Question: Interaction Term

#1
Short question on how to interpret the results of a linear regression with regards to interaction terms.

Model without interaction: y = c + b1x1 + b2x2
Results: both independent significant ( x1 & x2)

When introducing interaction term: y = c + b1*x1 + b2*x2 + b3*x1*x2
Results: b1*x1 and interaction term significant but not b2*x2.

How do I interpret such results? and what does it tell me?

Best regards,
Preppy
 

hlsmith

Omega Contributor
#2
You keep the base terms in the model for the multiplicative interaction term, but now you only care about and interpret the interaction term. So you no longer need to try and interpret the base terms since they are actually dependent on each other. Another common practice that is missed some times it that if the base terms are continuous, it is good practice to center them.