Choose a survey period for a community health indicator

A county health team wants to describe households without a vehicle. Decide first whether the question is about recent conditions in a large area or a stable measure for a small area. The American Community Survey (ACS) five-year estimate describes the full collection period. It is not an estimate for the final December, nor an arithmetic average of five published one-year percentages.

Make a period decision you can defend

1. Write the target population, geography, collection interval and decision date. A household transportation measure has a different denominator from a person-level measure.
2. Check product availability for every required geography. A standard one-year estimate generally requires an area of 65,000 people or more; supplemental tables have different coverage. Do not substitute a nearby city for a missing neighborhood.
3. Inspect both the estimate and its margin of error. A longer collection period can improve precision while obscuring a sudden change. Decide which tradeoff matters for this decision.
4. Save the product, period and universe alongside the number. Keep like-length periods when comparing releases and check for overlap.

Hypothetical decision

An outreach planner needs estimates for three tracts and their county. Only the county has the desired one-year table. Use the same available five-year product for all four areas; put the county’s one-year result in a separately labeled recent-conditions note if useful. A 2020–2024 window and a 2021–2025 window share four collection years. Their difference is not a clean before-and-after contrast.

If a service started in 2024, label an estimate spanning 2020–2024 as mixed exposure. A narrower research question or another data source may be more defensible than assigning the whole period to the program.

Primary source checked 7 November 2026: U.S. Census Bureau: Choosing ACS one-year or five-year estimates.

Use this in your study: Survey period comparison screen · Download: Survey product decision record.