Pooling data after MI

[h.sumnall]

Dear Colleagues, I have imputed missing values using the SPSS (v17)
multiple imputation technique. In accordance with the missing values
manual my data is split by Imputation_ and in SPSS Edit>Options i have
requested that Pooled results are shown in the output.

However after running my required repeated measures ANOVA my output
shows the results of each imputation, but not the pooled results.
Manipulating the pivot table in accordance with the manual does not
allow me to view the pooled results, only all imputations
independently or all together.

I wonder if anyone has faced this problem before? Using the tutorial
data and running a MLR i was able to show pooled data as expected, but
not it seems with this method.

All help appreciated,

Thanks

Harry

Dr Harry Sumnall
Reader in Substance Use
Centre for Public Health
Liverpool John Moores University
Kingsway House
Hatton Garden
Liverpool
L3 2AJ

Tel: +44 151 231 8096
Fax: +44 151 231 8020
http://www.cph.org.uk/harrysumnall

[Bruce Weaver]

Harry posted the same question to the SPSSX-L mailing list, and has no
doubt seen this response posted there. But for the benefit of those
who do not read SPSSX-L, here it is.

--- Response in SPSSX-L archives ---

Dear Harry,

To date, no combination rules for MI in (repeated measures) ANOVA or
MANOVA have been defined. For ANOVA you may resolve the problem by
using
a multilevel model in Mixed models, where repeated measures are nested
within subjects. SPSS can give pooled results within Mixed models. For
MANOVA there is not really a solution at hand yet.

Best regards,

Joost van Ginkel

--
Bruce Weaver
bwe...@lakeheadu.ca
http://sites.google.com/a/lakeheadu.ca/bweaver/
"When all else fails, RTFM."