|
|
Measurement model - selected observed... |
|
Message/Author |
|
|
Dear Linda, I would like to know if there is any rules that I can follow if I want to use only some of the observed variables to form latent factor. The thing is I have about 15-20 survey items for each factor and I can't use them all to specify the model, becuase this will cause a poor fit model. Please adivse. thanks. |
|
|
It sounds like not all of the factor indicators measure the factors well. I would do an EFA and use the factor indicators that have high loadings on their factor and low loadings on the other factors. |
|
Back to top |
|
|