Glossary
Sampling frame
The list or source, such as a voter file or address database, from which a survey's respondents are actually drawn.
A sampling frame is the actual list, database, or contact method used to identify and reach potential survey respondents, such as a registered-voter file, a list of residential addresses, or a set of phone numbers, as distinct from the target population the researcher actually wants to describe.
A frame is good to the extent it fully and accurately covers the target population, without excessive duplication or ineligible entries. Frame error occurs whenever the frame's coverage diverges from the population it is meant to represent, for example a cellphone-only frame that misses landline-only households, or a voter file that lags behind recent movers and new registrants tied to a particular census tract. Random selection is then applied within the frame, so results are only as representative as the frame allows; people who had no chance of being selected cannot be recovered later.
A well-chosen frame can matter more than a larger sample size, because survey weighting can rebalance a sample's known demographics but cannot fix the absence of people the frame never had a chance of reaching. This is a distinct concept from sampling bias, the resulting distortion in who ends up responding, of which poor frame coverage is one major cause alongside nonresponse bias and a low response rate.
Last reviewed September 22, 2026