[R] Factor Analysis?

John Maindonald john.maindonald at anu.edu.au
Thu Feb 3 00:01:46 CET 2000

Peter Dalgaard wrote:

> Erm, while there is a certain truth to what you're saying, in that
> factor analysis is often used by people who find anything non-GUI too
> difficult, I don't think that is a valid cause for further supporting
> that state of affairs... Such users will have no grasp of what a
> factor analysis means and will likely overinterpret the results. 
> There are good reasons to support factor analysis and other variants
> of "structural equations modelling" and there are also good reasons to
> make R more accessible to people with simple statistical needs, but
> oversimplifying complicated models is a real danger. (And I have
> several scary stories of the "I just filled in the forms the way it
> seemed to be intended" variety, mostly with SPSS users.)

One could easily get the impression that these sorts of use are almost
the only ways in which factor analysis that goes beyond principal
components kinds of uses, and structural equation modelling, are used.
It is very hard to find examples of the use of these techniques where
there is some genuine and plausible attempt at model validation.  It
is also very hard to devise and implement good diagnostic statistics
and plots.  Such concerns are well articulated in David Freedman's
paper (with extensive following discussion) "As Others See Us; A Case
Study in Path Analysis", in J. Educ. Statistics 12: 101-223

I'd very much like to learn of published examples of the use of
these methods which do address these concerns.  If these can be
found, they would be good references to include when and if these
techniques are implemented.

One reason for the current popularity of qualitative research methods
in the social sciences may be a disaffection with the way that these
methodologies are used to provide results whose meaning and
interpretation make sense only to high priests of these arts.  There's
a good deal of unease about much of the way these methdologies are
used in the social sciences, but those who are uneasy rarely have the
technical skills needed to go beyond unease to articulate criticism.

(By the way, I think qualitative research has a pretty important role.
I do not want to leave the impression that I see qualititative and
quantitative approaches as alternatives, though this is the way they
are presented in much of the literature.  What is needed and has long
been needed is an effective marriage of quantitative approaches with
well-conducted qualitative research.)

John Maindonald               email : john.maindonald at anu.edu.au        
Statistical Consulting Unit,  phone : (6249)3998        
c/o CMA, SMS,                 fax   : (6249)5549  
John Dedman Mathematical Sciences Building
Australian National University
Canberra ACT 0200
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at stat.math.ethz.ch

More information about the R-help mailing list