Correcting for selective nonresponse in the National Longitudinal Survey of Youth using multiple imputation

Survey attrition and nonresponse, particularly when selective. present uniq ue challenges to researchers interested in studying developmental processes and longitudinal change, Four distinct patterns of nonresponse on children 's psychosocial adjustment and lifetime poverty experiences and family hist ories are identified using principal components analysis. In turn, membersh ip in these four groups is significantly predicted by the child's demograph ic characteristics, family experiences, and previous values on adjustment v ariables, indicating selective nonresponse and raising the possibility of b iased estimates based on listwise deletion of missing data. We then examine a set of latent growth curve models that interrelate children's family exp eriences and psychosocial adjustment using listwise deletion (LD) and multi ple imputation (MI) procedures. Implications for treatment of nonresponse i n national longitudinal surveys are discussed.