[R] glm with nesting

[Berton Gunter]

Jeffrey:

Please... May I repeat what Peter Dalgaard already said: consult a local
statistician. The structure of your study is sufficiently complicated that
your stat 101 training is inadequate. Get professional help, which this list
is not set up to provide (though it often does, through the good offices and
patience of many wise contributors).

Bert Gunter
Genentech Nonclinical Statistics
South San Francisco, CA 94404

 

> -----Original Message-----
> From: r-help-bounces at stat.math.ethz.ch 
> [mailto:r-help-bounces at stat.math.ethz.ch] On Behalf Of 
> Jeffrey Stratford
> Sent: Thursday, October 05, 2006 7:46 AM
> To: HDoran at air.org; r-help at stat.math.ethz.ch
> Subject: Re: [R] glm with nesting
> 
> Harold and list,
> 
> I've changed a few things since the last time so I'm really starting
> from scratch.  
> 
> I start with
> 
> bbmale <- read.csv("c:\\eabl\\2004\\feathers\\male_feathers2.csv",
> header=TRUE)
> box <-factor(box)
> chick <- factor(chick)
> 
> Here's a sample of the data
> 
> box,chick,julian,cltchsz,mrtot,cuv,cblue,purbank,purban2,purba
> n1,pgrassk,pgrass2,pgrass1,grassdist,grasspatchk
> 1,2,141,2,21.72290152,0.305723811,0.327178868,0.003813435,0.02
> 684564,0.06896552,0.3282487,0.6845638,0.7586207,0,3.73
> 4,1,164,4,18.87699007,0.281863299,0.310935559,0.06072162,0.208
> 0537,0.06896552,0.01936052,0,0,323.1099,0.2284615
> 4,2,164,4,19.64359348,0.294117388,0.316049817,0.06072162,0.208
> 0537,0.06896552,0.01936052,0,0,323.1099,0.2284615
> 7,1,118,4,13.48699876,0.303649408,0.31765218,0.3807568,0.43624
> 16,0.6896552,0.06864183,0.03355705,0,94.86833,0.468
> 12,1,180,4,21.42196378,0.289731361,0.317562294,0.09238011,0.13
> 42282,0,0.2430127,0.8322148,1,0,1.199032
> 12,2,180,4,18.79487905,0.286052077,0.316367349,0.09238011,0.13
> 42282,0,0.2430127,0.8322148,1,0,1.199032
> 12,3,180,4,12.83127682,0.260197475,0.292636914,0.09238011,0.13
> 42282,0,0.2430127,0.8322148,1,0,1.199032
> 15,1,138,4,20.07161467,0.287632782,0.318671887,0.07046477,0.03
> 355705,0.03448276,0.2755622,0.6577181,0.8275862,0,1.503818
> 15,2,138,4,17.61146256,0.305581768,0.315848051,0.07046477,0.03
> 355705,0.03448276,0.2755622,0.6577181,0.8275862,0,1.503818
> 15,3,138,4,20.36397134,0.271795667,0.30539683,0.07046477,0.033
> 55705,0.03448276,0.2755622,0.6577181,0.8275862,0,1.503818
> 15,4,138,4,20.81940158,0.269468041,0.304160648,0.07046477,0.03
> 355705,0.03448276,0.2755622,0.6577181,0.8275862,0,1.503818
> 
> As you can see I have multiple boxes (> 70).  Sometimes I 
> have multiple
> chicks per box each having their own response  to mrtot, cuv, 
> and cblue
> but the same landscape variables for that box.  Chick number 
> is randomly
> assigned and is not an effect I'm interested in.  I'm really not
> interested in the box effect either.  I would like to know if 
> landscape
> affects the color of chicks (which may be tied into chick
> health/physiology).  We also know that chicks get bluer as the season
> progresses and that clutch size (cltchsz) has an effect so 
> I'm including
> that as covariates.  
> 
> Hopefully, this clears things up a bit. 
> 
> I do have the MASS and MEMS (Pineiro's) texts in hand. 
> 
> Many thanks,
> 
> Jeff
> 
> ______________________________________________
> R-help at 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.
>