Premium
On the Use of Two‐stage Cluster Samples in Epidemiological Population Studies
Author(s) -
Chambless Lloyd E.
Publication year - 1988
Publication title -
biometrical journal
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.108
H-Index - 63
eISSN - 1521-4036
pISSN - 0323-3847
DOI - 10.1002/bimj.4710300310
Subject(s) - statistics , cluster sampling , sampling (signal processing) , rule of thumb , cluster (spacecraft) , population , variable (mathematics) , mathematics , variance (accounting) , sampling scheme , econometrics , sample (material) , sample size determination , multistage sampling , computer science , medicine , algorithm , environmental health , economics , mathematical analysis , chemistry , accounting , filter (signal processing) , chromatography , estimator , computer vision , programming language
This paper discusses the need to account for the sampling scheme in an analysis of epidemiological population data which were collected by a two‐stage cluster sample. An example is presented where for validity reasons one should generally account for the sampling scheme, though a rule‐of‐thumb is given to estimate the effect of not doing so. The example concerns one of the centers (Augsburg, F. R. G.) participating in the WHO MONICA Project, which was designed to study the relationship between changes in risk factor levels, as measured by several surveys in each center, and changes in cardiovascular incidence rates, as measured by a registry system for each center. Variance estimation methods which either account for or ignore the sampling scheme are compared for a particular sampling scheme. Easily computable upper bounds on the effect on variances of ignoring the sampling scheme are presented, both over all possible variables, and for a particular variable.