s-news
[Top] [All Lists]

contrasts in interaction term

To: s-news@wubios.wustl.edu
Subject: contrasts in interaction term
From: Ana Bio <a.bio@geog.uu.nl>
Date: Thu, 28 Jun 2001 13:47:10 -0700
Thanks to Ruud Koning who knew the answer for a two-level factor.

Does anyone, by any chance, know the default contrasts for a 4-level factor?
The default contrasts for the factor itself result in k-1 (i.e. 3) coefficients

contrasts(datafin$soil)

     [,1] [,2] [,3]
K   -1   -1   -1
L    1   -1   -1
V    0    2   -1
Z    0    0    3

The interaction term results in 4 coefficients

f1 <- glm(Caltpalu ~ soil:mean, data = datafin, family = binomial)

coef(f1)

   (Intercept)   soilKmean   soilLmean     soilVmean   soilZmean
 -0.8134896782 16.36298448 2.315063712 -0.7620437211 10.34777897


Which contrasts did S+ use here?

Thanks in advance,

Ana



<Prev in Thread] Current Thread [Next in Thread>