Hello
I have an experience where participants perform a stress task with three time measurements (resting - task-recovery). I measure anxiety at these 3 time periods as well as alexithymia only before the experiment.
I obtain a triple interaction with multilevel modeling. the graph is below. I would test the slope differences between resting and task for high and low anxiety at High alexithymia. I'm OK with marginal means but it s not what I want..despite reading the excellent book from Mitchell (interpreting and visualizing regression models), I can't find the solution. Any help? thanks a lot.
carole
I have an experience where participants perform a stress task with three time measurements (resting - task-recovery). I measure anxiety at these 3 time periods as well as alexithymia only before the experiment.
I obtain a triple interaction with multilevel modeling. the graph is below. I would test the slope differences between resting and task for high and low anxiety at High alexithymia. I'm OK with marginal means but it s not what I want..despite reading the excellent book from Mitchell (interpreting and visualizing regression models), I can't find the solution. Any help? thanks a lot.
carole
