IPUMS USA Full Count Census Data
IPUMS, University of Minnesota
Every person counted in U.S. censuses from 1850 through 1950, coded and ready for analysis.
Complete-count microdata from historical U.S. censuses, built with genealogy partners and harmonized by IPUMS. Each row is a person in a household, with age, sex, race, birthplace, occupation, and more; 1790-1840 are household-level only. It makes it possible to study small places, rare groups, and individual life paths.
Access
- Cost
- FreeFree extracts after registering; versions with names are available only under restricted licenses.
- License
- Provider's own termsIPUMS terms of use (citation required, no redistribution)
- Sign-up
- Free account
- How you get it
- API and Bulk downloadCSV, Stata, SPSS, SAS, fixed-width
What's in it
- One row is
- One row per person (1850-1950) or per household (1790-1840)
- Covers
- United States, down to county and often enumeration district
- Years
- 1790-1950
- Updated
- Irregular
- Size
- Hundreds of millions of person records across census years (1950 alone covers about 150 million people)
Quality
The gold standard for historical U.S. population microdata, with extensive harmonization and documentation.
- Names are withheld from the public files
- Transcription errors and undercounts vary by year and group
Connects to
- Ancestry U.S. Federal Census Collection: Same enumerations with page images and names, behind a paywall
- Chronicling America: Local newspapers for the same places and years
Questions it could answer
- How did occupational mobility from fathers to sons change between 1880 and 1940?
- Where did Black Southerners move during the first Great Migration?