Abstract:
|
One approach to assessing the effect of changes in a survey's sample size and/or the number of primary sampling units (PSU) is by modeling. Another approach, if prior survey data are available, is to simulate the changes using the data, producing estimates and standard error estimates, and repeating the process. We discuss our research using National Health and Nutrition Examination Survey data and simulations to estimate the effect (change in effective sample size) of reducing the nominal sample size and/or increasing the number of PSUs. We compared our results to the effect previously estimated using modeling.
|