Hello,
I am performing a BACI analysis with ANOVA using the following glm:
fit1<-glm(log(Cucs_m+1)~(BA*Otter)+BA+Otter+ID+Primary, data=b1)
The summary(aov(fit1)) shows significance in the interaction; however, now I
would like to determine what combinations of BA and Otter are significantly
different (each factor has two levels). ID and PRIMARY substrates are
categorical and included in the model to help explain some of the variation
in the data. The data is unbalanced so I plan on using Tukey Kramer post
hoc analysis. Here is how my data is laid out, it is a fairly substantial
data set:
Subdistrict T Year Cucs_m Primary Persistence Otter
Fishing BA ID
109-41,42 9 2010 0.00 sil 3
1 1 A 109-41,42
109-41,42 13 2010 2.75 rck 3
1 1 A 109-41,42
109-41,42 16 2010 2.00 rck 3
0 1 A 109-41,42
109-41,42 18 2010 8.25 rck 3
0 0 B 109-41,42
I am assuming this is an appropriate pairwise comparison analysis and I
cannot get the code to work with my data. I am *unclear how to code it to
work with the interaction*; however, even when I attempt to use it only for
a single factor, it does not work (see below).
x<-aov(glm(Cucs_m~as.factor(BA),data=cuc))
glht(x, linfct=mcp(BA="Tukey"))
....................................
Error in mcp2matrix(model, linfct = linfct) :
Variable(s) ‘BA’ have been specified in ‘linfct’ but cannot be found in
‘model’!
Can anyone off suggestions on potential problems with my approach and/or
script issues?
Thank you very much in advance.
--
View this message in context:
http://r.789695.n4.nabble.com/Tukey-Kramer-with-ANOVA-glm-tp4633314.html
Sent from the R help mailing list archive at Nabble.com.
______________________________________________
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.
