Can someone explain what it means to test for significant differences between interaction terms in a multiple regression? I understand testing to see if each interaction term is significant, but I don’t know how you test for differences between interaction terms. The situation is experiment with three conditions, a continuous measured variable (let’s say religiosity), and dependent variable (let’s say risk ratings). Here is the description of the analysis: “We conducted a multiple regression analysis, entering dummy-coded variables representing experimental condition, religiosity (continuous), and the interaction of each dummy-coded variable with religiosity as predictors of risk ratings. The analysis yielded a significant difference between the Condition 1 × religiosity and both the Condition 2 × religiosity, β = x.xx, p < .01, and the Condition 2 × religiosity interactions, β = −0.21, p < .01.” Thanks in advance!


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.