| Abstract: | The objective of the study was to examine the effects of three independent variables ‐ accountability, audit workpaper structure, and type of control deviations ‐ on auditors' detection failure rates during control tests in a purchases, payables, and payments cycle. The experimental design used a between‐subjects manipulation of accountability and workpaper structure, and a within‐subjects manipulation of deviation type. Consistent with prior research, we observed an alarmingly high detection failure rate of 42.3 percent. This failure rate was not affected by levels of accountability or workpaper structure, although postexperiment evidence suggests that these variables were successfully manipulated. Failure rates did depend on the type of seeded control deviation, with nonmonetary deviations being overlooked most frequently. In addition to replicating prior research, our study makes two further contributions. First, we provide empirical evidence that supports Hirst's (1992) speculation that successful manipulations of accountability may not affect auditor performance because auditors may self‐induce levels of accountability that create a ceiling effect on auditor performance. Second, we observe that although auditors perceived that highly structured workpapers allowed them to be more effective and efficient when performing tests of controls, their actual audit performance was not more effective and, on average, was less efficient. |
