The U.S. Census of Population and Housing, mandated by Article I of the Constitution, counts every resident in the United States every ten years. The data are used primarily to allocate seats in the U.S. House of Representatives and to distribute federal funds to local communities.
Each decennial census results in a number of standard and special data products. Most of the standard products are summary files, which are essentially grouped tables of numbers.
Summary Files 1 and 2 are enumerations of all U.S. residents (100-percent data products). (Note that in 2020, this naming convention changed — SF1 is now called the Demographic and Housing Characteristics (DHC) file.) SF1 / DHC presents counts and cross-tabulations of basic information collected from all people and housing units. Characteristics include:
Summary File 2 adds detail for specific race and ethnic groups.
Summary File 3, a sample data product, contains tables with social, economic, and housing characteristics compiled from about 19 million housing units in 2000. SF3 contains much more detail than SF1. Some SF3 characteristics include veteran status, employment, place of birth, citizenship, migration, disabilities, household vehicles, and housing value, to name only a few.
Summary File 3 was last produced for the 2000 census. In 2005, the Census Bureau introduced the American Community Survey (ACS), a sample survey covering the same characteristics as SF3, but released on a yearly schedule.
As 100-percent data products, the summary files (and DHC) are not samples. Data for these products are collected via a list of ten or fewer questions — the so-called "short form". The census attempts to count all people at their "usual residence" on Census Day (April 1 of the census year).
SF3 data was collected via a longer, more detailed set of questions — the "long form" — which represents a sample of about one out of every six households in the U.S.