How Reliable are the Hurtt Professional Skepticism Scale and the Rotter Interpersonal Trust Scale for Audit Experimental Research?

This study uses an experimental context to compare the results obtained from two well accepted scales that have been designed to measure stable personality traits - the Hurtt Professional Skepticism Scale (HPSS) and the Rotter Interpersonal Trust Scale (RITS). The RITS was developed outside of accounting, whereas the HPSS was developed specifically for measuring professional skepticism in accounting. Recently, the RITS has been proposed by Quadackers, Groot and Wright (2014) as a substitute for the HPSS on the grounds that the inverse of the RITS is a more useful measure of skepticism for audit purposes as it measures the presumptive doubt form of skepticism (Nelson 2009) rather than the neutral form of skepticism measured by the HPSS and is more responsive to risk information. To test this assertion, we created three case scenarios involving three different auditing tasks adapted from current literature. For each scenario, we administered both scales, varying the order of case and scale administration to monitor the stability that would be expected of measures of stable traits. Our findings indicate that HPSS and RITS measures of skepticism and trust, respectively, vary by case presented and order of risk tolerance measurement. Skepticism, as measured per the HPSS is not affected by the order of HPSS administration; however, it is affected by case and order of risk tolerance measurement. Trust, as measured by the RITS, is affected by order (timing) of RITS administration. Specifically, measured Trust decreases when the scale is administered after cases are completed compared to the level obtained when Trust is measured before the case is completed. RITS is also marginally affected by case and order of risk tolerance measurement. This study contributes to our understanding of the HPSS and RITS, challenges the reliability of both scales as measures of stable personality traits, and supports order stability for the HPSS, but not the RITS. Both scales may have components that are sensitive to states rather than being pure measures of traits. Also, HPSS and RITS do not appear to be substitutes for one another, as the RITS does not behave as the inverse of the HPSS across cases and orders of administration. Lack of trust (the inverse of RITS) does not appear to be synonymous with skepticism as measured by the HPSS