CFA of Ordinal Data via Adaptive Quad... PreviousNext
Mplus Discussion > Confirmatory Factor Analysis >
Message/Author
 Fred B. Bryant posted on Monday, May 12, 2014 - 4:17 pm
I have four questions concerning the use of adaptive quadrature (AQ) to conduct confirmatory factor analysis of ordered categorical data via Mplus 7.

1. When using AQ to estimate CFA models with ordered categorical data, which specific link functions are available in Mplus 7? For example, does Mplus 7 offer probit, logit, log-log, and complimentary log-log link functions?

2. What options are available to handle missing data, when using AQ to conduct CFA with ordered categorical data via Mplus 7? Does Mplus 7 offer full-information estimation with missing data?

3. Is it possible to estimate multigroup CFA models when using AQ to analyze ordered categorical data via Mplus 7?

4. When using AQ to conduct CFA with ordered categorical survey data, can Mplus 7 implement weights? Specifically, can Mplus 7 implement both case-wise and design-wise weights when running AQ analyses?
 Bengt O. Muthen posted on Monday, May 12, 2014 - 8:33 pm
1. Probit and logit

2. Yes, FIML (that is, MAR) as well as NMAR (see my 2011 Psych Methods paper).

3. Yes, via type=mixture with knownclass.

4. Yes, both.

See also the paper on our website:

Muthén & Asparouhov (2013). Item response modeling in Mplus: A multi-dimensional, multi-level, and multi-timepoint example.
 Fred B. Bryant posted on Monday, May 12, 2014 - 8:49 pm
Thanks very much for your rapid and helpful answers to my questions.
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: