Re: Q: dichotomous variables

Hi Erkii,

certain things in mind. Gorsuch (Factor Analysis 2nd Ed. , 1983) and
Rummel (Applied Factor Analysis, 1970) give such examples and present
some difficulties but do not deny the use of dichotomous variables
categorically. I know that LISREL, Mplus and other programs offer
sophisticated solutions to the problem.

You have the option to factor analyze the matrix of tetrachoric
correlation coefficients. I have a free program, TetMat, much simpler
than LISREL or Mplus for this:

http://ourworld.compuserve.com/homepages/jsuebersax/tetra.htm#soft

Factor analysis of tetrachoric correlation coefficients is discussed
here:

http://ourworld.compuserve.com/homepages/jsuebersax/sem.htm

Note than when Rummel and Gorsuch wrote their books, tetrachoric
correlation coefficients were considered difficult to calculate. Now,
with more software and better computers, that is trivial. In 1970,
people were doing factor analysis with manual calculations!

HTH

John Uebersax PhD

.