Johnstone, D. J., Audit risk in terms of probabilities: The AUP24 model, School of Accounting & Finance, University of Wollongong, Working Paper 4, 1994.
The AUP24 audit risk model defines audit risk implicitly as the joint probability of three independent events: (i) a material error occurring in an account balance, (ii) that error not being corrected by internal control procedures, and (iii) the uncorrected balance being accepted by the auditor. A more apposite risk measure, relating to these same possible events, is the conditional probability of a material-error given that the stated balance has been subject to internal control procedures and accepted by the auditor. The two risk measures so defined are related by the laws of probability, through Bayes' theorem specifically, but are not the same and exhibit no necessary correlation. Calculation of the conditional ('Bayesian') risk measure requires consideration of both the type I (alpha) and type II (beta) error probabilities of the auditor's substantive test procedure. Unless both error characteristics are taken into account, it is not possible to interpret a test result (acceptance or rejection) in terms of the probability of the stated account balance being materially correct.