A CASE-STUDY ON THE USE OF MULTIPLE IMPUTATION

Multiple imputation is a relatively new technique for dealing with mis sing values on items from survey data. Rather than deleting observatio ns for which a value is missing, or assigning a single value to incomp lete observations, one replaces each missing item with two or more val ues. Inferences then can be made with the complete data set. This pape r presents an application of multiple imputation using the 1987-1988 N ational Survey of Families and Households. We impute several binary in dicators of whether the respondent's elderly mother/mother-in-law is m arried. Descriptive statistics are then presented for the sample of ad ult children with an unmarried mother or mother-in-law.