[R-sig-ME] cross-over design
Thierry.ONKELINX at inbo.be
Wed Feb 16 15:39:53 CET 2011
lmer(TGA ~ Treatment + Time + (1|sub) ,family=binomial) is the correct syntax
ir. Thierry Onkelinx
Instituut voor natuur- en bosonderzoek
team Biometrie & Kwaliteitszorg
Research Institute for Nature and Forest
team Biometrics & Quality Assurance
tel. + 32 54/436 185
Thierry.Onkelinx at inbo.be
To call in the statistician after the experiment is done may be no more than asking him to perform a post-mortem examination: he may be able to say what the experiment died of.
~ Sir Ronald Aylmer Fisher
The plural of anecdote is not data.
~ Roger Brinner
The combination of some data and an aching desire for an answer does not ensure that a reasonable answer can be extracted from a given body of data.
~ John Tukey
> -----Oorspronkelijk bericht-----
> Van: r-sig-mixed-models-bounces at r-project.org
> [mailto:r-sig-mixed-models-bounces at r-project.org] Namens Soyeon Ahn
> Verzonden: woensdag 16 februari 2011 10:11
> Aan: r-sig-mixed-models at r-project.org
> Onderwerp: [R-sig-ME] cross-over design
> Dear all,
> I have a repeated binary measurements and can't figure out
> how my data will work.
> Response (dependent, Y/N) was repeatedly recorded across Time
> (categorical : 3 time points) with different Treatments (categorical :
> 2 levels).
> The problem is that each subject was only assigned either the
> first or the second time point. All were assigned into the
> third time point, though. Thus, it seems to me that Subject
> (random) is cross-over withn Time (fixed) effect.
> For example,
> Treatment Time Res
> sub1 A 1 Y
> sub1 B 1 N
> sub1 A 3 Y
> sub1 B 3 Y
> sub2 A 2 Y
> sub2 B 2 Y
> sub2 A 3 N
> sub2 B 3 N
> So far, I have played around the following models and I feel
> that the fist one produces feasible outcomes, even though the
> last one seems to be the right syntax.
> lmer(TGA ~ Treatment + Time + (1|Time) ,family=binomial)
> lmer(TGA ~ Treatment + Time + (1|sub) ,family=binomial)
> lmer(TGA ~ Treatment + Time + (1|sub) + (1|Time) ,family=binomial)
> Can anyone suggest a solution to this? And please correct me
> if I'm wrong
> R-sig-mixed-models at r-project.org mailing list
More information about the R-sig-mixed-models