I have a dataset with a large amount of missing data that I want to analyze in SPSS using generalized linear mixed models via the GENLINMIXED procedure. I am planning to use multiple imputation to address the missing data issue. However, it seems that the GENLINMIXED procedure ignores "split file" commands and as such would not provide pooled estimates from the multiply imputed datasets. Therefore I am looking for a way to override the problem with "split file" and the GENLINMIXED procedure. I was wondering if anyone knows how to do this. Thanks!