Skip to content

Tetrachoric and polychoric ceofficients (for sem) - any tips?

2 messages · Michael Dewey, John Fox

#
About two years ago there was a thread about this which suggested that at 
that time nobody had these coefficients ready to go.
(a) has anyone in the meanwhile programmed them?
(b) I think I can see how to do the tetrachoric one with mvtnorm on similar 
lines to an example on the help page so will try that if nobody else 
already has
(c) looking at the polychoric one makes me realise yet again that I wish I 
knew more mathematics. I would also find it difficult to test it as I do 
not have any worked examples. Does anyone have any tips about how to 
program it and how to test the resultant code?
(d) I appreciate this last item is not strictly an R question, but my 
intention is to use these as input into the sem package for structural 
equation models. If anyone thinks that is misguided I would be intersted to 
here.


Michael Dewey
m.dewey at iop.kcl.ac.uk
#
Dear Michael,

I'm not aware of pre-existing R code for tetrachoric or polychoric
correlations. I may at some point incorporate such functions into the sem
package but I don't have concrete plans for doing so. On the other hand, I
don't think that it would be very hard to do so. (A discussion, references,
and an example are in Kotz and Johnson, eds., Encyclopedia of Statistics,
Vol 7.)

Tetrachoric and polychoric correlations are estimates, and in subsequently
estimating a SEM from these, one should take that into account. 

Regards,
 John

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