Also, Amelia doesn't always produce that odd result, but it tends to do it more frequently for some variables than it does for others.  I'm happy to provide data, scripts, etc. 


On Sat, Oct 17, 2009 at 10:25 AM, Donald Braman <dbraman@law.gwu.edu> wrote:
Hi Folks, 

I'm getting some odd levels in my post-imputation data: 

> levels(as.factor(a.out$imputations[[1]]$facts_convict))
[1] "1" "1" "2" "3" "4" "5" "6"

The "facts_convict" variable is imputed as ordinal, but the double "1"s aren't there prior to imputing: 

> levels(as.factor(my.data$facts_convict))
[1] "1" "2" "3" "4" "5" "6"

Also, I can't seem to cure this by converting to numeric and back: 

> levels(as.factor(as.numeric(a.out$imputations[[1]]$facts_convict)))
[1] "1" "1" "2" "3" "4" "5" "6"

Not sure how/why the double "1" crept in during imputation.  As you can imagine, ologits end up being odd, with two estimates for "1".  Any thoughts? 

Don

Donald Braman
phone: 413-628-1221
http://www.culturalcognition.net/braman/
http://ssrn.com/author=286206
http://www.law.gwu.edu/Faculty/profile.aspx?id=10123