Bioequivalence and Bioavailability Forum

Dear ElMaestro,

I was lying awake all night trying to figure this one out.
It appears that options(contrasts = c("contr.helmert", "contr.poly")) does the trick, then the model fits the mean as intercept. I got the inspiration for this here. Let it be said, I am not totally comfortable with contrasts (or with BE or with R or ...) but if the contrasting principle of summing to zero applies here then I would guess that the coefficient for Trt1 is -Trt2; the summary readout is consistent with this.
Aaaah, life is one long bloody learning process. HS, I guess Karl Popper or Kermit the Frog or some other fancy philosopher has a quote that fits on me in this situation?!

Best regards
ElMaestro - "Roses are red, violets are blue, I'm a schizophrenic and so am I"
(attributed to Oscar Levant)