Method variance as an artifact in self-reported affect and perceptions at work: Myth or significant problem?

Method variance is an artifact of measurement that biases results when relations are explored among constructs measured by the same method. The existence of method variance was explored for affective and perceptual constructs frequently used in organizational research. Data from multitrait–multimethod analyses, studies of social desirability and acquiescence, and relation of self-report and records of absenteeism were presented. Little evidence for method variance as a biasing problem was found with these measures. I conclude that properly developed instruments of the type studied here are resistant to the method variance problem, but that validity of these instruments cannot be assumed on the basis of these results.