Abstract:
|
Multiple imputation is a commonly used method to deal with incomplete data sets and is used by researchers on many different analytical levels. Imputation substitutes missing data with some values instead of discarding the entire case from the analysis. Multiple imputation based on log-linear modeling is a useful way to solve missing-data problems for categorical variables (Schafer 1997), though log-linear models have limitations especially when dealing with large data sets with more than a few incomplete categorical variables because of sparseness issues. . The latent class model is a plausible multiple imputation tool to solve this problem (Vermunt 2008), as is hot-deck imputation (Rubin 1987). The latter does not rely on model fitting for the variable to be imputed, and thus is potentially less sensitive to model misspecification than are imputation methods based on a parametric model, such as with regression imputation (Andridge & Little 2010). In this study, we are proposing a restricted latent class model as a multiple imputation tool and comparing it with the unrestricted latent class model and the hot-deck imputation method.
|
ASA Meetings Department
732 North Washington Street, Alexandria, VA 22314
(703) 684-1221 • meetings@amstat.org
Copyright © American Statistical Association.