David Studer
2012-Feb-15 10:50 UTC
[R] Multiple linear Regression: Standardized Coefficients
Ein eingebundener Text mit undefiniertem Zeichensatz wurde abgetrennt. Name: nicht verf?gbar URL: <https://stat.ethz.ch/pipermail/r-help/attachments/20120215/9099c528/attachment.pl>
Kenn Konstabel
2012-Feb-15 11:36 UTC
[R] Multiple linear Regression: Standardized Coefficients
It's a bit dangerous to call them "betas" in this list. Standardized regression coefficients sounds much better :) A simple way is to first standardize your variables and then run lm again. lm(scale(height)~scale(age) + factor(sex)) # or, depending on what you want: lm(height~scale(age)+factor(sex)) # or, but only for Statistica and SPSS users: lm(scale(height)~scale(age) + scale(sex)) # But of course, it's pointless in your case: standardization makes sense when units don't matter # but years and meters (even feet and inches, for that matter!) make much more sense than sd # "units" of an unknown sample. KK On 2/15/12, David Studer <studerov at gmail.com> wrote:> Hello everybody, > > Can anyone tell me, how to obtain standardized regression coefficients > (betas) for > my independent variables when doing a multiple linear regression? > > height<-c(180,160,150,170,190,172) > sex<-c(1,2,2,1,1,2) > age<-c(40,20,30,40,20,25) > > fit<-lm(height~age+sex) > summary(fit) > > I already heard about the "QuantPsyc"-Package, which, unfortunately, > produces an error > (it says "sd(<data.frame> is deprecated"). > > > Thank you very much! > David > > [[alternative HTML version deleted]] > > ______________________________________________ > R-help at r-project.org mailing list > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide http://www.R-project.org/posting-guide.html > and provide commented, minimal, self-contained, reproducible code. >