|
|
| Message/Author |
|
|
| Zi Yan posted on Sunday, June 04, 2017 - 10:45 pm
|
|
|
Dear Prof. Muthén,
I want to combine Rasch model and SEM in my analysis. Specifically, I want to use Rasch analysis to model the three latent variables A, B, and C; and then to examine the relationship among them (A and B predict C). I learn from your previous posts that “To get a Rasch model, you need to hold the factor loadings equal and fix the factor variance to one.” Could you please advise whether the following Mplus control lines are correct? Thank you!
VARIABLE: NAMES ARE a1-a5 b1-b5 c1-c5;
CATEGORICAL ARE a1-a5 b1-b5 c1-c5;
MODEL: A BY a1-a5* (1);
A@1;
B BY b1-b5* (2);
B@1;
C BY c1-c5* (3);
C@1;
C ON A B; |
|
|
|
|
| This is correct. |
|
|
| Zi Yan posted on Monday, June 05, 2017 - 5:38 pm
|
|
|
| Thank you very much! |
|
|
| Back to top |
|
|
