I'm looking for ways to display information about many multiple comparisons
in a compact form.
I know how to plot confidence intervals for multiple comparisons, for
example
> mc<-multicomp(...)
> plot(mc)
With 20 treatments, this will produce about 10 pages of plots of confidence
intervals.
I'm interested in other ways to display and interpret this data. A couple
of ideas:
1. Display the treatment means along a number line and use underscores to
group non-significant means. For example:
---A--B----C-D
---- ---
--------
2. Create a matrix with cell entries that are the estimates of the
difference in means. Use a special character to denote estimates that are
significantly nonzero. For example:
A B C D
A 3 8* 10*
B 5 7
C 2
Does anybody know of S+ functions that will perform tasks like this?
Kevin Wright, Pioneer Hi-Bred Int'l
-----------------------------------------------------------------------
This message was distributed by s-news@wubios.wustl.edu. To unsubscribe
send e-mail to s-news-request@wubios.wustl.edu with the BODY of the
message: unsubscribe s-news
|