SRC error

[scott.raynaud]

I have a data set which I've run imputation on several times successfully.  Now, I'm suddenly getting the following error when running imputation:

 

Error: Character variable CPLKIND not categorial

I find this odd since the variable is defined as a numeric transfer variable.  The data file was a csv exported from SAS 9.4 (I changed the extension to txt as in the past) and CPLKIND was defined as numeric in the metadata step and transfer in the impute step.  I can't seem to figure out how to solve this problem.

 

[scott.raynaud]

I was able to get this to run by passing the SAS file thorugh StatTransfer to convert it to text. I had to declare CPLKIND as integer to get IVEWARE to run. That's very curious to me.

I've noticed another problem in the run log. I have set the following restriction:

restrict CASHYR(CASHMEV=1) IHMOBYR(IHMOB=1);

I have also run this restriction several times without problem, but now it's indicating no cases. Crosstabs of CASHYR with CASHMEV reveals 21234 cases and IHMOBYR with IHMOB 3156 cases, so there are cases present. I wonder if other variables are coming inot play or if my file is somehow corrupted. Thoughts?

 

[Trivellore Raghunathan]

If CPLKIND has more than 20 categories then it will give you the kind of error you saw the first time. It is described in the manual. It assumes that you have declared a continuous or count variable as categorical by mistake.

The second error may be that all cashyr might be missing with cashmev=1 or ihmobyr=missing when ihmob=1.  Other possibility is that the data is corrupted. 

Raghu

--
You received this message because you are subscribed to the Google Groups "IVEware" group.
To unsubscribe from this group and stop receiving emails from it, send an email to iveware+u...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

--

Trivellore Raghunathan (Raghu)

Research Professor, Survey Research Center

Institute for Social Research

Professor, Department of Biostatistics

University of Michigan

Phone: (734)-647-4619

"A good life is filled with selfless actions full of compassion knowing well that we are all one"

[scott.raynaud]

Thanks for the feedback.  Just interesting that I never encoutered this error before, but StatTransfer seems to have solved the problem.

 

There were definitely non-missing cases for the restrict statement, so, as before, I used StatTransfer to configure those variables as integers and then declared them numeric in the metadata procedure in IVEWARE.  This sees to have fixed the problem.