Re: st: -factor- with binary variables

[Ron�n Conroy <rconroy@rcsi.ie>]

Patricia Sourdin wrote:

> Hi Statalist
> just a query on -factor-.
> I am trying to construct an index where I have five variables which 
> are binary
> indicators.
> I have read somewhere that it is not appropriate to use factor 
> analysis if the
> variables are binary.  Can anyone confirm, please?
Better still, try -msp- and use Mokken scaling. Jean-Benoit Hardouin's 
site http://anaqol.free.fr is a great resource here. Essentially, if you 
are building or evaluating a scale it makes more sense to use item 
response theory to evaluate the scale as a scale. Mokken scaling is a 
more 'intelligent' approach than just doing KR20s, as it tries to build 
a coherent scale from your items, and, when it can no longer add items, 
it tries to use the remainder to build another scale, until the item 
pool is exhausted.


--

Ronan M Conroy (rconroy@rcsi.ie) 
Senior Lecturer in Biostatistics 
Royal College of Surgeons 
Dublin 2, Ireland 
+353 1 402 2431 (fax 2764) 
-------------------- 
Just say no to drug reps 
http://www.nofreelunch.org/

*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/