Dear all,

I struggling with MCMC glmm models and I have some problems to understand the output of the models I perform, in particular because I have different categorical predictors that interact with each other.

I performed a controlled condition experiment with three temperature treatments, a defence induction treatment and different plant species originating from different elevations that I classified as low-, mid- and high-elevation species.

My model predictor variables and their respective levels are the following:

- temperature treatment: mid, warm, cold

- induction treatment: non induced, induced

- elevational class: high, mid, low

The model looks like that: response variable ~temp_treat*induction_treat*elevational_class

Now, since that my predictors are categorical, the model takes one level and put it in the intercept, and show me the coefficients for the other levels.

I ordered my predictors variables in order to have always: the mid temperature, the non induced plants and the high-elevation plants in the intercept.

My question now is: how do the model calculate the coefficients for the interactions between variables? Or more clearly: if the model gives me coefficients for the interaction "warm_temp x induced x low_elev" is it related to the interaction "mid_temp x non induced x high_elev" or does it only consider the last part (i.e. high vs. low at warm temperature and induced)?

I put in attachment an example of output so that you can look by yourself. Any suggestions on how should I interpret this would be very useful for me!

Thank you in advance,

Janisse