On Sat, 2007-12-08 at 22:51 +0100, Irene Mantzouni
wrote:> Hi all!
>
> I am fitting a (mixed) model with a factor (F) and continuous response and
predictor:
> y~F+F:x
>
> (How) can I check the significance of the model at each factor level (i.e.
the model could be significant only at one of the levels)?
>
> Thank you!
Irene
If I understand your doubt is necessary only use a summary command.
Example
set.seed(123)
x<-rnorm(300,sd=2)
F<-sample(rep(letters[1:3],100))
y<-rnorm(300,mean=2,sd=1.5)
model<-lm(y~F+F:x)
summary(model)
(..)
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 2.118128 0.151543 13.977 <2e-16 ***
Fb -0.052866 0.213729 -0.247 0.805
Fc -0.229110 0.213726 -1.072 0.285
Fa:x 0.036987 0.074549 0.496 0.620
Fb:x 0.059439 0.083823 0.709 0.479
Fc:x 0.003769 0.082373 0.046 0.964
---
Signif. codes: 0 ?***? 0.001 ?**? 0.01 ?*? 0.05 ?.? 0.1 ? ? 1
(...)
In this case only the Intercept is significant
--
Bernardo Rangel Tura, M.D,MPH,Ph.D
National Institute of Cardiology
Brazil