Sampling Practices of Internal Auditors at Corporations on the Standard & Poor's Toronto Stock Exchange Composite Index*

Abstract

The purpose of this study is to find out how often statistical and nonstatistical audit sampling practices are used by internal auditors in companies listed on the Standard and Poor's (S&P) Toronto Stock Exchange (TSX) Composite Index and how such practices are related to the training and background of the respondents. We adapted the questionnaire used by Hall, Hunton, and Pierce (2002) in their survey of U.S. auditors in public accounting, industry, and government.Although 20 percent of companies responding do not have an internal audit department, the other 80 percent use statistical methods to plan sample sizes 15 percent (± 5 percent) of the time, random sample selection methods 23 percent (± 5 percent) of the time, but statistical evaluation methods only 10% (± 4%) of the time. Despite the low percentage use, almost half of the respondents reported substantial training in statistical sampling and evaluation methods. Moreover, we found statistically significantly higher proportions of respondents with substantial training in audit sampling methods among companies crosslisted on U.S. exchanges compared with companies listed only on the TSX.Finally, respondents with a chartered accountant designation tend to have a negative impact on the use of statistical methods in audit sampling, and companies cross-listed on U.S. exchanges tend to have larger internal audit departments than companies listed only on the TSX.

Priors in a Bayesian audit: How integration of existing information into the prior distribution can improve audit transparency and efficiency

“…Recently, Yoon and Pearce (2020) argue that such analytical procedures and statistical sampling both have unique benefits, and that they are best used to complement each other. Statistical sampling therefore remains a much-used technique due to its efficiency in larger populations (Christensen et al, 2015;Hitzig, 1995; International Auditing and Assurance Standards Board (IAASB), 2018; Maingot & Quon, 2009;Srivenkataramana, 2018;van der Nest et al, 2015). After all, statistical sampling enables the auditor to infer a conclusion about a certain characteristic of the population based on only a small subset of this population.…”

Priors in a Bayesian Audit: How Integration of Existing Information into the Prior Distribution Can Improve Audit Transparency and Efficiency

“…Recently, Yoon and Pierce (2020) argue that both analytical procedures and statistical sampling have unique benefits, and that they are best used to complement each other. Statistical sampling remains a much used technique due to its efficiency in larger populations (Christensen, Elder, and Glover 2015;Hitzig 1995;IAASB 2018, Maingot andQuon 2009;Srivenkataramana 2018;van der Nest et al 2015). After all, statistical sampling enables the auditor to infer a conclusion about a certain characteristic of the population based on only a small subset of this population.…”

The use of generalised audit software by internal audit functions in a developing country: A maturity level assessment

“…Internal auditors had to revisit the manner in which they collect audit evidence in order to achieve the predetermined engagement objectives in an efficient manner. The internal auditor can collect audit evidence for the purpose of evaluating the effectiveness of internal controls in one of the following three ways:  Full population testing (i.e., a 100 percent examination of all items in the population with the use of technology-based tools such as GAS);  Selection of specific items (i.e., the inclusion of specific items from the population based on predetermined criteria), and  Random sampling for the purpose of performing a statistical test (often called statistical sampling) or simply for generalization and/or extrapolation (often called non-statistical sampling) (Maingot & Quon, 2009:218; Aghili, 2011:21; AICPA, 2012:23; Smidt, 2014:85; IAASB, 2015 ISA 530 par.5 (g)).…”