Re: RM ANOVA and other methods

On Mar 13, 4:30 am, greg <jsg...@xxxxxxxxxxx> wrote:
I have a study design question I was having trouble with.
I have 2 groups (placebo & case) and 3 timepoints.
I could use 1 2-way Repeated Measures ANOVA to analyse the data, but
the timepoints would (surely) simply be treated as 3 groups, rather
than 3 ordinal points.
Is there a method that will do this for me?
(And if there isn't, how would I interpret a result that went up and
then back down again when I'm expecting a dose response?)

Look up (polynomial) trend analysis. The basic concepts are discussed
here in the context of one-way (between-Ss) ANOVA, but it can also be
used for repeated measures factors:

http://davidmlane.com/hyperstat/intro_ANOVA.html

It sounds like you are expecting a significant quadratic trend.

--
Bruce Weaver
bweaver@xxxxxxxxxxxx
www.angelfire.com/wv/bwhomedir
"When all else fails, RTFM."
.

Relevant Pages

  • RM ANOVA and other methods
    ... I have a study design question I was having trouble with. ... I could use 1 2-way Repeated Measures ANOVA to analyse the data, ... the timepoints would simply be treated as 3 groups, ...
    (sci.stat.edu)
  • Problem with cross correlation - xcorr
    ... I am having a problem with the output from xcorr, as it doesn't give me what I am expecting. ... and vice verse. ... I analyse these time series in the following way: ...
    (comp.soft-sys.matlab)