A local analyst prepares two poverty figures for the same state and finds they do not match. One comes from the Current Population Survey, or CPS; the other comes from the American Community Survey, or ACS. That difference does not automatically mean one number is wrong. The surveys are separate instruments with different designs and uses, so the first task is to identify what each estimate measures before comparing the results.
A poverty rate depends on the survey
The Census Bureau publishes poverty data from several household surveys and programs, and provides guidance for choosing among them U.S. Census Bureau. Its current materials distinguish the CPS Annual Social and Economic Supplement, the ACS and the Survey of Income and Program Participation. The fact that all can produce poverty estimates does not make them interchangeable series.
The CPS ASEC is the basis for the Census Bureau’s national poverty report, while the ACS supports estimates for states and metropolitan areas, among other geographies. The Bureau’s 2025 state and metro brief uses 2024 and 2025 ACS one-year estimates; its 2025 national poverty report uses information collected in the 2026 and earlier CPS ASEC U.S. Census Bureau. A reader comparing publications therefore needs to check the survey and the data period, not only the place name and the word “poverty.”
Questionnaires and samples shape the result
Survey programs differ in questionnaire length and detail, the number of households included and methodology, according to the Census Bureau’s income overview U.S. Census Bureau. Those design choices affect what data are collected and how estimates are produced. A figure is an estimate from a particular survey system, not a complete count of every household in a state or metro area.
The reference period and intended use also matter. A statistic built from one survey’s collection and reporting process may answer a different policy or descriptive question from a statistic built by another. The Census Bureau explicitly directs data users to guidance on the differences among poverty sources and on choosing the right estimate for a need U.S. Census Bureau. Without examining those definitions, a side-by-side comparison can make two valid but distinct measures look like competing answers to precisely the same question.
Choose the source for the decision
Suppose a hypothetical state analyst is preparing a statewide briefing and also needs a metro-level comparison. The analyst should first identify the geography and population needed, then select a survey estimate designed to support that use. The ACS state and metro publication is an obvious starting point for those areas; a national poverty trend question may call for the CPS ASEC series used in the national report U.S. Census Bureau. This is an example of source selection, not a claim that either survey always answers every question best.
The analyst should state the survey, period and geography alongside the estimate. If a report changes from CPS to ACS, it should mark the change instead of presenting the new point as a seamless continuation. Separate figures can be placed next to each other for context, but should not be averaged, spliced or described as one series unless the Census Bureau provides a method that supports that operation. The two surveys’ different sample sizes and methodologies are reasons for care, not evidence that one must be discarded U.S. Census Bureau.
Read the measure without overclaiming
The official poverty concept itself is based on a comparison: a family is considered in poverty when its total income is below the official threshold for a family of that size and composition U.S. Census Bureau. That definition does not by itself explain every aspect of material hardship, and it does not turn an area rate into a description of every resident. A statewide estimate summarizes people covered by the chosen method; it cannot identify a particular household’s circumstances.
For readers, the practical checklist is modest: locate the survey name, data year or reference period, geographic coverage and intended use. Then compare like with like across time or places. When two numbers differ, ask whether the surveys gathered information in different ways or covered different populations and periods. The Census Bureau maintains several poverty series precisely because a single estimate cannot serve every question. Clarity about the measure makes a disagreement informative rather than confusing.
Sources & further reading
- Poverty ↗U.S. Census Bureau
- Income in the United States ↗U.S. Census Bureau
