Thank you both for the extensive amount of help! I am sorry it has taken me a
bit to respond, but i've been trying to plug away at this. I still have a
few questions, if you don't mind giving me some pointers:
Here is the fake data again:
treatment
feeding
avoid
noavoid (all visits which didnât result in avoidance)
Control
nofeeding
1
357
Control
chum
2
292
Control
Satiation
4
186
Proc. Control
nofeeding
15
291
Proc. Control
chum
25
288
Proc. Control
Satiation
17
140
Magnet
nofeeding
87
224
Magnet
Chum
34
229
Magnet
Satiation
46
151
Here is the coding:
model1 <- glm(cbind(avoid, noavoid) ~ treatment, family=binomial, Â data=avoid)
summary(model1)
glm(formula = cbind(avoid, noavoid )~treatment , family = binomial,
data = avoid)
 #Predicting avoid, no avoid from just treatment
model2 <- glm(cbind(avoid, noavoid) ~ feeding, family=binomial, data=avoid)
summary(model2)
 glm(formula = cbind(avoid, noavoid )~feeding , family = binomial,
data = avoid)
#Predicting avoid, no avoid from just feeding
model3<-glm(cbind(avoid,noavoid)~treatment*feeding,family=binomial,data=avoid)
summary(model3)
glm(formula = cbind(avoid, noavoid )~treatment*feeding , family = binomial,
data = avoid)
#Predicting avoid,noavoid from the interaction between treatment and feeding
However, when I run all the models, I notice that my "control" data is not
incorporated in the output. I just receive my procedural control and magnet
data. I cannot figure what is causing this.
Secondly, when I run model 3 I am not receiving what I anticipated. I thought
I would get the following output data:
treatmentmag                Â
treatmentproc               Â
feedingnofeed                  Â
feedingsat                      Â
treatmentmag:feedingnofeed      Â
treatmentproc:feedingnofeed     Â
treatmentmag:feedingsat        Â
treatmentproc:feedingsat
What happened to the following:Â treatmentcon, feedingchum,
treatmentcon:feedingnofeed, treatmentcon:feedingsat, treatmentmag:feedingchum,
treatmentproc:feedingchum, treatmentcon:feedingchum ?  I have a strange
feeling that the issue is arising from my "family=binomial" term. Any
feedback would be greatly appreciated.
Kind Regards,
Craig
----- Original Message -----
From: "Steve Lianoglou" <mailinglist.honey...@gmail.com>
To: "David Winsemius" <dwinsem...@comcast.net>
Cc: "Craig P O'Connell" <coconne...@umassd.edu>, r-help@r-project.org
Sent: Tuesday, November 27, 2012 11:54:48 PM
Subject: Re: [R] GLM Coding Issue
Hi,
On Tuesday, November 27, 2012, David Winsemius wrote:
[snip]
<blockquote>
`cbind`-ing doesn't make much sense here. What is your target (y)
variable here? are you trying to predict `avoid` or `noavoid` status?
Sorry, Steve. It does make sense. See :
?glm  # First paragraph of Details.
</blockquote>
Indeed ... I've tried to,send a follow up email salvaging my bad call with some
hopefully useful tidbits, but it "matched some headers" and is stuck in the
mailman queue. It might come through eventually.
Don't be sorry, though ... I learned something new :-)
Still, I do apologize for the flawed advice re: the cbind-ing thing
-Steve
--
Steve Lianoglou
Graduate Student: Computational Systems Biology
 | Memorial Sloan-Kettering Cancer Center
 | Weill Medical College of Cornell University
Contact Info: http://cbio.mskcc.org/~lianos/contact
[[alternative HTML version deleted]]
______________________________________________
R-help@r-project.org 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.
