Regression after multiple imputation

[johnstatistic]

Hello every one. I want to perform simple binary logistic regression after multiple imputation but I constantly get this message and the execution of the command stops: For split file Imputation Number = 1 the dependent variable has more than two non-missing values. For logistic regression, the dependent value must assume exactly two values on the cases being processed. How can I solve this problem. Thank you in advance.

 

[hlsmith]

What program are you using?


Regardless, it seems like it is telling you that in a dataset the DV takes on more than two values. Perhaps ( ., 0 ,1)?

 

[janessa642]

depends on program..

 

[Camryn64]

What program are you using?


Regardless, it seems like it is telling you that in a dataset the DV takes on more than two values. Perhaps ( ., 0 ,1)?

can you explain with the choice of programs

 

[johnstatistic]

I have a binary DV(0,1). However, after MI it has more than these two values in iterations. Moreover, I do not get pooled data for some regressions. this is the message:For at least one model, pooled estimates could not be computed because model parameters vary by imputation. IK am working with SPSS last version.

 

[hlsmith]

Well I would imagine you did something wrong if you have "after MI it has more than these two values in iterations". Not familiar with MI in SPSS.