In a regression, is it possible to have a significant R but non significant b coefficients for both X1 & X2? What circumstances would this occur? Give research example