Mixed Model – How to Interpret Posthoc Table Output in R Package Phia for Mixed Model Interaction (Covariate+Factor)

interactionmixed modelr

In R, using package lme4, I have used the following 2 mixed models to determine I have a signifacnt interaction between a covariate (continous, normally distributed) and a factor (three levels: herbivores, plants, predators):

test1 <- lmer( mode ~ sr * func.group + (1|community), data=nr.test, REML="FALSE")
test2 < -lmer( mode ~ sr + func.group + (1|community), data=nr.test, REML="FALSE")
anova(test1, test2)

Data: nr.test
Models:
test2: mode ~ sr + func.group + (1 | community)
test1: mode ~ sr * func.group + (1 | community)
     Df    AIC    BIC  logLik  Chisq Chi Df Pr(>Chisq)    
nr.test2  6 77.458 82.093 -32.729                             
nr.test1  8 62.570 68.751 -23.285 18.887      2  7.919e-05 ***

I have obviously plotted the interaction for each level of factor, and all three levels are positive relationships which cross over each other (i.e. different levels of slope steepness).

However, I would like to be able to identify which level of the factor are significantly different from one another, and which are also significantly different from a slope value of zero (i.e. which slopes are significant).

I have installed the R package phia, and used the command 

testInteractions(test1, pairwise="func.group", slope="sr")

and recieved the following output:

Adjusted slope for sr 
Chisq Test: 
P-value adjustment method: holm
                     Value Df   Chisq Pr(> Chisq)    
    herbivore-plant 0.11878  1  4.3061     0.03798 *  
 herbivore-predator 0.40567  1 65.3283   1.902e-15 ***
     plant-predator 0.28689  1 30.3474   7.224e-08 ***

It would appear for each level of factor the slopes are significantly different from one another, but how can I know which ones are significantly different from zero, and how would I report this?

Best Answer

how can I know which ones are significantly different from zero, and how would I report this?

To obtain the slope estimate and its statistical significance for each level of a factor, you can perform the following tests:

testInteractions(test1, custom=list(func.group=c(1,0,0)), slope="sr", adjustment="none")
testInteractions(test1, custom=list(func.group=c(0,1,0)), slope="sr", adjustment="none")
testInteractions(test1, custom=list(func.group=c(0,0,1)), slope="sr", adjustment="none")

You may report the results of the individual slopes the same way as the pairwise comparisons of slopes (e.g., tables).