[R] Confirmatory Factor Analysis in Non-Normal case

John Fox jfox at mcmaster.ca
Wed Mar 23 02:27:25 CET 2005


Dear Liu,

I'm not aware of any way of doing what you want in R at present. You can
base the analysis on polychoric correlations, which can be computed using
the polycor package, and you can get consistent estimates of the factor
loadings and factor correlations from the sem packages, but the likelihood
won't be right. 

I hope that this is of some help,
 John

--------------------------------
John Fox
Department of Sociology
McMaster University
Hamilton, Ontario
Canada L8S 4M4
905-525-9140x23604
http://socserv.mcmaster.ca/jfox 
-------------------------------- 

> -----Original Message-----
> From: r-help-bounces at stat.math.ethz.ch 
> [mailto:r-help-bounces at stat.math.ethz.ch] On Behalf Of yyan liu
> Sent: Tuesday, March 22, 2005 7:24 PM
> To: r-help at stat.math.ethz.ch
> Subject: [R] Confirmatory Factor Analysis in Non-Normal case
> 
> Hi:
>   I am doing a confirmatory factor analysis now. In the 
> analysis, I have null hypothesis test which specify some 
> special structure for the loading matrix.
> And the alternative is there is no such special structure. 
> Then the log likelihood ratio test can be used. The problem I 
> have is my data comes from a questionnaire and all the 
> variables are discrete from
> 1 to 7, which makes the normality assumption a problem. I 
> wonder if there is any other test can do this job or at least 
> I can make some sensible transformation to make the variables 
> "normal"?
>   Thank you very much!
> 
> liu
> 
> ______________________________________________
> R-help at stat.math.ethz.ch mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide! 
> http://www.R-project.org/posting-guide.html




More information about the R-help mailing list