| Sumario: | A methodology is presented for minimizing the mean error variance-covariance component in studies with resource constraints. When designing measurement studies, two important statistical issues—power and measurement error—must be considered. Power is concerned with the number of observations to employ, and measurement error is concerned with the best method of allocating a fixed number of conditions to the different facets of observation. The mean error variance-covariance component is an estimate that takes into account both of these statistical issues. This method is illustrated with a one-facet multivariate design, and extensions to other designs are explored.
|
|---|