Social Desirability Response Bias in Accounting Ethics Research

Accounting ethics researchers commonly use self-report behavioral responses to scenarios as a research tool to examine ethical judgments. A problem with self-report behavioral intention scales is the susceptibility of these scales to social desirability response bias. Specifically, some individuals have a tendency to present responses that reflect a socially desirable response, not their true beliefs and values. Some accounting ethics researchers have attempted to minimize social desirability response bias by phrasing behavioral intention items in the third person. However, prior to the current study, no direct empirical evidence was available relative to the commonly held belief that changing a statement phrased in the first person to one phrased in the third person reduces social desirability response bias. The results of this study clearly indicate that responses to behavioral intention items phrased in the first person have a statistically significant correlation with social desirability response bias. As expected, there is a lack of a statistically significant correlation between responses to the other phrased behavioral intentions items and social desirability response bias. Researchers are recommended to use behavioral intention items phrased in the third person to avoid problems with social desirability response bias. If researchers use behavioral intention statements phrased in the first person, then procedures to control for social desirability response bias need to be incorporated into the research design.