Furthermore, if you stick to doing sensible tests, the tests are independent of
the coding. For example, interaction tests are invariant to coding and so are
global tests of combined main effect + interaction (e.g., H0: age is not a risk factor
for
either sex vs. Ha: age is a risk factor for at least one sex, not restricting the age
slope to be equal for the two sexes). -Frank Harrell
Paige Miller wrote:
> Burke Johnson wrote:
> >
> > Hi,
> >
> > A student of mine is getting ready to develop a GLM prediction model that will
>include a mixture of categorical and quantitative predictor variables. We will
>probably not include interaction terms in the model (i.e., it will be a main effects
>only model).
> >
> > Here's my question: Do you suggest using dummy coding (0,1) or effects coding
>(1,0,-1) for the categorical variables included in the model?
> >
> > The reason I'm asking is because dummy coding does not always give the same result
>for a factorial design as does ANOVA and effects coding, and, hence, Pedhazur
>recommends using effects coding rather than dummy coding in the factorial case. Do
>you know if the choice of dummy or effects coding matters for a main effects only
>model with multiple categorical and quantitatively scaled predictor variables?
>
> Your predictions will be the same regardless of the coding of the dummy
> variables. The parameter estimates (and thus the associated t-tests or
> F-tests) may change depending on the coding, but the parameter estimates
> are not unique anyway when categorical variables are used.
>
> --
> Paige Miller
> Eastman Kodak Company
> [EMAIL PROTECTED]
> "It's nothing until I call it!" -- Bill Klem, NL Umpire
--
Frank E Harrell Jr
Professor of Biostatistics and Statistics
Division of Biostatistics and Epidemiology
Department of Health Evaluation Sciences
University of Virginia School of Medicine
http://hesweb1.med.virginia.edu/biostat
