictionary below is the verified one - field names, types, definitions and example values as observed in the housing slice. Records are DCAT-native, so semantics follow the federal metadata standard rather than a vendor's renaming of it, and the ten-field spine stays identical whether a record comes from a federal agency or a county office.

Below the headline dictionary sit search-layer attributes that vary query to query - relevance scores, popularity counts, facet tallies, per-resource media-type detail. They are confirmed present in current pulls but not uniform across records, so they ship as additional fields on request: ask for them by name with your sample and we confirm exactly what each exposes for this slice.

• Geography · United States across federal, state, county, city and tribal publishers, with a bounding-box geographic filter offering Within and Intersect matching - so a Sun Belt metro pull stays inside its actual footprint instead of dissolving into state averages.
• Temporal · Declared per record: every listing carries its own temporal span, from static archives to rolling quarterly windows, plus both the publisher's modification date and the catalog's re-check timestamp - staleness audited listing by listing rather than assumed.
• Granularity · One catalog record per dataset, each referencing one or more distribution resources. This is a directory layer over the housing-data estate, not a row-level feed - which is precisely why it pairs cleanly with the row-level products sitting next to it on the shelf.

This slice scores 7/10 on Datadory's quality scale: the DCAT dictionary is fully verified, while individual listings vary in depth - the inherent trade of a federated catalog.

API, files, or your warehouse. Daily, weekly, or hourly.

You choose the channel and the cadence; the ten-field dictionary above travels unchanged across all three. Bulk files suit teams loading the whole housing-slice index once and joining it to portfolio or CRM tables. Scoped extracts suit products that surface publisher-matched records inside an app - one agency, one state, one bounding box. Warehouse delivery suits analysts running market-by-market models in SQL against continuously refreshed tables. Cadence changes are a settings conversation, not a re-integration project, and every sample arrives mapped to the exact dictionary above.

Market mapping before site selection. Single-family rental is a two-pure-play public business - Invitation Homes and American Homes 4 Rent - competing metro by metro, and every underwriting model starts with knowing which agencies hold which records. Publisher and organization-type fields turn 552,271 listings into an agency-by-agency map of who publishes housing data where: HUD program counts, Census tenure and vacancy series, FHFA price indexes, county registers. Pair the map with the Census Housing Vacancies and Homeownership (CPS-HVS) feed for the vacancy numbers themselves.

Underwriting input discovery. Instead of hunting agency properties one at a time, the accrual-periodicity field separates monitoring-grade series from static baselines before a pipeline gets built around the wrong feed. The mortgage-side twin, the Data.gov Federal Mortgage Datasets Hub, extends the same directory posture to lien and lending records.

Institutional positioning. The SEC's Form N-PORT data sets - twenty-seven quarterly ZIP distributions spanning 2019Q4 through 2026Q2 - sit inside this slice, making fund-level disclosure tracking a catalog lookup rather than a standing research project.

Metro-footprint monitoring. Bounding-box filtering with Within and Intersect matching scopes pulls to the operating footprint itself, so county and city records arrive matched to metros rather than averaged up to states.

Market researchers & consultants. The publisher roster plus bureau and program codes make agency-by-agency publishing footprints auditable - who ships housing statistics, at what declared cadence, in which formats. The market researchers breakdown ranks this slice against alternatives for the industry.

Competitive-intel & product teams. Tracking which disclosures the pure-play operators must file means knowing where those disclosures live; the organization field plus contact points resolve the owning office behind every record. Deeper comparisons live at the competitive intel page.

Investors & quants. Accrual periodicity and temporal coverage work as pre-flight checks: screen candidate series for cadence and span before committing them to a model. The investors and quants page carries the industry ranking.

Data scientists & ML engineers. DCAT-native dictionaries drop straight into schemas without manual annotation, and the ten-field spine joins cleanly to the row-level housing feeds already in the warehouse. The data scientists page ranks the slice for feature work.

Notes and adjacent reading:

• Single-Family Residential REITs data hub - the pooled industry view, from federal catalogs to loan-level tapes.
• Best single-family residential REITs datasets - where this slice sits in the ranked shortlist.
• Single-family residential REITs data guide - the landscape tour before the field-level dive.
• vs Nareit REIT Directory - a government-wide index against the sector association's own member roster.
• Data.gov source profile - the aggregator behind the catalog, and everything else it publishes.
• Census Housing Vacancies and Homeownership (CPS-HVS) - the row-level tenure and vacancy series this catalog points at.
• Data.gov Federal Mortgage Datasets Hub - the same directory posture, aimed at lien and lending records.