R help - Jan 2012 - computing scores from a factor analysis

Haj

i try to perform a principal component analysis by using a tetrachoric
correlation matrix as data input

tetra <- tetrachoric (image_na, correct=TRUE)
t_matrix <- tetra$rho
pca.tetra <- principal(t_matrix, nfactors = 10, n.obs = nrow(image_na),
rotate="varimax", scores=TRUE)

the problem i have is to compute the individual factor scores from the pca.
the code runs perfect if i do not ask for the scores

if i ask for the scores i get an error message
"Error in scale(x.matrix): object 'x.matrix' not found"

can somebody help me?

cheers
wolfgang

--
View this message in context:
http://r.789695.n4.nabble.com/computing-scores-from-a-factor-analysis-tp4306234p4306234.html
Sent from the R help mailing list archive at Nabble.com.

Wolfgang, 

Since you seem to be doing this in the psych package, it would have been faster
to directly ask the author (me).  Luckily, I saw the question on R-Help.
 

The principal components step is being done on the correlation matrix, not on
the raw data matrix, thus, it is not able to find scores.

However, since you have the components solution, you also the scoring weights.

Taking your analysis:
> tetra <- tetrachoric (image_na, correct=TRUE)
> t_matrix <- tetra$rho
> pca.tetra <- principal(t_matrix, nfactors = 10, n.obs = nrow(image_na),
> rotate="varimax", scores=FALSE)
scores <- image_na %*% pca.tetra$weights

Bill






On Jan 18, 2012, at 4:27 AM, wolfgang wrote:
> Haj
> 
> i try to perform a principal component analysis by using a tetrachoric
> correlation matrix as data input
> 
> tetra <- tetrachoric (image_na, correct=TRUE)
> t_matrix <- tetra$rho
> pca.tetra <- principal(t_matrix, nfactors = 10, n.obs = nrow(image_na),
> rotate="varimax", scores=TRUE)
> 
> the problem i have is to compute the individual factor scores from the pca.
> the code runs perfect if i do not ask for the scores
> 
> if i ask for the scores i get an error message
> "Error in scale(x.matrix): object 'x.matrix' not found"
> 
> can somebody help me?
> 
> cheers
> wolfgang
> 
> --
> View this message in context:
http://r.789695.n4.nabble.com/computing-scores-from-a-factor-analysis-tp4306234p4306234.html
> Sent from the R help mailing list archive at Nabble.com.
> 
> ______________________________________________
> 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.
> 
William Revelle		           http://personality-project.org/revelle.html
Professor			           http://personality-project.org
Department of Psychology   http://www.wcas.northwestern.edu/psych/
Northwestern University	   http://www.northwestern.edu/
Use R for psychology             http://personality-project.org/r
It is 5 minutes to midnight	   http://www.thebulletin.org

William,
I had a problem similar to Wolfgang and I solved it through your help.
Many thanks!

Just an observation which sounded strange to me ( I am not a statistician,
just a wildlife biologist)
I have noticed  that  running the pca using principal with raw data  (and
therefore using scores=TRUE in the command line) gives different pca scores
than running the same pca with the correlation matrix (using scores=FALSE in
the command line and therefore calculating the scores in the way you
suggested to Wolfgang). Is that normal?

Thanks

francesca




--
View this message in context:
http://r.789695.n4.nabble.com/computing-scores-from-a-factor-analysis-tp4306234p4372993.html
Sent from the R help mailing list archive at Nabble.com.