As long as you have a second factor with at least two indicators as well then the model should be identified. It would seem more likely that the model is misspecified if it is not converging. How many factors and indicators are there overall?
Paul R. Swank, Ph.D. Professor, Developmental Pediatrics Medical School UT Health Science Center at Houston -----Original Message----- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Irina Sent: Sunday, December 14, 2003 1:02 PM To: [EMAIL PROTECTED] Subject: LISREL - <3 indicators for a latent variable in CFA Hi there, I am running confirmatory factor analysis in LISREL and one of my factors has two indicators only. The model does not converge because of this so I am trying to figure out how to deal with that. I am running the whole thing as eta-lambda-y. I know that there is a way to have single indicator latent variables but the technique for those does not seem right for comfirmatory factor analysis. Any ideas? you can email me at iasATcmuDOTedu if you can't answer in this format. Thanks a LOT! Irina . . ================================================================= Instructions for joining and leaving this list, remarks about the problem of INAPPROPRIATE MESSAGES, and archives are available at: . http://jse.stat.ncsu.edu/ . ================================================================= . . ================================================================= Instructions for joining and leaving this list, remarks about the problem of INAPPROPRIATE MESSAGES, and archives are available at: . http://jse.stat.ncsu.edu/ . =================================================================
