Having used the Multiple Imputation feature of SPSS 17, I have a question regarding F-values and pooled results.
When conducting a repeated measures ANOVA, the output reports the main and interaction effects for each of the multiple imputations (say for the five different imputations) but does not pool them, so I do not have a single F value, significance level etc to report, which would obviously be required for an article/report.
I have read that SPSS does not yet support the pooling of these results but I was wondering if there are any rules that I might use to do the pooling myself? Or is there a total different track that I should be taking? Any advice welcome,
Sarah
|