Guidebook for Imputation of Missing Data. Technical Report No. 17.

This guidebook is designed for data analysts who are working with computer data files that contain records with incomplete data. It indicates choices the analyst must make and the criteria for making those choices in regard to the following questions: (1) What resources are available for performing the imputation? (2) How big is the data file? (3) What is the purpose for imputing missing data? (4) What structures exist in the recorded variables? (5) What is the pattern of missing data? (6) What assumptions are acceptable for this imputation? Answers to these questions constitute recommendations for imputation procedures. Several alternative recommendations and the conditions that determine the appropriateness of use are considered. The final section of the guidebook contains instructions for using PROC IMPUTE created by the Statistical Analysis Group in Education for the National Center for Education Statistics, and for interpreting its results. Appendices include: (1) Processing Time by Numbers of Variables and PROC IMPUTE; and (2) Sample Statistical Analysis System Program to Reweight for Total Nonresponse. (JAZ)