[R] AOV/LME

[janesense]

Hi everyone,

I'm having a little trouble with working out what formula is better to use
for a repeated measures two way anova. I have two factors, L (five levels)
and T (two levels). L and T are both crossed factors (all participants do
all combinations). So, I do:

    summary(aov(dat~L*T+Error(participant/(L*T)),data=dat4))

But get different results with:

   anova(lme(dat~L*T,random=~1|participant,data=dat4)).

Rather, the lme results are the same as those with the formula

   summary(aov(dat~L*T+Error(participant),data=dat4))

I know most people advocate the use of lme rather than aov, but I think I
need to include factors L and T in the error term and can't figure out how
to do this with lme, so possibly the first aov is more correct. I have a
balanced design so I don't think it would be a problem to use aov.

The first aov shows an expected significant interaction (unlike lme and the
second aov), but I don't want to use this if it's incorrect!

If anyone could help, that would be great.




-- 
View this message in context: http://r.789695.n4.nabble.com/AOV-LME-tp3044118p3044118.html
Sent from the R help mailing list archive at Nabble.com.