[R] Incomplete Factorial design

Thomas Lumley tlumley at u.washington.edu
Fri Feb 6 16:30:50 CET 2004

On Fri, 6 Feb 2004, Simon Fear wrote:

> One could also fit
> fit <- lm(y~A*B - 1, data.frame(y=..., A=..., B=..,)
> which will give a direct a:b term (as the negative of the
> intercept in Spenser's formulation). Arguably this is more
> natural in a setting where there is no placebo so that
> an intercept term has a less obvious interpretation.

I would have though the natural analysis was to compare A+B to B alone and
A+B to A alone with two separate t.tests, and power the analysis for
these.  It's not really a factorial design at all.


More information about the R-help mailing list