[R] Is there any R package to calcualte "Power" for ANCOVA

[Greg Snow]

I don't know of any existing packages or functions to compute power for
general ancova, but here are 2 approaches using existing tools:

1.  If the main focus is on the anova/t-test portion and the covariates
are only there to adjust for potential differences in the groups then
just use power.anova.test or power.t.test and add an additional
subject/unit for each degree of freedom to be used by the covariates.

2.  Write a function that simulates data like you expect to see and does
the analysis of interest.  Run the function a bunch of times to see how
often it rejects.  Repeat for other sample sizes.

Number 2 is how I do most my power calculations these days.  Hope this
helps,


-- 
Gregory (Greg) L. Snow Ph.D.
Statistical Data Center
Intermountain Healthcare
greg.snow at intermountainmail.org
(801) 408-8111
 

-----Original Message-----
From: r-help-bounces at stat.math.ethz.ch
[mailto:r-help-bounces at stat.math.ethz.ch] On Behalf Of Tony cogswell
Sent: Monday, November 20, 2006 9:28 AM
To: r-help at stat.math.ethz.ch
Subject: [R] Is there any R package to calcualte "Power" for ANCOVA

Dear list members:

I searched the R-help for packages to calculate power for an ANCOVA
problem I have. I have found power.t.test, power.anova.test. But it
seems that I can not found one with ANCOVA.

I have two datasetsets with variables: univariate response(one data with
continous response and one with 0 and 1), treatment(two levels),
covariates(x1,x2,x3).

I would appreciate your help.

Tony

	[[alternative HTML version deleted]]

______________________________________________
R-help at stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide
http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.