>>> Duck-Hye Yang <dyang@CHAPINHALL.ORG> 10/4/2006 12:05 pm >>> wrote

<<<<

My data has somewhat small sample size (N=79). A variable of primary interest as an independent variable is categorical, having three values. 1=have condition A (N=30) 2=have condition B (N=40) 3=have both A and B (N=9)

As you see, the third one has a small sample size. And there may be multicollnearity among the three groups. Is the use of multiple regression ok under these circumstances? >>>

As usual, it would be nice to have some context.

Is this the only IV in the model?

What are these 'conditions'?

Does anyone have neither condition?

Is this from a survey (you KNEW David Cassell would ask, but I am on the East Coast, so I am asking first :-)) etc.

I don't understand what you mean by collinearity in this context. Collinearity is a relation among multiple IVs. It sounds like you have ONE IV with three levels. But perhaps not

If you write back to SAS-L with answers to these questions, then perhaps someone will be able to help (maybe even me).

Peter

Peter L. Flom, PhD Assistant Director, Statistics and Data Analysis Core Center for Drug Use and HIV Research National Development and Research Institutes 71 W. 23rd St http://cduhr.ndri.org www.peterflom.com New York, NY 10010 (212) 845-4485 (voice) (917) 438-0894 (fax)