Excuse me, everyone. I would like to share my current research on a linear mixed effect model involving three factors (A, B, C). Here are the properties of these factors: factor A has four levels (a, b, c, d: repeated measurement), factor B has two levels (group measurement), and factor C is a continuous variable representing language proficiency scores. I performed the analysis using the lmer function and observed significant interactions between A and B, A and C, as well as between B and C.
Now, I am interested in conducting a simple main effect analysis. Since both factors A and B are grouped variables, I analyzed them using the emmeans package. However, when I attempted to use the simple slopes analysis from the reghelper package, which handles continuous and grouped variables together, I only obtained the analysis results for factors A and C.
I would greatly appreciate any guidance on how to set up a simple slope analysis to investigate the interaction between factors B and C. Thank you for your assistance.
I'd combine A and B into a new factor AB with 8 levels, so you can estimate and test the slope for each level-combination individually. In R, you can generate such a combined factor with AB
Why not plot the marginal effects or predictions from the model using something like sjPlot https://strengejacke.github.io/sjPlot/articles/plot_interactions.html ?
To Thom Baguley
Thank you for your reply. I'm not sure if it's acceptable to only plot without reporting the statistical results.
A plot of marginal effects would include confidence bands (which would be more informative than arbitrary selection of simple slopes at various points, I think). Presumably you'd report the model.
- Badges
- Science topic
- Similar topics
- Psychology