ACS characteristics have a survey period
The American Community Survey publishes survey estimates, including population characteristics, with reference periods and margins of error. An ACS one-year or five-year product is not interchangeable with the Population Estimates Program's July 1 annual resident total. A current PEP headline does not make an older ACS panel current.
When selecting an ACS characteristic, identify the product, period, table and variable, geography, universe and margin of error. Median age, household population and household size use definitions that must be preserved. Dividing all residents by households does not correctly calculate average household size when the numerator includes people living in group quarters.
Match products and definitions
Compare the same product type and compatible periods. A one-year estimate and a five-year estimate describe different time coverage and reliability. Variable definitions, geographic boundaries and category systems can change, so a matching label is not sufficient proof of compatibility. Census race and Hispanic-origin categories may overlap unless the table explicitly supplies mutually exclusive groups.
For change analysis, Census guidance favors comparing nonoverlapping five-year periods where appropriate. Overlapping periods share years of observations, which complicates interpretation of apparent change. Check the specific product's comparison guidance and any restrictions for unusual years or geographic releases.
Uncertainty belongs in the analysis
Display the published margin of error where supplied, along with the estimate. An observed difference does not by itself establish a statistically meaningful change. Evaluating a difference may require a statistical test using the source's guidance, assumptions and standard-error calculations. Do not manufacture survey-style margins for a PEP total.
Derived shares and combinations can require their own uncertainty calculations. A denominator must describe the same universe as the numerator; otherwise, the calculated share can be misleading even before uncertainty is considered. Avoid treating a population stock, such as foreign-born residents, as a direct count of annual arrivals.
How this launch uses ACS
ACS observations are not loaded into this site's initial annual population series or demographic panels. This guide documents the selection rules for a later characteristics release. Where characteristics are unsupported, profiles explain the gap rather than borrowing a county figure or labeling an ACS observation as PEP.
Primary sources: Comparing ACS data and Choosing an ACS product. Use the measurement guide to distinguish totals, survey characteristics and future scenarios before building a combined table.