CFA and Rasch model PreviousNext
Mplus Discussion > Confirmatory Factor Analysis >
 Seunghee Kook posted on Sunday, March 25, 2007 - 6:20 pm

I analyzed a data using CFA with Mplus and then analyzed it using Rasch model(1PL) with WINSTEPS. However, I heard that CFA in Mplus is similar to a two parameter IRT model. So, did I make a mistake?

Thank you in advance.
 Linda K. Muthen posted on Monday, March 26, 2007 - 8:46 am
If you want a Rasch model in Mplus, you should hold the factor loadings equal and set the metric of the factor by fixing the factor variance to one.
 Andreas Schulz posted on Wednesday, February 19, 2014 - 6:38 am

I'm doing a 3-dimensional Rasch analysis with MPlus, please can you give me a hint for this question:

How do I get any reliability indicators such as PSI (person separation index, cf. RUMM2030 software) oder EAP/ PV (cf. Conquest software)?

Thank you very much in advance,
 Bengt O. Muthen posted on Wednesday, February 19, 2014 - 8:52 am
Those statistics are not given in Mplus.
 Andreas Schulz posted on Thursday, February 20, 2014 - 7:31 am
Thank you for your answer!

How is item fit (according to the Rasch model) evaluated in MPlus?

I found standardized residuals that might indicate significant misfit when they exceed +/- 1,96. Is that correct?
Are there other fit statistics concerning the Rasch model in MPlus?

Thank you, Andreas
 Linda K. Muthen posted on Friday, February 21, 2014 - 8:25 am
The TECH10 residuals you are looking at are the best way to assess model fit with categorical items and maximum likelihood estimation. There are no absolute fit statistics.
Back to top
Add Your Message Here
Username: Posting Information:
This is a private posting area. Only registered users and moderators may post messages here.
Options: Enable HTML code in message
Automatically activate URLs in message