Correlation between factors PreviousNext
Mplus Discussion > Confirmatory Factor Analysis >
Message/Author
 Janine Swain posted on Tuesday, June 05, 2012 - 7:47 pm
This will seem like a really silly question to you, but here goes ...

I am running a pretty simple CFA with two continuous factors with three indicators each. In the output section, I ask for standardized results and all modification indices.

The fit looks pretty good. CFI is 0.979 and SRMR 0.043. So its looking good according to Hu and Bentler.

I am worried that the factors might be highly correlated (I hope they're not), but want to be sure. Where can I determine this in the output? What is a correlation high enough for me to consider a second-order construct?
 Janine Swain posted on Tuesday, June 05, 2012 - 8:31 pm
BTW, Here's my input:

TITLE: CFA MODEL 2
DATE:
VARIABLE: Long List
USE VARIABLES ARE
Benev1
Benev2
Benev4
Reli1
Reli2
Reli3;
ANALYSIS: TYPE = general;
MODEL: BENEV BY Benev1@1 Benev2 Benev4;
RELIAB BU Reli1@1 Reli2 Reli3;
BENEV WITH RELIAB;
OUTPUT: STANDARDIZED; MODINDICES (ALL);
 Linda K. Muthen posted on Wednesday, June 06, 2012 - 4:49 pm
Please restrict posts to one window.

See TECH4 for factor correlations. You cannot have a second-order model with only two first-order factors.
 Patchara Popaitoon posted on Tuesday, June 12, 2012 - 11:46 am
Dear Linda,

I would like to know if Mplus can provide the significance level or p-value for the correlations reported in TECH4. Thanks.
Pat
 Linda K. Muthen posted on Tuesday, June 12, 2012 - 12:15 pm
No, we don't give standard errors for TECH4.
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: