EXTERNAL AUDITOR'S EVALUATION OF INTERNAL AUDIT WORK A CASCADED INFERENCE APPROACH
by
Ganesh Krishnamoorthy
A Dissertation Presented to the
FACULTY OF THE GRADUATE SCHOOL UNIVERSITY OF SOUTHERN CALIFORNIA
Trang 2UMI Number: 9621619
UMI Microform 9621619
Copyright 1996, by UMI Company All rights reserved This microform edition is protected against unauthorized
copying under Title 17, United States Code
UMI
Trang 3THE GRADUATE SCHOOL UNIVERSITY PARK LOS ANGELES, CALIFORNIA 90007
This dissertation, written by Ganesh Krishnamesrthy
Trang 4ACKNOWLEDGMENTS
Tam indebted to many individuals who provided immense support to enable me complete this major project First and foremost, my deepest thanks and gratitude to Professor Theodore J Mock, whose guidance and support has been invaluable Special thanks are also due to my committee members, Professors Ward Edwards, Karen V Pincus, and Paul R Watkins for their input and constructive comments I am also thankful to the Institute of Internal Auditors for the financial support provided during the writing of this dissertation
Trang 5ACKNOWLEDGMENTS LIST OF TABLES LIST OF FIGURES ABSTRACT a ed a ÓÔỐẢÔẨỐ COCO HHO mee em me Hem eH OH HH Emo re meee eH ER Hee mm Ome OD Ce Oe me er Or OO eH OO Om Om eee HO ee eho re mee ro Oe
Come mR Om ee Oe ere Oe me mmm meme ree mee meee O oe
oR Oe ree me ee mm Owe me we re Or Ome moe
2.1 Professional Standards
2.2 External Auditor’s Evaluation and Use of Internal Audit Work 2.3 Conclusions 0 c cece cee e cece eens ececccerecsece
CHAPTER 3: A MULTI-STAGE APPROACH TO THE EVALUATION OF INTERNAL AUDIT FUNCTION
Trang 6iv
CHAPTER 4: A CASCADED INFERENCE MODEL FOR THE
EVALUATION OF INTERNAL AUDIT REPORT .- 76
4.1 Introduction 2.0 2 eee eee eee eee erence 76 4.2 — Graphical Representation eee 79 4.3 Research Questions 83
44 Model Development 84
4.4.1 When IA Reports that the Account is Fairly Stated 88
4.4.2 When IA Reports that the Account is not Fairly Stated 89
4.5 — Analysis and Findings .- co cà 9Ị 46 — Model Extension HQ HS 110 4.6.1 When IA Reports that the Account is Fairly Stated 113
4.6.2 When IA Reports that the Account is not Fairly Stated 114
4.6.3 Analysis 2.02 ee ee cece nh sở Lis 4.7 — Summary and Conclusions 117
CHAPTER 5: SUMMARY AND CONCLUSIONS 5.1 Summary and Implications - 121
5.1.1 Evaluation of Internal Audit Function 121
5.1.2 Evaluation of Internal Audit Report .- 124
5.1.3 Implications for Audit Judgment Research 127
5.1.4 Implications for Audit Practice 130
3.2 — Limitations and Suggestions for Future Research 132
APPENDIX A: Ranking of Factors Used By External Auditors In Evaluating Internal Audit Function: Summary From Selected Studies 136
Trang 7Table 3-1 3-2 4-1 4-2 Convergent Evidence vu ch Conflictng Evidence co Marginal Impact of Each Factor: Convergent Evidence oe ee em em meee oe
Marginal Impact of Each Factor: Conflicting Evidence eee ee eh em ewe ee
Trang 8Figure 3-1 3-2A 3-2B 3-3A 3-3B 3-4 3-5 3-6 4-6 LIST OF FIGURES vi
A Single-Stage Structure for the Evaluation of Internal Audit Function 32
A Multi-Stage Structure for the Evaluation of Internal Audit Function 38
A Multi-Stage Structure for the Evaluation of Internal Audit Function 43
Evaluation of Internal Audit Function When Convergent Evidence is Observed_ 46
Evaluation of Internal Audit Function When Conflicting Evidence is Observed 55
Confiicting Evidence: Impact of Competence 61
Conflicting Evidence: Impact of Work Performance 65
Conflicting Evidence: Impact of Objectivity 69
Evaluation of Internal Audit Report 80
A Cascaded Inference Model For Evaluation of Internal Audit Report 82
Trang 9This dissertation is a theoretical examination of external auditor's evaluation of internal audit work using cascaded inference theory The thesis consists of two major parts The objective of the first part is to understand the relative importance of the three factors (objectivity, work performance, and competence) identified by auditing standards and by prior research, in determining the strength of the internal audit function of a client In addition, the impact of evidence types convergent (all positive) and conflicting (positive and negative) is examined
Results reveal that the importance of evidence relating to the three factors varies with the type of evidence (positive or negative) observed by the external auditor and is contingent on other factors, i.e., on the interrelationships among the factors
Results also reveal that evidence of negative competence has a severe impact on the assessment of internal audit strength when there is positive evidence relating to work performance, but not when there is negative evidence relating to work
performance Results also indicate that auditors should focus more on looking for negative evidence relating to objectivity of the internal auditor Collectively, this suggests that evidence evaluation is non-compensatory when either objectivity or competence is negative Finally, results imply a compensatory processing strategy when all evidence is positive
Trang 11INTRODUCTION
Financial statement auditing (external auditing) has been characterized by many as a judgmental process where the auditor obtains, evaluates, and aggregates evidence about financial statement assertions and accounts with the ultimate objective of expressing an opinion about the fairness of client's financial statements The American Accounting Association Committee on Basic Auditing Concepts, has defined auditing as:
a systematic process of objectively obtaining and evaluating evidence regarding assertions about economic actions and events to ascertain the degree of correspondence between those assertions and established criteria and communicating the results to interested users (ASOBAC
1973, 2)
In order to express an opinion on the financial statements the auditor must collect, evaluate, and aggregate evidence from a number of sources The auditor, however, seldom has the opportunity to directly observe economic events, and must draw inferences from indirect evidence of actions and events Thus, the auditor must deal with multiple stages in the reasoning process between what is observable to the auditor and the ultimate hypotheses of interest
Trang 12prompted the application of cascaded inference theory to examine inferential issues in several disciplines such as law and medicine Ashton et al (1988) have indicated that audit researchers should explore the use of cascaded inference structures for examining normative and descriptive issues in audit inference and decision making
This dissertation examines external auditor's evaluation of internal audit (IA)
work using cascaded inference theory The study uses analytical models to understand the manner in which the external auditor should use the work of internal auditors
External auditors often evaluate internal audit work in the process of formulating an opinion about the fairness of client's financial statements Optimal utilization of internal audit work can improve both the efficiency and effectiveness of external audits and can enhance the value of internal auditors to the client organization
Since the issuance of SAS 9 (AICPA 1975), the relationship between external and internal auditors has been the focus of many studies The Institute of Internal
Auditors has issued a Statement of Internal Auditing Standard (No.5) addressing the
relationship between internal auditors and independent auditors More recently, the AICPA has re-emphasized the importance of this topic with the issuance of SAS 65 (AICPA 1991) which superseded SAS 9 (AICPA 1975) The topic is therefore both important and timely
Trang 13function These factors are competence, objectivity, and work performance of the internal auditor
Professional auditing standards outline several ways in which internal auditors’ work may affect the external audit process First, the external auditor may evaluate the overall strength of the internal audit function with the ultimate objective of evaluating and relying on the internal control structure of the client Second, the external auditor may evaluate internal auditors’ reports relating to specific audit tests and procedures, with a view to assessing the risk of material misstatement In this role, the external auditor may utilize internal auditors as assistants
Almost all prior literature has focused on external auditor's evaluation of the strength of internal audit function Specifically, prior research investigated the relative importance that external auditors place on the three factors (objectivity, competence, and work performance) for assessing the strength of the internal audit function Very few studies' have addressed external auditor's use of internal auditors as assistants in performing external audits
In part due to the lack of a theoretical basis, the results from prior studies have been mixed and inconclusive While some studies suggest that external auditors consider the work performance as the most critical factor (e.g., Brown 1983; Margheim
1986; Schneider 1984), other studies indicate competence as the most significant
Trang 14component (e.g., Messier and Schneider 1988; Schneider 1985a), while yet others reveal the objectivity factor to be the most significant (e.g., Abdel-Khalik, Snowball, and Wragge 1983)
While prior researchers have clearly chosen an important research issue, they have approached it in a rather simplistic manner by focusing on the relative importance of the three SAS 9 factors SAS 9 was intended for professional guidance and was neither a descriptive model of how external auditors actually evaluate the internal audit function nor a normative model of how they ought to evaluate it
It seems reasonable to expect that no single factor will dominate external auditor's evaluation of the internal audit function under all conditions, without regard to the interactions and interrelationships among the factors Audit judgments are complex and we must explicitly consider the potential interrelationships among the factors in order to develop descriptive and normative models in this and other contexts
Trang 15Fourth, notions of cascaded inference have been used in other disciplines (e.g., law,
intelligence analysis) where evidential reasoning is critical, and significant theoretical and practical insights have been obtained Finally, it is well suited for addressing the research issues in this study because the strength of the internal audit function and the process leading to the internal audit report are seldom directly observable to the external auditor
The first major objective of this research project is to understand the relative importance of evidence relating to the three factors (objectivity, work performance, competence) identified by auditing standards and by prior research in evaluating the strength of internal audit function and the effect on the inferential value of evidence, due to interrelationships among these factors This issue is addressed in Chapter 3 In addition, the impact of evidence types convergent (all positive) and conflicting
(positive and negative) is examined ‘n this chapter An important contribution of this
research is to evaluate the strengths of multistage, cascaded inference models,
particularly in explicitly defining the variables and their interrelationships in a given task, and understanding the manner in which evidence types combine to determine the inferential value of evidence
Results in Chapter 3 reveal that the importance of evidence relating to the three
factors (competence, work performance, objectivity) used in the evaluation of the
Trang 16observed by the external auditor, and as expected, contingent on the state of other factors, i.e., interrelationships among factors
First, the importance of evidence relating to competence is examined Evidence suggesting negative competence has a severe impact on the assessment of the internal audit strength when there is positive evidence relating to work performance, but not when there is negative evidence relating to work performance This should be expected because information suggesting that the internal auditors are not competent should reduce the value that can be placed on their work performance and this reduction will be more severe when work performance is positive
Second, analysis reveals little value from evidence relating to work performance if either objectivity or competence is negative This suggests that evidence evaluation relative to work performance is non-compensatory when either objectivity or
competence is negative There are interesting implications for audit practice because evidence is often collected and evaluated in a sequential manner and the results from this analysis suggest that when either competence or objectivity is negative, the value derived from evidence relating to work performance is minimal However, when either objectivity or competence is weak (but positive) or when both objectivity and
Trang 17Finally, results indicate that auditors should focus more on looking for negative evidence relating to objectivity If no negative evidence is found, this analysis suggests that the marginal value of the strength of positive evidence relating to objectivity is likely to have little value
The second major objective of the project is to understand external auditor's evaluation of internal audit reports Chapter 4 addresses this research objective Results of the analysis suggest high sensitivity of the inferential value of internal audit report, to changes in internal auditor reporting bias, but relative insensitivity to changes in internal auditor reporting veracity.’ Further, the inferential value of internal audit report is more sensitive at lower levels of reporting bias than at higher levels Results also suggest that the high sensitivity of the internal audit report to even slight changes in reporting bias is conditional on the level of risk of optimistic judgment, a measure of internal auditor competence, thus suggesting significant interaction effects between reporting bias and the risk of optimistic judgment The inferential value of the internal audit report is much more sensitive to variations in reporting bias at lower levels of the risk of optimistic judgment than at higher levels
Results of the analysis also revealed that the inferential value of internal audit report is significantly more sensitive to the risk of optimistic judgment than to the risk of pessimistic judgment Further, the sensitivity of the internal audit report to changes in
Trang 18the risk of optimistic judgment is not contingent on the level of either the risk of pessimistic judgment or internal auditor's reporting veracity, thus suggesting no significant interaction effects Results also indicate that the inferential vaiue of the internal audit report does not vary significantly with changes in the risk of pessimistic judgment
Finally, results suggest that audit procedure diagnosticity is critical in evaluating the value of internal audit report When audit procedure diagnosticity is low, the inferential value of internal audit report is minimal, irrespective of the levels associated with other parameters in the model The implications of these findings are discussed in Chapter 5
Trang 19LITERATURE REVIEW
The nature of this study requires review of both professional standards and academic research Section 2.1 provides a review of the professional standards Section 2.2 discusses the extant research concerning external auditors’ evaluation and use of internal audit work Finally, Section 2.3 provides concluding comments for this chapter
2.1 Professional Standards
In 1991, the American Institute of Certified Public Accountants (AICPA) issued Statement on Auditing Standards (SAS) No 65: The Auditor's Consideration of the Internal Audit Function in an Audit of Financial Statements SAS 65 superseded SAS 9: The Effect of an Internal Audit Function on the Scope of Independent Audit, which was in effect since its issuance in December 1975
The major objective of SAS 65 was to provide the external auditor "with guidance on considering the work of internal auditors and on using internal auditors to provide direct assistance to the auditor in an audit performed in accordance with generally accepted auditing standards."
Trang 2010 appropriately plan the nature, extent, and timing of audit tests and procedures (SAS 55:
Consideration of the Internal Control Structure in a Financial Statement Audit) SAS
55 details the procedures that the external auditor must follow to obtain an
understanding of the client's internal control structure and indicates that the internal audit function is a part of an entity's control environment Thus, the strength of the internal audit function provides information relating to the internal control structure of the client
SAS 65 states that when considering “how the internal auditor's work might affect the nature, timing, and extent of audit procedures, the auditor should assess the competence and objectivity of the internal audit function in light of the intended effect of the internal auditors’ work on the audit.” When evaluating the competence of the internal auditors, SAS 65 states that the external auditor should consider such factors as the educational level, professional certification, and professional experience of the internal auditors; practices regarding assignment of internal auditors; supervision and review of internal auditors’ activities; and evaluation of internal auditors’ performance
When evaluating the objectivity of the internal auditors, SAS 65 states that the organizational status of the internal auditors should be considered including whether the organizational level to which the internal auditors report is sufficiently high so as to ensure adequate audit coverage and consideration of actions on findings and
Trang 21relating to areas audited by the internal auditor, including policies specifically
preventing internal auditors from auditing areas where they were assigned recently or where they are likely to be assigned in the near future
While SAS 65 outlines the manner in which each factor impacting internal audit function may be evaluated, it does not specify how the factors should be weighted and combined in order to determine the overall strength of the internal audit function Further, SAS 65 does not provide guidance on the interrelationships among the factors (objectivity, competence, and work performance) and how such an interrelationship should affect assessment and reliance on the internal audit function
Other professional accounting bodies, both within and outside the United States, have deliberated on this issue The Institute of Internal Auditors has issued a Statement of Internal Auditing Standard (No.5) outlining the relationship between internal and external auditors Professional institutes in other nations have recognized
the importance of this issue: the Australian Accounting Research Foundation (1983) has issued a standard for using the work of an internal auditor (AUP 2); The Canadian
Trang 2212 2.2 External Auditor's Evaluation and Use of Internal Audit Work
This section discusses academic studies that relate to how external auditors evaluate, weight, and combine the three factors (objectivity, competence, and work performance) in evaluating and utilizing internal audit work
Gibbs and Schroeder (1979) was the first study that sought to determine the criteria that auditors use in evaluating the three factors specified by SAS 9 Based on surveys and experiments involving 148 CPAs and 111 internal auditors, a ranking was elicited to determine the importance auditors place on a number of variables when evaluating internal auditor competence, objectivity, and work performance factors ANOVA revealed that two variables were more important in explaining variance related to internal auditor competence: (1) the internal auditors’ knowledge of company operations; and (2) supervision over internal audit staff The organizational level to which internal audit department reports was the most significant factor in determining internal auditor objectivity Finally, management's readiness to act on internal audit recommendations and management's support for internal auditors were the two most important factors related to evaluating the work performance of the internal auditors
Clark, Gibbs, and Schroeder (1979) extended the work of Gibbs and Schroeder (1979) but focused on examining auditors’ judgments of internal auditor objectivity " The experimental cases were administered to 25 partners and managers of CPA firms,
Trang 23level to which internal audit staff reports, and ability to investigate any area of company's activity were the three most important factors identified, in that order
Thus, both Gibbs and Schroeder (1979) and Clark, Gibbs, and Schroeder (1979) focused on external auditors’ evaluation of internal auditor competence,
objectivity, and work performance These studies, however, did not address how these factors are weighted and combined
Brown (1983) performed one of the first studies directed at examining the manner in which external auditors weight and combine the three factors in determining the reliance to be placed on the internal audit function
Experimental packages were mailed to 101 subjects from four large public accounting firms Subjects were provided with a brief description of a manufacturing firm and were then given preanswered ("yes" or "no") for each of the six questions in the study These questions were selected based on discussion with practicing auditors Two questions related to the competence factor, three related to the work performance factor, and one question related to the objectivity factor
Subjects were required to make judgments relating to the degree of reliability that they would place on the internal audit function based on each of the 48 different combinations of the six questions presented to them.’ The experiment did not require the subjects to make audit planning decisions
Trang 24
14 Results revealed work performance (specifically, the question relating to whether or not the internal audit function was deemed satisfactory in prior year audits) and objectivity of the internal auditors were the most important factors, in that order Competence was a distant third, explaining only 9 percent of the variance Further, Brown did not find any significant interaction effects between the factors
Abdel-Khalik, Snowball, and Wragge (1983) examined the impact of three EDP audit techniques (the test data method, generalized audit software technique, and integrated test facility) and two organizational variables relating to internal auditor objectivity (organizational independence and internal audit review of program changes), on the extent to which internal auditors are utilized as assistants by the external
auditors and the extent to which it impacts planned testing by the external auditors The subjects in the study were audit seniors and mangers from large public accounting firms, with experience in audit of computerized information systems Two experiments were conducted where subjects were required to make judgments about planning audit programs for accounts receivable and accounts payable
Results indicated that: (a) the objectivity factor, operationalized as the
Trang 25Schneider (1984) attempted to identify descriptive models that can help represent the manner in which external auditors combine the factors identified in SAS 9 The level of consensus among external auditors was also examined
This study consisted of three experiments The first experiment consisted of a 2° complete factorial design to test which of the five characteristics’ was considered most important in determining each of the three factors in SAS 9 (Competence, Objectivity, and Work performance) Each subject rated profiles using a ten-point
ordinal scale Multidimensional scaling (MDS) was used to determine the
characteristics that accounted for maximum variance relating to a given factor The purpose of the second experiment was to select realistic and materially different levels for the characteristics identified in the first experiment For each factor, eight different levels for each of the top two characteristics identified in Experiment | were used so that subjects could make dissimilarity judgments among the levels MDS analysis was used in this experiment to identify four levels for each of the two
characteristics that were perceived by the subjects to have maximum discriminatory power Finally, the third experiment required the subjects to rate (on a 100 point ordinal scale) the strength of the internal audit function based on 64 profiles.®
° These characteristics were based on results of prior studies, SAS 9, and input from audit practitioners
Trang 2616 Subjects were 26 auditors from seven of the Big Eight auditing firms in the Midwest Questionnaires for all three experiments involved a hypothetical
manufacturing company which was financially strong Subjects were required to make evaluations determining the nature, extent, and timing of external auditors’ work related to the revenue cycle
The results of these experiments revealed that auditors viewed work
performance as the most important factor, followed by the competence and objectivity factors Work performance variables accounted for 42% of the explained variance, followed by the variables relating to the competence factor which accounted for 35% of the variance Finally, the objectivity factor variables accounted for 23% of the explained variance
Schneider (1985a) examined the relationship between external auditors’
evaluation of the internal audit function and their reliance in audit planning decisions The objectives of this research were twofold First, Schneider attempted to understand the relative importance of factors that external auditors use in evaluating the internal audit function Second, external auditor's reliance on the internal audit function in adjusting the audit plan, was examined
Trang 27the second experiment, the same subjects were asked to provide reliance judgments on the same [6 profiles Reliance decision was operationalized as the "extent of total audit work” relating to the revenue cycle
Conjoint analysis revealed the following weights placed on the three factors: Competence: 36.8%; Objectivity: 25.4%; and Work Performance: 37.8% The results implied that auditors perceived competence and work performance factors to be almost equally important, and the objectivity factor to be less but nevertheless a significant factor Further, results revealed that external auditors did rely on internal auditors and reduced their work by approximately 38%
Schneider (1985b) examined the degree of consensus among external auditors in evaluating the internal audit function Eighteen audit managers, non-randomly selected from large public accounting firms participated in this study Subjects were asked to rate various case profiles relating to an internal audit function An additive model was presumed appropriate and a compensatory processing strategy was implied
Trang 2818 concluded that there is significant consensus among auditors in evaluating internal audit function
Margheim (1986) examined the factors that external auditors consider
important in their decision to rely on internal auditors and if such reliance results in adjustment of their audit plan Internal auditor's reliability was defined as internal auditor competence-work performance and objectivity Competence and work performance were treated as one factor in order to avoid unrealistic combinations such as low competence and high work performance
Margheim used a between-subjects design in an experiment that required external auditors to complete an audit program plan (evaluation and reliance decision) for the accounts receivable area of a hypothetical company Internal auditor
competence/work performance and objectivity (independent variables) were varied across cases These two factors were tested at high and low levels, together with a control case with no internal auditor involvement, resulting in a five case between- subjects design
Experimental materials were mailed to 1000 Big Eight auditors from 30 largest metropolitan areas (200 subjects randomly assigned to each case version) Subjects were required to budget audit hours (dependent variable) to compliance and
Trang 29Results revealed that external auditors did not rely on internal auditors to adjust planned audit hours in the low competence-work performance condition However, they did reduce planned audit hours in response to high competence-work performance level of internal auditors Finally, they were insensitive to the degree of internal auditor objectivity and no significant interaction effects were found between competence-work performance and objectivity factors
Messier and Schneider (1988) examined the relative importance of the
attributes that are considered important by external auditors in evaluating the internal audit function A hierarchy of attributes was developed and the Analytical Hierarchy
Process (AHP) was used in the this study
Based on previous research and with the aid of practicing external auditors, attributes at three levels were identified using the Schneider (1984, 1985a) case Subjects were then asked to evaluate each set of attributes in a pairwise fashion using Saaty's nine point response scale Twenty-two subjects from Big-eight accounting firms participated in this study with an average experience of 7.6 years Each subject had the experience of making at least one reliance decision on internal audit function in the past
Trang 3020 significant measure of objectivity Finally, scope of internal audits was considered the most important measure of work performance
Harrell, Taylor, and Chewning (1989) examined the ability of the management of a firm to bias the objectivity of the intemal auditors The task required the subjects to evaluate the strength of payroll internal control subsystem after providing six decision cues, similar to the ones used by Ashton (1974) Each participant made 32 decisions in a 2° X 1/2 factorial design Fifty-eight internal auditors with an average experience of 7 years participated in the study Thirty-five of these were members of the Institute of Internal Auditors
Participants were randomly assigned to either the experimental group (n=30) or the control group (n=28) Experimental materials provided to the two groups were identical except that participants in the experimental group were provided a statement from the bank's management desiring that equal weights be placed on the six factors and they were also provided a consonant outcome feedback indicating the judgments that would result from employing equal weights Thus, management's specification of the weights was an attempt to bias the professional objectivity of the internal auditor
Trang 31who have knowledge of management's desired internal control system evaluations, the Institute of Internal Auditors (IIA) members will be less susceptible to agree with management than those who are not ITA members
Results provided support for the first hypothesis Results also supported the second hypothesis thus indicating that although management can bias the objectivity of internal auditors, ITA members are much more likely to resist such pressure from the management than are non-ILA members
Edge and Farley (1991) was the first academic study conducted outside the United States The major objective of the study was to determine the relative importance of factors that external auditors in Australia place in evaluating client's internal audit function In addition, the study also examined the consistency with which external auditors made this evaluation
A questionnaire required external auditors to make judgments on numerous scenarios representing different configuration of factors of an internal audit function The independent variables used in the study were: 1 Organizational status; 2 Scope of function; 3 Technical competence; 4 Due Professional care; and 5 Previous Audit work The first four factors are deemed important per Australian Statement of
Auditing Practice (AUP) No 2, while the fifth factor was included due to the results of
Brown (1983) The dependent variable was a reliance decision made by auditors
("Degree of use to be made of work of internal audit function") on a 7-point Likert
Trang 3222 of audit plan in terms of the nature, extent, and timing of tests), it is not clear as to what the subjects may have responded to
The experimental package included all possible Yes/No combinations of the independent variables resulting in a 25 (32) responses In addition 16 combinations were repeated in order to make stability assessments The experimental package was sent to 406 members of audit staff (computer audit staff were excluded) of four large audit firms in Melbourne Two hundred and five responses were received resulting in
184 usable questionnaires
Results revealed that technical competence was considered the most important factor by the auditors, followed by work performance.’ “Previous audit work" was considered the third most dominant factor Objectivity (organizational status) was considered the least important of the factors Consistent with the findings of Brown (1983), a high degree of consensus was found across the subjects
Finally, Maletta (1993) investigated external auditors’ decisions to use internal
auditors as assistants Specifically, the study investigated the impact of inherent risk on the extent to which internal audit objectivity, competence, and work performance affect auditors' decisions to use internal auditors as assistants
This is the first study to recognize that auditors use a complex configural decision process and that the importance of any given factor should be contingent upon
other factors Maletta (1993) states:
Trang 33
For example, if low-quality internal auditors assist the independent auditor in performing certain audit tests, the overall strength of the tests will be lower because the results of those tests will be less reliable Therefore, the effect of using internal auditors as assistants on the strength of the test should be a function of the quality of the internal auditors, which should depend not only on the objectivity and prior work of the internal auditors but also on their competence
However, Maletta's hypotheses were not guided by a theoretical model capable of predicting the conditions under which significant interaction effects will be observed
Maletta utilized audit managers from Big Six accounting firms with at least four years of audit experience Inherent risk was manipulated as a between-subject variable, while internal auditor competence, objectivity, and work performance were
manipulated as within-subject variables Subjects were required to react to a series of cases requiring audit planning decisions in the context of accounts receivable of a medium-sized manufacturing company The dependent variable was the percentage of budgeted audit hours assigned to internal auditors
Results indicate that when inherent risk is high, auditors seem to use a complex configural process and consider the impact of work performance of the internal auditors only when objectivity is high In low inherent risk conditions, no significant interaction effects were found between work performance and objectivity Further, the results indicated that across the inherent risk conditions, internal audit competence was the most significant of the three factors, followed by objectivity and work performance
Maletta and Kida (1993) investigated the effect of environmental factors on configural
Trang 3424 While this study did not directly address the issue of how auditors evaluate the strength of internal audit function and how the three factors are combined and weighted, their study made significant contribution in understanding the manner in which evaluation of internal audit function impacts audit planning decisions Their results indicated that external auditors rely on internal auditors based on a configural relationship between inherent risk and control strength and that consideration of environmental risk factors is critical in understanding the link between internal audit function and audit planning decisions
Finally, Whittington and Margheim (1993) examine the impact of inherent risk, materiality, and the subjectivity of evidence, on external auditors’ reliance on internal auditors A case study experiment involving managers in a large accounting firm was conducted where subjects were required to make judgments about relying on internal auditors to perform extemal audit work Inherent risk and materiality were
manipulated on a between-subjects basis at two levels: high and low The subjects completed a audit time allocation sheet indicating the amount of audit time to be assigned to internal and extemal audit staff
Contrary to Maletta (1993), the results in this study indicate that inherent risk factors are not significant Whittington and Margheim (1993) entertain the possibility that this may be due to the manner in which they operationalized the variable relating to
Trang 352.3 Conclusions
A review of the literature reveals that most studies have focused on
understanding the relative importance that external auditors place on the three factors (objectivity, competence, and work performance), and the manner in which these factors are combined to form an overall assessment of the strength of the internal audit function All prior research has employed a descriptive, experimental approach, with no formal model guiding the research hypotheses Based on the results of studies discussed in this chapter, Appendix A depicts the ranking of factors used by external auditors in evaluating the internal audit function
Appendix A reveals that the results from prior studies have been inconclusive While one study (Abdel-Khalik, Snowball, and Wragge 1983) reveals that objectivity is the most significant factor, others (Brown 1983; Schneider 1984; Schneider 1985a; Schneider [985b) reveal that the work performance factor is the most significant of the factors More recent studies (Margheim 1986; Messier and Schneider 1988; Edge and Farley 1991; Maletta 1993), however, reveal that external auditors consider
competence as the most significant factor in evaluating the internal audit function Thus, a total of nine studies has not lead us to a state when we can even make tentative conclusions about how external auditors seem to weight and combine these factors in order to make assessments relating to the internal audit function
Trang 36design while others employed a between-subjects (e.g., Margheim 1986) design Studies using different designs may not always be comparable due to possible "demand effects" (Pany and Reckers 1987) in a within-subjects context
Second, the nature of the task required of the subjects in the experiments varied from requiring auditors to “evaluate” the strength of the internal audit function based on a profile of cases (e.g., Brown 1983), to asking auditors to make “reliance and planning” decisions in determining the nature, extent, and timing of audit tests (e.g., Margheim 1986), although some studies involved both (e.g., Schneider 1985a) Since studies in other decision contexts (e.g., Biggs, Mock, and Watkins 1988; Mock and Wright 1993) reveal that reliance and audit planning decisions need not correlate with evaluation judgments, the results of these studies must be interpreted with caution
Third and perhaps most importantly, I believe that prior research has approached this issue in a rather simplistic manner by largely ignoring the interrelationships that should exist among factors specified in SAS 9 and SAS 65 While simple, parsimonious models are desirable and sometimes outperform
complicated models in explaining or predicting phenomenon, simplistic approaches can be perilous in understanding complex judgments In fact, some early studies called for more research that can consider the complexity involved in this issue Brown (1983,
Trang 37Further studies in this area should not attempt to deal with these factors (objectivity, competence, and work performance] on an isolated basis because auditors tend to view an intermingling relationship among key internal audit function characteristics and evaluative dimensions
reported in SAS No 9 (bracket added)
Despite these calls, research in this area has not addressed the complexities but has made simple additive assumptions For example, Schneider (1985b, 299) states:
additive model was presumed to be an appropriate representation of auditors’ judgments on the internal audit function: Additivity implies that all three SAS No 9 criteria would have independent and
compensating effects as if one could just add values of ‘competence’, ‘objectivity’ and ‘work’ together to arrive at a value of overall internal audit strength
Trang 38CHAPTER 3
A MULTI-STAGE APPROACH TO THE EVALUATION OF INTERNAL AUDIT FUNCTION
3.1 Introduction
External auditors routinely evaluate the strength of the internal audit (LA)
function of the client organization, with the ultimate objective of understanding and evaluating the internal control structure of the client The focus of this chapter is to model external auditor's evaluation of the internal audit function as a multistage
inferential task A significant amount of prior research has examined this issue Almost all prior research has been descriptive The hypotheses were largely derived from professional literature (e.g., SAS 9, SAS 65) which prescribed the three most important factors that external auditors must consider in evaluating the IA function These factors were competence, objectivity, and work performance of the internal auditor
Trang 39Perhaps due to the lack of a stronger theoretical basis, the results from prior studies have been mixed and inconclusive While some studies suggest that external auditors consider work performance as the most critical factor (e.g., Brown 1983; Margheim
1986; Schneider 1984), other studies reveal competence as the most significant
component (e.g., Messier and Schneider 1988; Schneider 1985), while yet others reveal the objectivity factor to be the most significant (e.g., Abdel-Khalik, Snowball, and Wragge 1983)
The rest of the chapter is organized as follows Section 3.2 discusses the research issue The model is developed in Section 3.3 Section 3.4 presents the analysis and findings, followed by summary and conclusions in section 3.5
3.2 Research Issue
Prior research has clearly chosen an important issue, but the approaches employed were theoretically simplistic We should expect that no single factor will dominate the external auditor's evaluation of the IA function under all circumstances and in all situations, without regard to the interaction and interrelationship among the factors For instance, ordinarily an external auditor may increase his belief in the proposition that the LA function is strong, if the work performance’ is satisfactory
Trang 40
30 However, if the external auditor learns that the internal auditor is not objective,’ the inferential value from work performance should be reduced because the internal audit work may not be reliable
Audit judgments are complex and we must explicitly consider the potential interrelationships among the factors in order to develop descriptive and normative
models in this and other contexts Messier and Schneider (1988, 351) recognize the
importance of the interrelationships:
While further research should attempt to empirically derive the interrelationships among cues, our recognition of interrelationships represents an advancement over past internal auditing research, which has not addressed specifically the issue of how the cues relate to each other
Therefore, the following research question is addressed in this chapter: What is the relative importance of evidence relating to the three factors (objectivity, work performance, competence) identified by auditing standards and by prior research, and what is the effect on the inferential value of evidence, due to interrelationships among
these factors?