[R-sig-ME] How should the model be written in R

[ONKELINX, Thierry]

Dear Petter,

Is it correct that quadrant is a combination of rewetting and tree removal? And thus quadrant has 4 levels? In that case you should at least add the interaction with site, then quadrant will have 12 levels. Do you have multiple measurements of the releves? If not you should drop the obsplot factor.

If you don't trust the output, did you will need to give us more details. What are the numbers that you don't trust and what do you expect them to be?

Best regards,

Thierry

ir. Thierry Onkelinx
Instituut voor natuur- en bosonderzoek / Research Institute for Nature and Forest
team Biometrie & Kwaliteitszorg / team Biometrics & Quality Assurance
Kliniekstraat 25
1070 Anderlecht
Belgium
Thierry.Onkelinx at inbo.be
www.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 Petter Hedberg
Verzonden: donderdag 8 december 2011 11:04
Aan: r-sig-mixed-models at r-project.org
Onderwerp: [R-sig-ME] How should the model be written in R

Have a question regarding how to accurately put this experimental design in the correct way in R.
I have attempted this before using lme, but eventualy we tried SAS. 

Response variable ~ site + rewetting + tree removal + distance along ditch + distance along ditch*tree removal + distance from ditch/rewetting, random=~1|quadrant/transect/obsplot)).

What puzzled me was that the data I got when using the lme formula was not the same, and I clearly must have written it the wrong way.

There are 3 sites, each treated in a split plot design with rewetting and tree cutting. This results in each site being divided up into 4 different treatment combinations that we refer to as quadrants. Each quadrants has 4 transects, and Each transect has 5 relevees that were monitored. Hence the nesting of the random variable quadrant/transect/obsplot in the end. I would be greatly thankful for any help in what the correct way to write this in R would be. Was I right to go after lme. I thought I was, but obviously the way I put it in lme was wrong.

Any input on this is greatly appreciated.

Best regards, Petter H

--
This email was Anti Virus checked by Astaro Security Gateway. http://www.astaro.com

	[[alternative HTML version deleted]]

_______________________________________________
R-sig-mixed-models at r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models