Dear people,
i want to conduct a LCA (latent class analysis) using -gllamm in Stata (14.1).
The purpose of this is, to validate a psychometric scale. On a theoretical level, four latent variables are assumed, which are categorical (yes/no). I have 13 observed variables which are ordinal ( 7-point-likert scales).
This is the basic model I want to test:
y1 y2 y3 ← c1 (c=latent class; y=observed variable)
y4 y5 y6 ← c2
y7 y8 y9 ← c3
y10 y11 y12 y13 ← c4
How is this possible using -gllamm in Stata? I tried to find examples which suit my purpose, but I couldn't.
I studied the manual and did find some references to LCA/Gllamm (eq. http://www.gllamm.org/faqs/models/lcmfit.html), but I didnt help me.
Best regards,
Anton
i want to conduct a LCA (latent class analysis) using -gllamm in Stata (14.1).
The purpose of this is, to validate a psychometric scale. On a theoretical level, four latent variables are assumed, which are categorical (yes/no). I have 13 observed variables which are ordinal ( 7-point-likert scales).
This is the basic model I want to test:
y1 y2 y3 ← c1 (c=latent class; y=observed variable)
y4 y5 y6 ← c2
y7 y8 y9 ← c3
y10 y11 y12 y13 ← c4
How is this possible using -gllamm in Stata? I tried to find examples which suit my purpose, but I couldn't.
I studied the manual and did find some references to LCA/Gllamm (eq. http://www.gllamm.org/faqs/models/lcmfit.html), but I didnt help me.
Best regards,
Anton
