Home › Learn
How ACS Estimates Work: The Survey Behind the Data
This guide is part of the ZIP Code Demographics learning library, written by our editorial team to help readers use ZIP-level Census data accurately. It explains how ACS estimates work in plain language, with the same care we apply to the data pages themselves: no invented figures, no borrowed statistics, and a clear line between what the Census Bureau publishes and what it does not.
A survey, not a census
The American Community Survey is a continuous survey run by the U.S. Census Bureau. Instead of counting everyone once a decade, the ACS interviews a sample of addresses every month — about 3.5 million addresses a year — and asks questions about income, education, employment, housing and dozens of other topics the decennial census no longer covers. The responses are weighted to represent the whole population, producing estimates for the nation, states, counties, places, census tracts and ZIP Code Tabulation Areas.
Why ZIP-level data is always five-year data
A single year of sampling is too thin to describe a small area reliably, so the Census Bureau publishes two products: one-year estimates for areas with at least 65,000 residents, and five-year estimates that pool sixty months of responses to cover every geography down to the block group. ZCTA figures on this site always come from the five-year product. The five-year window is a strength and a caveat at once: pooling makes small-area estimates possible and stable, but it also means the figures describe the whole period — our current vintage describes 2020–2024 — rather than a single point in time.
What the ACS asks
The demographic tables behind this site draw on the ACS questions about age, sex, race, Hispanic origin, household relationship, educational attainment, employment status, income in the past twelve months, housing value, rent and tenure. Income figures are adjusted to the final year of the period so that responses from earlier years are comparable. Home value is the householder's own estimate of what the home would sell for, and gross rent includes utilities the renter pays — definitions that matter when comparing ACS figures with listing sites or tax assessments, which measure different things.
How to cite and check ACS data
Every figure on this site can be traced to a published Census Bureau table: our methodology page lists the exact table numbers, from B01003 for total population to B25077 for median home value. If a number matters to you — for a grant application, a business plan or a news story — verify it in the Bureau's own data tools before citing it, and cite it as an ACS estimate with its vintage, not as a count. That habit protects you, because estimates are revised with each annual release and the vintage is part of the fact.
Frequently Asked Questions
What should I know about how ACS estimates work?
The American Community Survey is a continuous survey run by the U.S. Census Bureau. Instead of counting everyone once a decade, the ACS interviews a sample of addresses every month — about 3.5 million addresses a year — ...
Where does the data on ZIP Code Demographics come from?
Every figure on this site comes from the U.S. Census Bureau's American Community Survey (ACS) 2020–2024 5-year estimates, published as table-based summary files and retrieved in October 2026. No figures are modeled or taken from commercial sources.
Data: U.S. Census Bureau, American Community Survey (ACS) 2020–2024 5-year estimates, retrieved October 2026. Source files and table numbers are listed on our methodology page; the underlying data is published at census.gov.