Sampling design effects: Do they affect the analyses of data from the National Survey of Families and Households?

Most large national surveys, such as the National Survey of Families and Ho useholds (NSFH), involve clustered and stratified samples. These complex sa mple designs have consequences for data analysis techniques. Standard error s calculated using procedures that do nor adjust for design effects often a re too small and lead to incorrect inferences. We discuss design effects an d estimate them for a set of variables selected from the 1988 NSFH. Include d are examples of descriptive estimates and regression results with househo ld income and marital happiness as dependent variables. Statistical softwar e that adjusts standard errors in complex designs is discussed, as are issu es related to weighting and the analysis of subsamples.