Mplus Discussion >> SEM outcome variable with zero scores

Topics
Last Day
Last 3 Days
Last Week
Tree View

Edit Profile


SEM outcome variable with zero scores

Mplus Discussion > Structural Equation Modeling >

Message/Author

Jordan Marwood posted on Thursday, June 01, 2017 - 3:23 am

Hi,

this is my first time posting and I'm very new to SEM so please be kind!

I was wondering if anyone knows if SEM will allow zero entries for an outcome variable. For example if I am looking at how various factors are associated with binge eating, with binge eating frequency as the outcome variable (continuous), could I include participants who have not endorsed binge eating (e.g. have a binge frequency of 0) as part of the analysis?

Thanks in advance

Bengt O. Muthen posted on Thursday, June 01, 2017 - 6:57 pm

Yes, this is perfectly fine: see Chapter 7 of our new book on censored variables - there are many modeling alternatives. We give a workshop on the book in August.

Jordan Marwood posted on Monday, June 05, 2017 - 8:28 am

Thanks for the response. Will the model tolerate high numbers of zero entries in comparison to the number of participants who have endorsed binge eating?

Can you please provide the title of the book you are referring to?

Thanks

Bengt O. Muthen posted on Monday, June 05, 2017 - 3:45 pm

Q1: Yes

Q2: See

http://www.statmodel.com/Mplus_Book.shtml