The usual MI inferences are based on adding imputation variation to the squared standard error. But some exact small-sample procedures produce confidence intervals that are not based on standard errors.
Are there guidelines for adjusting such confidence intervals using multiple imputation? Best wishes, Paul von Hippel Paul von Hippel Statistician Department of Sociology / Initiative in Population Research Ohio State University 300 Bricker Hall 190 N. Oval Mall Columbus OH 43210 614 688-3768
