I'm trying to perform posthoc comparisons for my treatments but I keep getting this error when running the glht: "Error in modelparm.default(model, ...) : dimensions of coefficients and covariance matrix don't match".
Is there a better way of doing multiple pairwise comparisons? I've also tried using emmeans abut I'm not sure if that's the correct method.
This is a subset of my data:
mydata <- read.table(header=TRUE, text="
treatment total.bites hours rep
A 10 3.1 a1
A 1 3.2 a2
A 1024 3.22 a3
B 0 3.13 a1
B 16 3.15 a2
B 1305 3.24 a3
C 0 3.13 a1
C 0 3.26 a2
C 0 3.11 a3
D 2 3.25 a1
D 0 3.17 a2
D 3 3.21 a3
")
mC4 <- glmmTMB(total.bites~treatment + offset(log(hours)) +(1|rep), ziformula=~0, family=nbinom1, data=mydata)
summary(mC4)
summary(glht(mC4, mcp(treatment = "Tukey")))As you already mentioned emmeans you could do
library(emmeans)
pairs(emmeans(mC4, "treatment"))
#contrast estimate SE df t.ratio p.value
#A - B 0.323 8.94e-01 6 0.361 0.9824
#A - C 20.930 1.51e+04 6 0.001 1.0000
#A - D 0.892 9.05e-01 6 0.985 0.7631
#B - C 20.607 1.51e+04 6 0.001 1.0000
#B - D 0.569 9.92e-01 6 0.573 0.9365
#C - D -20.038 1.51e+04 6 -0.001 1.0000
#
#Results are given on the log (not the response) scale.
#P value adjustment: tukey method for comparing a family of 4 estimates
Here we condition on treatment and characterise all pairwise comparisons, correcting for multiple hypothesis testing using Tukey's method.