Imputation of missing values and factor analysis.

[lei]

Hi,

i'm trying to impute data for a patient experience survey to conduct factor analysis. For HCAHPS, they mentioned in a paper:

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1361249/

"In case of missing values for items on the questionnaire with an ordered response format, we imputed values under Missing at Random assumptions (Rubin 1976) using SAS software (PROC MI). Such imputations might be interpreted as predictions of the responses that might have been expected from these respondents had they appropriately answered the skipped items, using relationships observed among those who actually answered those items."

I am wondering if anyone can point me to the code or a description of the code for this. I have the missing values imputed. In total, I did 5 of them. I want to average them and replace missing values in the original data sets.

Any help would be appreciated

[Reeza]

The documentation is a good place to start:

SAS/STAT(R) 9.2 User's Guide, Second Edition