Montecarlo power simulation for a bif... PreviousNext
Mplus Discussion > Categorical Data Modeling >
Message/Author
 Max  posted on Friday, August 21, 2015 - 5:17 am
Dear Prof. and dr. Muthén,

With great interest I have read your article “How to us a monte carlo study to decide on sample size and determine power”. At the moment I am running a montecarlo simulation for my own model and since it contains categorical variables, I have used the method and syntax offered by Myers et al. I have adapted the syntax to my own model, everything works fine and I get a decent power.

However, one of the models that I am studying, is a bifactor model. I would also like to run a montecarlo simulation the bifactor model, but I struggle with the thoughts the possibility of running a simulation and the interpretation of it. In a bifactor model all the factors are forced to be orthogonal, therefore how would I interpret the power? (Since in the analysis of a correlated model one looks at the % sig column for the correlating factors.)

Therefore I would like to ask you whether you think it is possible to run a power-estimation montecarlo simulation for a bifactor model that contains categorical variables.
And if it would be possible, how would one interpret its power?
Thank you in advance for answering these questions.

Max
 Bengt O. Muthen posted on Friday, August 21, 2015 - 1:51 pm
You have to ask yourself what you want the power of.

This general analysis question may be better suited for SEMNET.
Back to top
Add Your Message Here
Post:
Username: Posting Information:
This is a private posting area. Only registered users and moderators may post messages here.
Password:
Options: Enable HTML code in message
Automatically activate URLs in message
Action: