ON THE USE OF POLYNOMIAL REGRESSION EQUATIONS AS AN ALTERNATIVE TO DIFFERENCE SCORES IN ORGANIZATIONAL RESEARCH.

Abstract

For decades, difference scores have been widely used in studies of congruence in organizational research. Although methodological problems with difference scores are well known, few viable alternatives have been proposed. One alternative involves the use of polynomial regression equations, which permit direct tests of the relationships difference scores are intended to represent. Unfortunately, coefficients from polynomial regression equations are often difficult to interpret. We used response surface methodology to develop an interpretive framework and illustrate it using data from a well-known person-environment study. Several important findings not reported in the original study emerged.