Are employee surveys biased? Impression management as a response bias in workplace safety constructs

Publication date: October 2019Source: Safety Science, Volume 118Author(s): Nathanael L. Keiser, Stephanie C. PayneAbstractSafety researchers often rely on employee surveys, which are susceptible to impression management. The purpose of this investigation was to estimate the prevalence of impression management in self-report measures of workplace safety constructs in three distinct samples. Results from an initial sample of research laboratory personnel indicated that impression management accounted for 28% of the variance in the relationships among safety constructs. In a second sample of research laboratory personnel, impression management was significantly related to safety constructs even after controlling for personality trait variance and accounted for more variance (35%) in the safety construct relationships than what was found in the first sample. A third sample of oil and gas personnel provided less support for the biasing effect of impression management, accounting for 11% of the variance in safety construct relationships. Overall, these data suggest that impression management accounts for up to one-third of the variance in workplace safety construct relationships. Safety researchers should seek to replicate these findings, examine interventions to reduce the prevalence and impact of impression management, and identify alternative sources or methods of assessment.

Source: Safety Science - June 6, 2019 Category: Occupational Health Source Type: research