Mplus Discussion >> Missing Items needed to compute total score

Topics
Last Day
Last 3 Days
Last Week
Tree View

Edit Profile


Missing Items needed to compute total...

Mplus Discussion > Missing Data Modeling >

Message/Author

Nicole Tarter posted on Thursday, February 19, 2015 - 10:38 am

Is it possible to run FIML and then use DEFINE to compute the new variable (total score)?

I am running a double mediator model, but my variables are based on total scores of certain questionnaires, i.e PSWQ, PCL, BDI.

Bengt O. Muthen posted on Thursday, February 19, 2015 - 2:25 pm

I don't understand the question - how does computing a total score relate to running FIML?

Nicole Tarter posted on Thursday, February 19, 2015 - 2:47 pm

I need to fill in my missing data before I can compute total scores. Is this possible?

Linda K. Muthen posted on Thursday, February 19, 2015 - 3:01 pm

FIML does not create values for missing values. The DEFINE command works with the observed variables directly. It does not work in conjunction with FIML.

Nicole Tarter posted on Thursday, February 19, 2015 - 3:18 pm

Is there a way in Mplus to create values for missing data?

Bengt O. Muthen posted on Thursday, February 19, 2015 - 3:58 pm

See our Multiple Imputation examples in Chapter 11 of the UG, starting with 11.5.

Nicole Tarter posted on Tuesday, February 24, 2015 - 4:35 am

When I ran the multiple imputation it only saved to the one file Name I requested. From what I read it was supposed to save as many data sets as I asked to impute, which was 5, each with the name I put in the syntax and a number to signify which imputation it was. Am I missing something? The one data set that did save is filled in.

Linda K. Muthen posted on Tuesday, February 24, 2015 - 7:36 am

Please send the output and your license number to support@statmodel.com.