I'm looking at Springtail community analysis and want to see what, if any, affect Litter Type (LT), Site Type (ST), and Sampling Day (SD) have on springtail abundance. LT and ST both have 3 factors each while SD has 2.
I'm still learning the ropes with R, but initially thought the best way to do this was along the lines of:
model1 <- glm(Abundance ~ LT * ST * SD, data=mydata, family=poisson())
summary(model1)
anova(model1, test="Chisq")
Does this seem logical? Am I on the right track? Further more I'd like to perform some sort of posthoc TUKEYHSD test to get more information on any variables which DO significantly affect Springtail Abundance. Is this possible?
Abundance? – whuber Nov 05 '18 at 14:01