Greg Snow Greg.Snow at imail.org
Tue Dec 9 17:08:25 CET 2008

ANCOVA usually is used when you have a numerical covariate that you want to adjust for, your description does not include any, so a regular anova (aov function) can be used.  If you do have 1 or more covariates to adjust for, then it can be thought of as a regression problem (ANCOVA is a special case of linear models/regression) and you can fit it using either the aov or lm function.

Gregory (Greg) L. Snow Ph.D.
Statistical Data Center
Intermountain Healthcare
greg.snow at imail.org

> -----Original Message-----
> From: r-help-bounces at r-project.org [mailto:r-help-bounces at r-
> project.org] On Behalf Of Samuel Okoye
> Sent: Tuesday, December 09, 2008 2:03 AM
> To: r-help at stat.math.ethz.ch
> Subject: [R] ANCOVA
> Hello,
> Could you please help me in the following question:
> I have 16 persons 6 take 0.5 mg, 6 take 0.75 mg and 4 take placebo! Can
> I use the ANCOVA and t-test in this case? Is it possible in R?
> Thank you in advance,
> Samuel
>         [[alternative HTML version deleted]]

More information about the R-help mailing list