```Date: Thu, 25 Jan 2001 10:08:19 -0800 Reply-To: "Dennis G. Fisher" Sender: "SAS(r) Discussion" From: "Dennis G. Fisher" Subject: Re: Proc GLM Comments: To: Dr Olaf Kruse Content-Type: text/plain; charset=us-ascii I have to weigh in on this one. Usually I would agree that ruining a perfectly good continuous variable by dichotomizing it is not a good thing to do and I once gave such advice to a grad student. It turned out that I was wrong. The variable was birthweight. This actually turned out to be a dichotomous variable, which is something I did not know at the time. Infants can be classified into low birth weight and non low birthweight. Low birth weight is a proxy (or perhaps an indicator) that there were problems with the pregnancy. So non-low birthweight infants mean that the indicators of lbw problems were not present. It does not mean that infants who are very heavy are somehow protected against these problems. In the case of this grad student, the infants should have been classified into low birth weight and non low birthweight. Weight should not have been treated as a continuous variable. You have to understand the meaning of the variable before giving an opinion about the analysis. So I guess I agree with Dr. Kruse. Just my 2 cents. Dennis Fisher Dr Olaf Kruse wrote: > cchang7814@my-deja.com wrote: > > > A colleagues of mine suggested to dichotomize "age" as categorical > > variable. Do you think that will help? > > Paige Miller replied > > >>In regression, I have yet to see an example where taking a continuous > >>variable and making it a class variable has helped. Nor am I aware of > >>a logical reason why it might be a good thing to do. > > IMHO it depends on the _true_ relationship between the variables. If you use > age as a continous variable for explaining > something that is related to the _true_ relationship "after/before > retirement", you are better off by dividing > age into two classes "before/after 65" (official retirement-age in Germany). > > Treating a variable as continous in a linear model can be a very restrictive > assumption, if the relationship is not linear. > Proper dichotomizing can help you (among other techniques) detect non-linear > relationships and outiers. > Both ways have its pros an cons. > > Cheers, Olaf > > +-----------------------------------------------------+ > Dr. Olaf Kruse > VST -Gesellschaft fuer Versicherungsstatistik > Roscherstr. 10 > 30161 Hannover/FRGermany > > mail: Olaf.kruse@vst-gmbh.de > phone:++49-511-339 599 21 > fax: ++49-511-388 57 13 > +-----------------------------------------------------+ -- Dennis G. Fisher, Ph.D. Director Center for Behavioral Research and Services 1090 Atlantic Avenue Long Beach, CA 90813 562-495-2330 562-983-1421 fax ```

Back to: Top of message | Previous page | Main SAS-L page