curvilinear relation moderated by two dummy variables
Hi all, I hope you can help me out with the following question:
I expect a moderated inverted u-shape relation. However my moderator is a variables that containts three categories (low, medium, high). Therefore, I dummy coded the variable in to two dummy variables: medium and high (low is the baseline category).
If my moderated quardratic interaction term is negative (and significant) for both dummies, how can I interpret these results? Normally, with a moderated dummy variable you can compare with the baseline, can I do the same here? (I did regression analysis)
Can I also do simple slope analysis for these effects?