n. a procedure in which several independent estimated values of a population characteristic are averaged, with or without weights, to obtain a single value. For example, the pooled variance is a single value for a variable’s dispersion produced by combining several independent estimates of that dispersion. Consider a researcher who uses multiple imputation to fill in missing data when examining correlations between children’s educational performance and maternal depression. He or she could calculate the value of the variance using the data set that exists at each stage of the imputation procedure and then average each of those individual variances to obtain a single overall value.