Assessing Method Variance in Multitrait-Multimethod Matrices: The Case of Self-Reported Affect and Perceptions at Work
Year
1990
Author(s)
Richard P Bagozzi, Youjae Yi
Journal
Journal of Applied Psychology
Volume
75(5)
Pages
547-560
P. Spector (see record 1987-33304-001) 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 L. J. Williams et al (see record 1989-31744-001) concluded that method variance was prevalent. These studies were extended by examining several important but often neglected issues in assessing method variance. A direct-product model is described 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. Reanalyses indicate that method variance in these studies is more prevalent than Spector concluded but less prevalent than Williams et al asserted. The methods can have multiplicative effects, supporting the claim made by D. T. Campbell and E. J. O'Connell (1967, 1982).