Many surveys are repeated on several occasions, and the associated estimates are used to analyze changes in variables over time. Major social and economic surveys, such as labor force and retail trade surveys (RTS), are conducted monthly or quarterly to identify changes in the level or rate of change of variables, including turning points. Many repeated surveys involve overlap in the sample between different time periods. The sample overlap induces a correlation structure in the sampling errors of the time series of estimates, which affects the analysis of changes in them and may be exploited in producing estimates. In a panel or longitudinal survey, an initial sample is selected and at each occasion that the survey is conducted an attempt is made to include all the members of the initial sample, even if they move. Longitudinal surveys are developed to permit analysis of changes at the individual level. In a repeated survey, there is not necessarily any overlap of the sample for the different occasions. When the emphasis is on estimates for the population and major subpopulations, an independent sample may be used on each occasion, which is often the case when the interval between the surveys is large. © 2009, Elsevier Inc. All rights reserved.
Steel, D., & McLaren, C. (2009). Design and Analysis of Surveys Repeated over Time. Handbook of Statistics, 29(PB), 289–313. https://doi.org/10.1016/S0169-7161(09)00233-8