ASSESSING METHOD VARIANCE IN MULTITRAIT MULTIMETHOD MATRICES - THE CASE OF SELF-REPORTED AFFECT AND PERCEPTIONS AT WORK

Spector (1987) concluded that there was little evidence of method variance in multitrait-multimethod data from 10 studies of self-reported affect and perceptions at work, but Williams, Cote, and Buckley (1989) concluded that method variance was prevalent. We extended these studies by examining several important but often neglected issues in assessing method variance. We describe a direct-product model that can represent multiplicative method effects and propose that model assumptions, individual parameters, and diagnostic indicators, as well as overall model fits, be carefully examined. Our reanalyses indicate that method variance in these studies is more prevalent than Spector concluded but less prevalent than Williams et al. asserted. We also found that methods can have multiplicative effects, supporting the claim made by Campbell and O'Connell (1967, 1982).