A comparative examination of auditor premature sign-offs using the direct and the randomized response methods

Prior research has examined the incidence of questionable acts by auditors using a direct method (DM) of inquiry. A potential problem with the use of the DM of eliciting responses to sensitive issues is respondent bias. This study assessed the degree of respondent bias by examining auditor premature sign-offs using both the DM of inquiry and the randomized response technique (RRT) proposed by Greenberg et al. (1971). The findings using the RRT were that, on the average, auditors prematurely signed off almost ten times over a period of one year. Over 78 percent of the auditors admitted to having prematurely signed off at least once. A significantly lower incidence of premature sign-offs was reported by auditors who responded to the DM of elicitation. While the DM of inquiry is an acceptable approach for examining many types of issues, the results suggest that the RRT is a preferred method for eliciting sensitive information.