dprkexchange@HOTMAIL.COM wrote [in part]: >I'm using survey data (from the California Health Interview Survey) to >estimate as best I can the independent effect of income on stroke. The >variables are coded in the interview data so that increasing values >correspond with better health indicators.

I see that some more-responsive posters have already helped you out. So let me gripe about your first paragraph.

You're using survey data. So PROC LOGISTIC has given you the wrong standard errors, and the wrong p-values. Oops.

You really need to be using PROC SURVEYLOGISTIC and all the design effect information that is available for the CHIS. Isn't it a stratified sample? Aren't there sample weights? You need to find out all that good stuff, so you can get the right answers.

HTH, David -- David L. Cassell mathematical statistician Design Pathways 3115 NW Norwood Pl. Corvallis OR 97330

_________________________________________________________________ On the road to retirement? Check out MSN Life Events for advice on how to get there! http://lifeevents.msn.com/category.aspx?cid=Retirement