“…Traditionally, difference scores between two constructs are often treated as congruent (that is, fit or match; Shin, Kim, Choi, Kim, & Oh, ) or incongruent (relative leader–member exchange: RLMX; Henderson, Wayne, Shore, Bommer, & Tetrick, ), and then used to predict the outcomes of interest. However, because traditional methods (such as, the algebraic, absolute, or squared difference between two measures) have several drawbacks (including reduced reliability, ambiguous interpretation, confounding effects, untested constraints, and dimension reduction) (Edwards, ), polynomial regression has been suggested as an alternative technique to deal with difference scores (Edwards, ; Edward & Parry, ).…”