On the Current Use of the Stringer Method of MUS: Some New Directions.

Abstract This paper presents the results of a survey on the use of the Stringer method of monetary unit sampling (MUS) by 12 large public accounting firms (including all "big eight" firms). Our survey results illustrate that the sample size planning methods and use of understatement errors by these firms differ significantly from the procedures presented in the 1983 AICPA Audit Sampling guide. Following up on these survey results, the paper presents some technical results on determining sample sizes and on using understatement errors with the Stringer method of MUS. The latter results are based on simulation studies of the reliability and comparative performance of two methods for using the understatement errors: the Meikle 1972 approach and the Leslie, Teitlebaum and Anderson (LTA) 1979 approach. While half of the 12 firms surveyed are using one of these two methods, to our knowledge no prior evidence has been made public on the comparative performance of these two methods. Both methods were found to be highly reliable over a wide range of simulations. In terms of efficiency, the LTA approach was found to have a relatively modest advantage over the Meikle approach.