Error when imputing unordered categorical responses

Welcome to the forum for REALCOM users. Feel free to post your question about REALCOM here. The Centre for Multilevel Modelling take no responsibility for the accuracy of these posts, we are unable to monitor them closely. Do go ahead and post your question and thank you in advance if you find the time to post any answers!

Go REALCOM (Developing multilevel models for REAListically COMplex social science data) >>
Post Reply
Posts: 4
Joined: Thu Oct 02, 2014 2:03 pm

Error when imputing unordered categorical responses

Post by odukesucl »


I am trying to impute 2 categorical responses (1 level 1, 1 level 2) and one binary (level 1). The categorical responses are ordered, but I want to impute them as unordered in case proportionality does not hold. I have already tried imputing them all as ordered categorical variables, which seemed to work (due to time conisderations I stopped the imputation process before completion). However, when I try to impute them as unordered, I get an error message before the imputation starts (the error message is attached). Any idea why this may be occuring?

Error message screenshot.pdf
(144.36 KiB) Downloaded 387 times
Posts: 11
Joined: Thu Oct 02, 2014 11:02 am

Re: Error when imputing unordered categorical responses

Post by feifeibu »

I think my problem is rather similar. I have a two-level model with missing unordered categorical variables. The imputation seems to work just fine. But I get error messages when the imputation is done. And no imputed data file is saved. The error message is attached. Thanks!
(300.31 KiB) Downloaded 429 times
Posts: 1265
Joined: Mon Oct 19, 2009 10:34 am

Re: Error when imputing unordered categorical responses

Post by ChrisCharlton »

The read error may be due to a permissions problem. When running Realcom-Impute you need to either change the working directory in the shortcut to somewhere that you can write or run the software with an account with administrative permissions. I am not sure what would be causing the other problem, however if you are able to send an example replicating the problem to Harvey Goldstein or myself then we can look into it in more detail. You can find contact details here:
Post Reply