SPSS Mult. Imputation "Mim" function?

[havingmycake]

Hi All,

After trying out Stata's multiple imputation commands, I've most recently used the multiple imputation tool in SPSS. I am trying to get my standard errors for my pooled model across multiple imputations.
Does a function exist in SPSS similar to 'mim' in Stata where you can get the statistics across multiply imputed models? Any tips/tricks would be much appreciated.

-El

[Alex Reutter]

When you build a model on multiply imputed data, the pooled output appears in the output of the procedure that's building the model.  See:

        http://publib.boulder.ibm.com/infocenter/spssstat/v20r0m0/topic/com.ibm.spss.statistics.cs/mi_telco_intro.htm

for a brief end-to-end example of producing multiply imputed results and analyzing it in SPSS Statistics.  If you want to skip past the production of the multiply imputed dataset, start here:

        http://publib.boulder.ibm.com/infocenter/spssstat/v20r0m0/topic/com.ibm.spss.statistics.cs/mi_analyze_telco_howto.htm


Note that not all procedures support multiple imputation.  There's a full list here: http://publib.boulder.ibm.com/infocenter/spssstat/v20r0m0/topic/com.ibm.spss.statistics.help/mi_analysis.htm

Alex



From:        havingmycake <[hidden email]>
To:        [hidden email]
Date:        12/12/2011 11:29 PM
Subject:        SPSS Mult. Imputation "Mim" function?
Sent by:        "SPSSX(r) Discussion" <[hidden email]>

---

Hi All,

After trying out Stata's multiple imputation commands, I've most recently
used the multiple imputation tool in SPSS. I am trying to get my standard
errors for my pooled model across multiple imputations.
Does a function exist in SPSS similar to 'mim' in Stata where you can get
the statistics across multiply imputed models? Any tips/tricks would be much
appreciated.

-El