Zero cells in path analysis PreviousNext
Mplus Discussion > Structural Equation Modeling >
Message/Author
 Sonia posted on Wednesday, April 29, 2015 - 12:16 pm
Both my outcome variable Y and the covariate X are dichotomous. There is one cell in the 2X2 table that is equal to zero. The chi-square is significant.

But when I run the path model that includes Y on X (and some other variables), X is not significant and I think this is because of the zero cell. Is that correct.

Conceptually, I need this variable in the model. What can I do?
 Bengt O. Muthen posted on Wednesday, April 29, 2015 - 5:53 pm
I don't think you want a zero cell for 2 such important variables in your model. I can only think of gathering a larger sample, or say that questions regarding the X impact cannot be answered in this sample.
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: