power for repeated-measures ANOVA lacking sphericity
Is there a function that calculates the power of a
repeated-measure ANOVA design, e.g., 2 groups, 4
within-subject factors, an average 0.40 correlation between
the 4 within factors, etc.
I don't think I can use power.anova.test() because it does
not consider corr=0.40.
I am hoping that someone has already implemented the method
by Muller & Barton (1989, JASA 84: 549-555) on how to
approximate statistical power for a univariate test in
repeated-measures ANOVAs when the sphericity assumption is
not met. If not, perhaps the MANOVA approach by O'Brien &
Kaiser (1985, Psych Bull: 97, 316-33).
library(pwr) does not seem to support either.
Many thanks in advance,
Please note that this e-mail and any files transmitted with it may be
privileged, confidential, and protected from disclosure under
applicable law. If the reader of this message is not the intended
recipient, or an employee or agent responsible for delivering this
message to the intended recipient, you are hereby notified that any
reading, dissemination, distribution, copying, or other use of this
communication or any of its attachments is strictly prohibited. If
you have received this communication in error, please notify the
sender immediately by replying to this message and deleting this
message, any attachments, and all copies and backups from your
Re: power for repeated-measures ANOVA lacking sphericity
On Fri, Mar 11, 2011 at 6:08 AM, Yuelin Li <[hidden email]> wrote:
> Is there a function that calculates the power of a
> repeated-measure ANOVA design, e.g., 2 groups, 4
> within-subject factors, an average 0.40 correlation between
> the 4 within factors, etc.
> I don't think I can use power.anova.test() because it does
> not consider corr=0.40.
The easiest approach is likely to be simulation: simulate data from
the design and compute the p-value, repeat a few thousand times.
Professor of Biostatistics
University of Auckland