A procedure for combining the advantages of the regular cross-sectional survey with the panel study, in national statistical surveys measuring trends over time, such as the USA National Crime Survey and certain Labour Force Surveys. A proportion of the initial survey sample is dropped from each subsequent interview sweep and replaced with a fresh sample of equal size. The complete sample is thus replaced gradually over a period of months or years. A variety of rotation designs are in use, some of them complex. For example, the USA Current Population Survey uses a 4–8–4 rotation system, ensuring that 75 per cent of the sample is common from month to month, and 50 per cent of the sample is common from year to year.