[R] Factor Analysis-factanal function

[John Fox]

Dear Ricardo,

Factor scores are linear combinations of the original variables and
therefore to get factor scores, factanal() needs the data, not just the
correlation matrix among the variables. 

Perhaps what you want is the factor-score coefficient matrix. Since you're
apparently using a varimax rotation, which is orthogonal, you could compute
the factor-score coefficients from the correlation matrix among variables
and the factor loadings. To do this for an oblique rotation, you'd need the
factor-correlation matrix, which factanal() sadly and inexplicably doesn't
report.

All this is academic, however, because if you don't have the original data
from which to compute factor scores, what use are the factor-score
coefficients?

I hope this helps,
 John

------------------------------
John Fox, Professor
Department of Sociology
McMaster University
Hamilton, Ontario, Canada
web: socserv.mcmaster.ca/jfox


> -----Original Message-----
> From: r-help-bounces at r-project.org [mailto:r-help-bounces at r-project.org]
On
> Behalf Of "Ocaña Riola, Ricardo"
> Sent: January-30-09 5:30 AM
> To: r-help at r-project.org
> Subject: [R] Factor Analysis-factanal function
> 
> Dear friends,
> 
> 
> 
> I'm using R to produce the following Factor Analysis:
> 
> 
> 
> > matriz.cor<-hetcor(matrix(as.factor(data), ncol=variables,
> byrow=T))$correlations
> 
> > factanal(x=data, factors=2, covmat=matriz.cor, scores='regression')
> 
> 
> 
> Then the screen output shows the following message:
> 
> 
> 
> Error en factanal(x = data, factors = 2, covmat = matrix, :
> 
>   requested scores without an 'x' matrix
> 
> 
> 
> Do you know some additional function in order to produce a matrix of
scores
> when 'covmat' parameter is used?.
> 
> Thank you.
> 
> 
> 	[[alternative HTML version deleted]]
> 
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.