Uniformity of Auditing Standards: A Replication

Abstract

The purpose of this paper is to report research results which replicate, extend and modify the conclusions of my prior study. Warren [1975] examined the of auditing standards by utilizing two uniformity surrogates across nine different groupings-a grouping for each of the Big Eight firms and a category for all others. The two uniformity surrogates were (a) the percentage of qualified opinions to the total number of audit opinions issued, and (b) the percentage of different types of qualified opinions to the total number of qualified opinions issued. The main and interactive effects of the groupings, client size, and industry were explored. To control for possible biasing effects three potentially confounding variables were also inserted into the analysis. These nuisance variables were (1) the existence of extraordinary items, (2) the direction of client's earnings, and (3) the client's fiscal year.1 The results of the original research concluded that ... . the likelihood of receiving a qualified opinion (the first uniformity surrogate) is independent of the auditor. Likewise, given that a client receives a qualification, the likelihood of receiving an uncertainty rather than a consistency exception (the second uniformity surrogate) is also independent of the auditor (Warren [1975]). The client size, industry, and three nuisance variables had significant effects in one or more analyses, but did not interact with any of the groupings. A major shortcoming of the original paper was the lack of a significant data base. The analyses were conducted for a one-year period only and only aggregate data were examined. These data were gathered from 10-