Thanks a lot for the response and explanation, Lorenz! It seems that
I can run the analysis with 'lm' without treating subject as a random
factor
lm(Resp ~ Group*Cov, TestData)
Thanks,
Gang
On Aug 2, 2007, at 5:09 AM, <[EMAIL PROTECTED]>
<[EMAIL PROTECTED]> wrote:
> I do not think anyone has answered this.
>
>> I'm trying to run a simple one-way ANCOVA with the lmer
>> function in R package lme4, but have encountered some
>> conceptual problem. The data file MyData.txt is like this:
>>
>> Group Subj Cov Resp
>> A 1 3.90 4.05
>> A 2 4.05 4.25
>> A 3 4.25 3.60
>> A 4 3.60 4.20
>> A 5 4.20 4.05
>> A 6 4.05 3.85
>> B 7 3.85 4.15
>> B 8 4.15 4.60
>> B 9 4.60 4.15
>> B 10 4.15 4.40
>> B 11 4.40 3.35
>> B 12 3.35 3.80
>> B 13 3.80 3.90
>>
>> Since I would like to treat subject (Subj) as a random
>> factor, my one- way ANCOVA (with covariate, Cov) is:
>> TestData=read.table("MyData.txt", header=T);
>> m1 <- lmer(Resp ~ Group * Cov + (1 | Subj), TestData)
>
> This is a conceptual problem, indeed. To be able to estimate a
> subject effect, you need to have repeated measurements (several
> lines) per observed subject ...
>
> Cheers, Lorenz
______________________________________________
R-help@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.
