This preview shows page 1. Sign up to view the full content.
Unformatted text preview: AUDITING: A JOURNAL OF PRACTICE & THEORY Vol. 24, No. 2 November 2005 pp. 4967 The Impact of Management Integrity on Audit Planning and Evidence
Timothy G. Kizirian, Brian W. Mayhew, and L. Dwight Sneathen, Jr.
SUMMARY: This study uses working paper data from 60 clients of a U.S. Big 4 auditing firm to directly examine the influence of auditor-assessed management integrity on auditor's assessments of risk of material misstatement, audit planning, and audit outcomes. We hypothesize that management integrity is related to preliminary risk assessments, and to the persuasiveness, timing, and extent of planned audit procedures. We also hypothesize that the management integrity assessment influences the auditor's evaluation of source credibility of management provided evidence beyond its influence via risk assessments (Beaulieu 2001). Finally, we expect the auditor's management integrity assessment to be associated with the discovery of client misstatements. When we add management integrity to empirical models motivated by Mock and Wright (1993, 1999), we find support for our hypotheses. However, all but two hypothesized associations disappear when we add an indicator variable for prior-year errors. We continue to find that management integrity impacts the persuasiveness of evidence sought beyond what is suggested by the auditor's risk assessment. Interestingly, even after controlling for prior-year errors, we find an inverse association between the auditor's assessment of management integrity and the likelihood of detecting misstatements, suggesting the management integrity assessment aids the auditor in ultimately discovering errors. The results support the importance of assessing management integrity in planning the audit and discovering misstatements. Keywords: audit risk model; management integrity; evidence. Data Availability: A confidentiality agreement with the data-granting firm precludes revealing its identity or disseminating the data. T INTRODUCTION his study uses archival data from 60 audits conducted by a U.S. Big 4 accounting firm to examine the association between the external auditor's assessment of management integrity, and the auditor's assessment of risk of material misstatement Timothy G. Kizirian is a Professor at California State University, Chico, Brian W. Mayhew is an Assistant Professor the University of WisconsinMadison, and L. Dwight Sneathen, Jr. is an Assistant Professor at Georgia Southern University.
We appreciate comments from the referees, Bill Messier, Kathy Hurtt, and Larry Rittenberg. We thank the datagranting firm for its generous provision of data. Funding from The University of Arizona Foundation and Department of Accounting is gratefully acknowledged. Submitted: December 2003 Accepted: June 2005 49 50 Kizirian, Mayhew, and Sneathen (RMM), audit planning, and the discovery of financial statement misstatements. Prior archival research provides weak support for the linkage between risk assessments and auditrelated judgments (Mock and Wright 1993, 1999). Our study provides a stronger test than prior studies. First, we use data taken directly from the working papers rather than through questionnaires. Second, we employ a key internal control characteristic (i.e., management integrity). Last, we focus on a combined measure of audit risk.1 We also go a step further than prior archival research by linking the management integrity assessment through the audit process to the discovery of misstatements. Experimental research generally indicates a linkage between risk assessments and audit planning, suggesting that auditors adjust their decisions based on evidence concerning management integrity (Beaulieu 2001; Ayers and Kaplan 1998; Schaub 1996). Our paper extends this research to an archival setting to examine the impact of auditor assessments of management integrity on audit planning. We examine this association in a field setting to get feedback on the relative importance of management integrity assessments in field-based settings. Experiments often focus on a single variable of interest. While experiments are useful in identifying potential influences and developing basic theory, they are limited in their ability to determine the relative importance of the treatment variable in more complex field-based settings. Our archival approach potentially enhances existing experiment-based theory. The linkage between an auditor's assessment of management integrity and auditor assessed risk of material misstatement (RMM), audit planning, and misstatement detection is important for three reasons. First, management integrity is a key determinant of the client's risk structure. It provides the foundation of internal control. Without management integrity, or the ``tone at the top,'' it is unlikely even the most proficient internal controls will be effective in reducing financial statement misstatement. As a result, it is important to evaluate evidence that auditors incorporate this risk component into their audit judgments. Second, auditors rely on management to provide a great deal of evidence during the course of the audit. The auditor must carefully evaluate management integrity to assess source credibility relating to client-supplied evidence. We examine whether auditors adjust the evidence source in response to lower management integrity. Finally, the recent enactment of the Sarbanes-Oxley Act (U.S. House of Representatives 2002), especially Section 404, requires that auditors evaluate and report on client internal controls, including management integrity. This requirement will increase the amount of information available to auditors in making evidential planning decisions. Auditors, therefore, need to understand the linkages between management integrity, audit risk, and evidence. We hypothesize that the auditor's management integrity assessment directly impacts the auditor's assessment of RMM. RMM is the combination of inherent risk (IR) and control risk (CR). We do not attempt to separate IR and CR as management integrity influences both components. After establishing a link between management integrity and RMM, we extend the analysis to evaluate management integrity's impact on audit planning beyond its indirect effect through RMM. Specifically, we test whether management integrity impacts the persuasiveness, timing, and extent of audit evidence while controlling for the auditor's RMM assessment. Management integrity impacts the perceived reliability (i.e., source credibility) of evidence gathered from management. This assessment can result in changes in the persuasiveness, timing, and extent of evidence gathered beyond the impact caused by
1 The use of a combined measure is consistent with authoritative guidance (i.e., SAS No. 47) and the more recent AICPA Exposure Draft--Proposed Statements on Auditing Standards, December 2, 2002 (AICPA 2002c). Auditing: A Journal of Practice & Theory, November 2005 The Impact of Management Integrity on Audit Planning and Evidence 51 increased RMM. Finally, we extend our analysis to the identification of financial statement misstatements. An association between management integrity and misstatement discovery is consistent with both auditors' accurately assessing management integrity and the increased RMM assessment. We find partial support our hypotheses. Auditor-assessed management integrity is negatively related to the auditor's RMM assessment (i.e., high integrity is related to low risk), but only when we exclude an indicator for prior-year error from the model. If we include prior-year errors in the model, then the management integrity assessment no longer influences RMM. This is consistent with prior literature that shows auditors rely more on a simple heuristic of whether there was a prior-year error in their overall risk assessment, rather than on a specific evaluation of management integrity (Wright and Ashton 1989; Kinney 1979). Our analysis of audit planning parallels the findings for RMM with one notable exception. After controlling for RMM and prior-year errors, it appears the management integrity assessment incrementally influences the planned persuasiveness of evidence. It appears that auditors adjust to their management integrity assessment by seeking more external evidence when management integrity is low. Finally, we find that the likelihood of discovered misstatements is higher for clients with low auditor-assessed management integrity, even after controlling for prior period errors and RMM. The paper proceeds as follows. The next section outlines our hypotheses. The third section describes the data and the empirical tests. The fourth section provides results and the fifth section summarizes our findings. HYPOTHESES We use the audit risk model to link the auditor's assessment of management integrity to RMM, then link the auditor's RMM assessment to the persuasiveness, timing, and extent of audit procedures. We then hypothesize that the impact of management integrity on the reliability of management-supplied evidence leads to an effect on audit planning beyond RMM. Finally, we look at whether the auditor's management integrity assessment is associated with the magnitude of audit differences discovered. Management Integrity and Risk Assessments The auditors in our study arrive at a management integrity measure by investigating factors relating to management's attitude toward reporting, controls, and the external audit, as well as their reputation in the business community. The firm's guidance states: ``Our evaluation [of management integrity and control consciousness] provides an indication of whether management is reliable and reputable, and therefore, whether representations and disclosure may reasonably be expected to be meaningful.''2 The Committee of Sponsoring Organizations (COSO) frames the ``tone at the top'' as arguably the most important component of an internal control structure (COSO 1999). Without management's willingness to lead and conduct follow-up procedures to ensure compliance with a strong internal control structure, it is unlikely that even the most proficient internal controls will be effective in detecting and preventing financial statement misstatements. The auditing standards include management integrity and control consciousness as key attributes to which auditors should be attentive (SAS No. 47, 53, 78, 82, 96; AICPA 1983,
2 The firm refers to their assessment as ``management tone.'' We use the more descriptive ``management integrity'' label throughout the paper. Auditing: A Journal of Practice & Theory, November 2005 52 Kizirian, Mayhew, and Sneathen 1988, 1995, 1997, 2002a). SAS No. 47 discusses the association between management integrity and IR and CR in greater detail, where inherent risk is defined as the ``susceptibility of an assertion to a material misstatement, assuming that there are no related controls.'' The assessment of inherent risk considers the client's operations, industry, the engagement, and characteristics of management. The inherent risk assessment takes management integrity into consideration as it pertains to the full and fair disclosure of financial information. When management integrity is assessed as high (i.e., high integrity), ceteris paribus, we expect inherent risk to be assessed at a lower level. SAS No. 47 defines control risk as ``the risk that a material misstatement that could occur in an assertion will not be prevented or detected on a timely basis by internal controls.'' Factors considered in the internal control assessment include the degree of competence of personnel, levels of oversight (i.e., board of directors and audit committee activities), the assignment of authority and responsibility, and management's commitment to the process of providing accurate financial information.3 The control environment promoted by management sets the tone of an organization, influencing the control consciousness of the entity's employees, and is the foundation for all other components of internal control (AICPA 2000). Therefore, control risk should be negatively related to management integrity. SAS No. 99 (AICPA 2002b) provides a conceptual basis for management integrity to impact auditor professional skepticism and resulting audit risk assessments.4 SAS No. 99 requires the auditor to adjust the audit plan based on fraud risk and consideration of the auditor's reliance on management-supplied evidence. Fraud risk is impacted by three components, incentive, opportunity, and attitude, which allow the individual to commit the fraud (SAS No. 99). The auditor's assessment of management integrity provides an indirect measure of management's attitude toward fraud. SAS No. 99 calls on the auditor to adjust the level of professional skepticism based on the assessed fraud risk. This includes adjusting the RMM (SAS No. 99 paras. 4648, 50). Based on our review of SAS No. 47 and 99, it appears that management integrity should impact RMM. This discussion leads to the first hypothesis, stated in alternative form: H1: The auditor's assessment of management integrity is inversely related to the risk of material misstatement assessment. Audit Planning The audit risk model is presented below. It is used to direct the auditor's audit planning. AR where: AR Audit Risk--the risk the auditor may unknowingly fail to modify his opinion on financial statements that are materially misstated (SAS No. 47, AU 312.02); RMM * DR 3 4 Internal control consists of five interrelated components for which management is responsible: the control environment, risk assessment, control activities, information and communication, and monitoring (SAS No. 78, AICPA 1995). While SAS No. 99 was not in effect at the time of the audits examined in this paper, it was a reflection of prior standards that were in place at that time. Auditing: A Journal of Practice & Theory, November 2005 The Impact of Management Integrity on Audit Planning and Evidence 53 RMM DR IR*CR, where IR is inherent risk and CR is control risk; and Detection Risk--risk the auditor will not detect a material misstatement. SAS No. 47 and 99 suggest that assessed RMM should directly affect the requisite persuasiveness, timing, and extent of audit evidence sought by the auditor to achieve a desired DR. Given an assessed RMM and a target AR, the auditor plans the persuasiveness, timing, and extent of audit procedures to achieve the necessary DR. The audit risk model theoretically guides the selection of audit procedures, producing an effective and efficient, riskfocused evidential plan that assists auditors in the opinion formulation process. The persuasiveness of audit procedures employed relates to the persuasiveness of the evidence required (AICPA 2000). A client with a high RMM requires the collection of more persuasive evidence and more direct observation than a client with a low assessed RMM. The timing of evidence collection reflects the auditor's reliance on the client's internal control structure, such that evidence gathered prior to year-end is expected to reasonably reflect year-end results and processes. The auditor uses more pre-year-end procedures when the client has strong internal controls. Finally, the connection between RMM and the extent of testing is direct. As RMM increases, the auditor needs to collect more evidence to reduce DR, to reach a specified level of AR. Source Credibility Management integrity also impacts the auditor's assessment of source credibility for management-provided evidence. The source credibility literature focuses on auditor judgments based on the reliability of evidence provided. Theories on aggregating evidence provide various statistical and behavioral belief adjustment models to determine how given evidence should be aggregated based on its reliability (see Krishnamoorthy et al. [1999)] for a discussion of these models). Hirst (1994) looks at how the reliability of evidence impacts judgment. The consistent theme from this stream of research is that less reliable information should be weighted less in the auditor's judgment process than information that is more reliable. Experiments appear to back up this contention. The auditor must consider the credibility of evidence supplied by management and adjust the audit plan accordingly (Beaulieu 2001, 1994). If management is deemed less credible (i.e., low integrity), then the auditor will either require more evidence to offset the lower weighting on the evidence, or require evidence be gathered from a different source that is more credible than management. Both options attempt to achieve at least the same level of confidence as when management is seen as a credible evidential source (i.e., has high integrity). A similar theme runs through SAS No. 99. As the RMM due to fraud increases, the standard calls for adjustments to the nature, timing, and extent of audit procedures (SAS No. 99, para. 48). SAS No. 99 discusses a dual effect on professional skepticism, both in terms of the overall attitude toward the RMM (captured in H1), and also through the questioning of evidence and information obtained from the client. SAS No. 99 appears to suggest that the auditor specifically adjust skepticism toward the evidence obtained. Paragraph 52 discusses in some detail the impact on the nature (we call persuasiveness), timing, and extent of audit procedures. Given that the management integrity assessment is a significant consideration in the planning phase, we explore whether management integrity has a dual effect on audit testing both through its impact on the auditor's assessment of RMM, and through its impact on the credibility of evidence supplied by management. We expect that as the assessment of Auditing: A Journal of Practice & Theory, November 2005 54 Kizirian, Mayhew, and Sneathen management integrity decreases, we should see the auditor react by placing a lower weighting on evidence provided by management. All things equal, we should see the auditor respond to low management integrity by increasing the extent of evidence collected, collecting more persuasive evidence from sources other than management, and replacing interim data with year-end data that more reliably reflects conditions as of the balance sheet date. As such, we test the following hypotheses: H2p: The auditor's assessment of management integrity is inversely related to the persuasiveness of audit procedures incremental to the risk of material misstatement assessment. The auditor's assessment of management integrity is inversely related to the timing of audit procedures (the proportion of audit testing performed at year-end) incremental to the risk of material misstatement assessment. The auditor's assessment of management integrity is inversely related to the extent of audit procedures incremental to the risk of material misstatement assessment. H2t: H2e: Misstatements We trace the management integrity assessment through the audit process to address our final question: Is management integrity associated with the magnitude and discovery of material misstatements? We expect management integrity to be associated with the likelihood and magnitude of misstatements for two reasons. First, the auditors perform a management integrity assessment to evaluate the likelihood of misstatements. We expect more misstatements when the auditor believes management to be of low integrity. Second, low management integrity assessments result in higher RMM assessments. Auditors adjust the audit plan accordingly, and therefore are more likely to find misstatements. In other words, the auditor tests more transactions or uses more persuasive audit tests, thereby increasing the probability of identifying more misstatements. Both explanations produce the same predictions, potentially compounding the effect. This leads to the last hypothesis: H3: The auditor's assessment of management integrity is inversely associated with the likelihood of detected misstatements. PROPRIETARY DATA, VARIABLE MEASUREMENT, AND MODEL SPECIFICATION Proprietary Data The financial statement audit documentation used in this study was acquired from a U.S. Big 4 firm. The firm granted access to its archived audit working paper records for a practice office that has a primarily technology client base.5 Sample audits were selected, by one of the authors, from a list of archived engagements using a random number generator. The archives contain audit files from 1996 to 1999. The firm provided audit data for 78 audit engagements from 78 different clients.6 The firm did not document a management integrity assessment on 18 of the clients, which resulted in a final sample of 60 clients
5 6 As a condition to accessing this data we agreed not to disclose the identity of this firm. We retained all rights to publish our findings without any review or approval from the supplying firm. The number of observations selected was not intentional, rather a result of time constraints imposed upon the researcher. Auditing: A Journal of Practice & Theory, November 2005 The Impact of Management Integrity on Audit Planning and Evidence 55 used in our tests.7 Within the sample, 54 clients are publicly traded. The firm has been auditing these clients for an average of 7 years (see Table 1). The data set does not contain any first-year audits. All sampled engagements received unqualified opinions, none of the sample clients has failed as of 2001, and there were no restatements or known allegation of audit failure as of 2001. The firm assisted in coding the variables used, and provided a subsequent multilevel review to facilitate consistent coding of the data.8 The data-granting firm considers revenue to be a ``significant account.'' Several sources corroborate the perceived higher risk of errors and fraud within the revenue cycle (e.g., COSO 1999). For example, SAS No. 99 requires the auditor to presume there is a high fraud risk related to revenue. The firm considers an account or cycle to be ``significant'' if it is critical to the prevention of audit failure, or if its examination will significantly decrease audit firm risk. Given these characteristics, revenue is a primary area in which risk assessment and related audit testing can vary, and therefore provides a more powerful setting to study management integrity decisions. Assessed risk and audit planning measures are drawn from the revenue cycle audit programs. The revenue cycle includes procedures applied to sales and accounts receivable. We include audit procedures applied to the recording of sales, accounts receivable and collection of cash as part of the revenue cycle examined. Variable Measurement and Model Specification To test H1, we employ the following equation: RMM
0 1 5 *MI 2 *REVENUE
6 3 *TENURE e. 4 *PYERR (1) *PUBLIC *INDUSTRY RMM is defined as the combination of ranked auditor-assessed revenue cycle inherent and control risks (SAS No. 47). The data-granting firm measures and documents inherent and control risk assessments as ``low,'' ``medium,'' or ``high.'' The risks are coded 1 for low, 2 for medium, or 3 for high. A multiplicative combination would result in risk of material misstatement values of 1, 2, 3, 4, 6, and 9.9 For purposes of interpretation, we use a ranking mechanism to create a RMM metric that facilitates using OLS as follows:10 7 8 9 10 Of the 78 client observations included in our sample, 18 did not receive a detailed control risk assessment, and therefore did not include a management integrity (MI) evaluation. There may have been an informal evaluation, but it was not documented in the working papers we examined. Descriptive statistics suggest the 18 firms are more risky, have shorter audit tenure, and are more likely to be private firms than the 60 firms with MI assessments. In additional analysis, if we assume that all 18 firms are low-MI firms consistent with auditors treating the firms as if they are high-risk firms and not formally evaluating MI, we find even stronger support of our hypotheses. As such, we feel that removing these firms from our analysis is conservative and if anything biases against us finding results. We did not perform any tests for inter-coder reliability. Instead we relied on the firm's help in coding the variables to insure they measure accurately the constructs we examine. The multiplicative RMM metric is not measured on a continuous scale and therefore a multinomial logit may be a more fitting statistical model than OLS. However, the marginal effects of the regressors on the probabilities are not equal to the coefficients, rendering coefficient interpretation to be difficult. More importantly, the model will fail to account for the ordinal nature of the dependent variable (Greene 1997). Nonetheless inferences based on a multinomial logit results are the same as the inferences based on our OLS approach. We thank William Waller and Mark Zimbelman for suggesting the combination approach. Prior research asked survey respondents to rate risk on a seven-point scale even though the firm they surveyed used the same classification scheme as our firm (Mock and Wright 1993, 1999). Our results do not change if we multiply our IR and CR coding rather than the ranking used above. Our ranking scheme makes no distributional assumptions other than increases in the qualitative measures of risk increase the quantitative measures. A multiplicative approach assumes that risk increases dramatically as the qualitative risk increases. Auditing: A Journal of Practice & Theory, November 2005 56
IR Medium 2 3 4 Kizirian, Mayhew, and Sneathen Low Low Medium High 1 2 3 High 3 4 5 CR The test variable, management integrity (MI), is assessed and documented as ``strong,'' ``moderate,'' or ``weak.'' This assessment is coded 2 for strong, 1 for moderate, and 0 for weak.11 The data-granting firm's management integrity assessment is made prior to the assessment of control risk and inherent risk. The assessment attempts to capture management's integrity and control consciousness. Auditors arrive at a management integrity measure by investigating factors related to management's attitude toward reporting, controls, and the external audit, as well as their reputation in the business community. The datagranting firm's guidance states ``the reputation, character, and integrity of senior management and the tone set for the overall control environment provides us with an indication of whether they are responsible and reputable, and whether meaningful representations and full disclosure can be reasonably expected during the engagement.'' Where RMM is the dependent variable, we expect the coefficient on MI to be negative, indicating that higher integrity assessments contribute to a lower preliminary risk assessment. We include several controls to mitigate the potential for a correlated omitted variable problem. We focus on known determinants of RMM. Prior literature has noted that the length of the auditor-client relationship may affect risk assessments due to learning over time (O'Keefe et al. 1994; Ashton 1991). We control for this by including the number of years the auditor has been auditing the client (TENURE). As a result of the potential for learning over time, we expect the coefficient on TENURE to obtain a negative coefficient. There is no obvious association between TENURE and MI although we expect that the auditor's assessment of MI to be more accurate as TENURE increases. To control for client size, we include the natural log of total client revenue (REVENUE) for the year under audit. Risk and audit effort measures used in this study are revenue cycle-specific, rendering the inclusion of REVENUE to be a fitting size control. While larger firms may have greater oversight leading to potentially lower control risk assessments, they may have more complex control structures and greater decentralization, potentially increasing CR. Prior literature has shown that the relationship between auditor effort and client size is nonlinear (O'Keefe et al. 1994). To address this issue, we utilize the natural log of revenue. It is unclear how these effects will aggregate and affect the RMM-REVENUE relationship or how size would influence MI. Accordingly, we do not have an expectation of the sign on REVENUE. Prior research documents a strong association between the discovery of prior-year errors and current period errors (Kinney 1979), and between prior-year errors and current year risk assessments (Mock and Wright 1999). We include a dummy variable labeled PYERR that is coded as 1 if there was a documented prior year audit difference. PYERR also can impact the assessment of management integrity (MI). This variable is a very simple risk measure based on the discovery of a prior-year misstatement. Twenty-five percent of our sample firms have a prior-year misstatements, and approximately 50 percent of those have misstatements in both years.
11 Alternatively using an indicator variable approach with two indicator variables to capture the three levels of MI yields similar inferences to the single variable approach we report in the paper. Auditing: A Journal of Practice & Theory, November 2005 The Impact of Management Integrity on Audit Planning and Evidence 57 Prior research suggests the auditor is more likely to be sued if the client is publicly held (e.g., St. Pierre and Anderson 1984). Additionally, incentives to overstate financial standing and results of operations are suggested to be greater for managers of public firms due to market driven compensation (O'Keefe et al. 1994). We control for this by including an indicator variable (PUBLIC) that takes on a value of 1 for public firms, and 0 otherwise. We expect PUBLIC to obtain a positive association with RMM. To control for potential systematic differences between industry groups, we include an indicator variable representing the two industry categories in our sample (biotech and hightech) (INDUSTRY). Because we do not have any evidence concerning major changes in audit approach by the data-granting firm between industry groups, there is no expectation for the coefficient on INDUSTRY. An auditor may address specific client characteristics by adjusting the evidential plan. This can be accomplished by altering the persuasiveness of evidence collected, the timing of evidence collection, and the extent of audit procedures conducted. To test H2, we examine the marginal impact of MI on the persuasiveness, timing, and extent of audit procedures independently when RMM is included in the model by employing OLS analysis in the following form: 12 PERSUASIVENESSi
0i 1 4 7 MIi 2 RMMi
5 3 REVENUEi
6 TENUREi INDUSTRYi
3 PYERRi PUBLICi (2) ei;
4 TIMINGi 0i 1 5 MIi PYERRi MIi 2 RMMi
7 TENUREi ei; (3) PUBLICi
4 EXTENTi 0i 1 5 2 RMMi
7 TENUREi ei. (4) PYERRi PUBLICi INDUSTRYi Prior literature has looked at specific audit tests, and used the total number of tests to measure the nature of audit procedures (e.g., Mock and Wright 1999). This study uses a metric for nature that focuses on characteristics defined in SAS No. 31 (AICPA 1980) and by the Public Oversight Board's Panel on Audit Effectiveness. When referring to the nature of audit evidence, SAS No. 31 states that evidence collected from independent sources is more reliable than information collected directly from the client. The Panel on Audit Effectiveness specifically uses the term ``persuasive'' when referring to the nature of audit evidence. We depart from prior research because we believe authoritative guidance suggests nature should reflect the independence and persuasiveness of evidence. We label our nature variable PERSUASIVENESS. We measure PERSUASIVENESS by categorizing audit procedures listed in the revenue cycle of the audit program using a three-point scale (outside evidence 3, outside/inside 2, and inside 1). ``Outside'' evidence is obtained, inspected, or observed completely independent of management. This is considered to be the most persuasive form of evidence.
12 Because client characteristics may simultaneously affect each dependent variable, the residuals from these three regressions may also be correlated, implying that Seemingly Unrelated Regression (SUR) should be utilized to improve the efficiency of the coefficient estimation. However, the SUR estimator reduces to OLS when the independent variables are identical for all equations (Amemiya 1985, 187). Auditing: A Journal of Practice & Theory, November 2005 58 Kizirian, Mayhew, and Sneathen An example of outside evidence is a cash confirmation sent directly from a bank to the auditor. ``Outside/inside'' evidence originated from a third party outside of the influence of management but has the potential to be manipulated by management. An example of outside/inside evidence is a bank statement sent by the bank to the client. This bank statement originated from the bank but passed through the hands of the client, and thus had the potential to be manipulated by the client. ``Inside'' evidence is obtained directly from the client. An example is a bank reconciliation prepared by the client. PERSUASIVENESS is the average of these rankings for all procedures listed in the revenue cycle audit program. Where PERSUASIVENESS is the dependent variable, we expect the coefficient on MI to be negative, indicating the auditor requires less persuasive evidence when management integrity is high. Persuasiveness is a complex construct to try to capture in a single variable. We acknowledge that an internal versus external categorization is a significant simplification; however, it captures a key determinant of persuasiveness--the data source. Our approach is also easy to replicate as it is straightforward to objectively implement. In the sensitivity section, we consider an alternative measure based on the sum of the individual audit procedure scores rather than the average reported in our main results. The dependent variable TIMING equals the proportion of audit testing hours conducted at the client's fiscal year-end relative to total audit testing hours.13 Audit testing hours includes both substantive tests and test of controls to the exclusion of time spent on audit planning and final engagement wrap-up. If, for a given audit, 60 percent of audit testing hours were conducted at the client's fiscal year-end, and 40 percent during interim work, then the observation receives a TIMING score of 60. When TIMING is a dependent variable, we expect the coefficient on MI to obtain a negative value, indicating the auditor requires less evidence collected at year-end in the presence of strong management integrity. The dependent variable EXTENT equals total revenue cycle audit hours divided by total audit hours. This variable is measured in the same manner as in Mock and Wright (1999). It captures the portion of total audit hours allocated to the revenue portion of the audit.14 We expect the portion of total audit time allocated to revenue to increase as the risk of the revenue cycle increases. When EXTENT is a dependent variable, we expect the coefficient on MI to be negative, indicating the audit requires less evidential effort in the presence of strong management integrity. To test H3, we employ logistic regressions in the following form: DIFFi DIFFi
0i 1 5 0i MIi PYERRi MIi PYERRi 2 RMMi
6 3 REVENUEi
7 4 TENUREi ei; EXTENTi (6) (5) TENUREi
4 1 5 9 2 RMMi
10 INDUSTRYi ei. TIMING or PERSUASIVENESS 13 14 Unlike our other dependent variables, the timing variable is not based on the revenue cycle alone. Management integrity also effects engagement and account level assertions and risks, so we expect an association between the MI assessment and timing regardless of whether timing captures account specific or engagement level assertions or risks. Note also that the data set does not include any cases where the auditor changed the proximity of the year-end engagement relative to the client's fiscal year-end. We also scaled revenue hours by total revenue and alternatively by the log of total revenue. These alternative measures give a more direct test of the change in the revenue account relative to the average effort exerted across all audits of the revenue cycle in our sample. Inferences based on these alternative measures do not differ from the reported results for EXTENT. Auditing: A Journal of Practice & Theory, November 2005 The Impact of Management Integrity on Audit Planning and Evidence 59 The dependent variable DIFF is an indicator variable where 1 indicates that the auditor discovered an audit difference during the audit. The DIFF measure is for revenue cycle audit differences only. These audit differences represent potential misstatements that should be adjusted or investigated further, not necessarily errors in the financial statements. The auditor collected all the differences identified during the audit regardless of magnitude to consider the potential aggregate effect of all identified differences.15 These were the actual differences the auditor discussed with client management. When DIFF is a dependent variable, we expect the MI coefficient to be negative, suggesting that a weak MI assessment is associated with more audit differences. Equation (5) examines the impact of MI on the probability of discovering current period errors without consideration of EXTENT, TIMING, or PERSUASIVENESS. To better assess the incremental impact of MI on discovering error, we control for MI's influence on audit planning. We examine Equation (6) where we include EXTENT, TIMING, or PERSUASIVENESS in the model. RESULTS Descriptive Statistics Table 1 provides descriptive statistics on our dependent and independent variables. Mean MI is 1.40 and only 9 (18) of the 60 firms have MI scores of 0 (1). Hence, the median firm in the sample has high management integrity and, on average, auditors assessed client integrity as moderate. This result is consistent with findings from the client acceptance literature that suggests auditors screen out very high-risk clients (Johnstone and Bedard 2003). The mean RMM of 1.50 suggests that auditors on average assessed the risk of material misstatement as low. The mean of PERSUASIVENESS (1.62) suggests that on average auditors sought information from a mix of internal and external sources, but relied more heavily on internal sources. The mean of TIMING (79.67) suggests on average a majority of the procedures were performed at year-end. Auditors identified an average of $180,973 in audit differences, but over two-thirds of the clients (69 percent) did not have any audit differences. The frequency of differences is consistent with prior research (for a summary see Eilifsen and Messier ). Table 2 presents the correlation between all the variables. The significant correlation between MI and RMM supports H1. The significant correlation between MI and PERSUASIVENESS, TIMING, and EXTENT support H2p, H2t, and H2e. Finally, as expected the negative correlation between MI and DIFF supports H3. While we note many significant correlations, all VIFs in our subsequent analyses are well below 10. Marquandt (1980) argues that a multicollinearity problem exists if VIF values exceed 10. Management Integrity and Auditor Effort Table 3 presents OLS regression results for Equations (1)(4). In Equation (1), MI 0.357) is insignificant with RMM as a dependent variable. MI is highly ( 0.034, p significant if we exclude PYERR from the model. Table 2 shows that PYERR and RMM are very highly correlated (r 0.94, p 0.001). PYERR is also highly correlated with MI
15 Some concern was raised that this model needs to control for the tolerable error (or materiality threshold of the auditor). We do not think this is the case as the auditor collects all identified misstatements to consider the overall impact on the financial statements. Nonetheless, we included the tolerable error in our regression and it had no impact on our inferences. Auditing: A Journal of Practice & Theory, November 2005 60 Kizirian, Mayhew, and Sneathen TABLE 1 Descriptive Statistics (n 60) Mean MI RMM PERSUASIVENESS TIMING EXTENT DIFF REVENUE TOTALREVENUE TENURE PYERR PUBLIC INDUSTRY 1.40 1.50 1.62 79.67 0.07 0.31 17.92 441,124,554 6.85 0.25 0.90 0.63 Standard Deviation 0.74 0.89 0.32 13.94 0.04 0.46 1.95 1,186,191,750 4.30 0.44 0.30 0.49 Quartile #1 1 1 1.40 70 0.03 0 16.31 12,167,500 4 0 1 0 Median 2 1 1.54 80 0.05 0 17.81 54,557,000 6 0 1 1 Quartile #3 2 2 1.93 90 0.08 1 18.99 177,326,538 8 1 1 1 MI is an auditor assessment of management's integrity and commitment to full and fair disclosure of financial information measured on a three-level categorical scale where MI takes on a value of 2, 1, or 0, where 2 represents high levels of integrity. Risk of material misstatement (RMM) is a metric measured on a scale from 1 5 where 1 is low risk and 5 is high risk. PERSUASIVENESS is the average persuasiveness score given to audit procedures in the revenue cycle (the persuasiveness score is given on a three-level scale: 3 for outside information, 2 for outside / inside information, and 1 for inside information). TIMING is the percentage of audit testing conducted at the end of the fiscal year of the client. EXTENT is the total hours spent performing test of controls and substantive testing in the revenue cycle during the audit divided by total audit hours. DIFF is an indicator variable where 1 indicates that the auditor discovered an audit difference during the evidential process. The control variables are: natural log of revenue (REVENUE), tenure as auditor (TENURE), prior year, revenue cycle, audit differences (PYERR), public or private ownership (PUBLIC), and industry categorization (INDUSTRY). TOTALREVENUE is total revenue, and is included for descriptive purposes only. as one would expect (r 0.58, p 0.001). These results suggest that the main driver of risk assessment is prior-year error. Prior-year error is a key determinant of MI, and it dominates MI when both variables are included in the RMM model. Equations (2)(4) indicate that when PERSUASIVENESS is the dependent variable MI is significantly negative ( 0.092, p 0.019), however when TIMING ( 1.154, p 0.252) 0.175) are the dependent variables the coefficient on MI is and EXTENT (0.005, p insignificant. RMM is significant when PERSUASIVENESS (0.136, p 0.019) and EXTENT 0.001) are dependent variables, but not when TIMING (3.153, p 0.111) is (0.049, p the dependent variable. This finding indicates that the auditor responds to low MI by varying the persuasiveness of audit tests requiring more independent (i.e., persuasive) evidence beyond what RMM would suggest. This evidence provides support for H2p. Examination of the change in adjusted R2 when MI is removed from the model supports our inferences concerning the marginal impact of MI beyond the effect of RMM (adjusted R2 decreases from 66.39 percent to 60.5 percent).16 Due to the source credibility issue, the auditor makes a change in the audit program that will increase the cost of the audit (i.e., more independent
16 Adjusted R2 does not change significantly when MI is removed from the H2t and H2e models consistent with no marginal effect of MI when RMM is included in the respective models. Auditing: A Journal of Practice & Theory, November 2005 The Impact of Management Integrity on Audit Planning and Evidence TABLE 2 Spearman Correlations (n 60) MI RMM PERSUASIVENESS Auditing: A Journal of Practice & Theory, November 2005 TIMING EXTENT REVENUE PYERR TENURE PUBLIC INDUSTRY DIFF 0.596 .001 0.573 .001 0.572 .001 0.335 0.009 0.433 .001 0.584 .001 0.373 0.003 0.220 0.091 0.140 0.285 0.761 .001 RMM 1.000 0.750 .001 0.750 .001 0.671 .001 0.509 .001 0.943 .001 0.486 .001 0.083 0.530 0.149 0.255 0.506 .001 PERSUASIVENESS TIMING EXTENT REVENUE PYERR TENURE PUBLIC INDUSTRY 0.750 .001 1.000 0.566 .001 0.543 .001 0.327 0.011 0.310 0.016 0.470 .001 0.032 0.807 0.061 0.643 0.500 .001 0.750 .001 0.566 .001 1.000 0.576 .001 0.702 .001 0.723 .001 0.682 .001 0.150 0.253 0.300 0.020 0.327 0.011 0.834 .001 0.677 .001 0.647 .001 1.000 0.456 .001 0.592 .001 0.292 0.024 0.015 0.257 0.191 0.018 0.397 0.002 0.509 .001 0.327 0.011 0.702 .001 0.456 .001 1.000 0.495 .001 0.603 .001 0.305 0.018 0.521 .001 0.247 0.057 0.943 .001 0.310 0.016 0.723 .001 0.592 .001 0.495 .001 1.000 0.431 .001 0.064 0.626 0.200 0.126 0.517 .001 0.486 .001 0.470 .001 0.682 .001 0.292 0.024 0.603 .001 0.431 .001 1.000 0.104 0.430 0.290 0.025 0.194 0.137 0.083 0.530 0.032 0.807 0.150 0.253 0.015 0.257 0.305 0.018 0.064 0.626 0.104 0.430 1.000 0.092 0.483 0.251 0.053 0.149 0.255 0.061 0.643 0.300 0.020 0.191 0.018 0.521 .001 0.200 0.126 0.290 0.025 0.092 0.483 1.000 0.002 0.985 DIFF 0.506 .001 0.500 .001 0.327 0.011 0.397 0.002 0.247 0.057 0.517 .001 0.194 0.137 0.251 0.053 0.002 0.985 1.00000 61 62 Kizirian, Mayhew, and Sneathen TABLE 3 OLS Regression Analysis Test of H1: RMM
0 1 5 *MI 2 *REVENUE
6 3 *TENURE e 4 *PYERR (1) *PUBLIC *INDUSTRY Test of H2p: PERSUASIVENESS
0 1 5 *MI *PYERR 2 *RMM
6 3 *REVENUE
7 4 *TENURE e (2) *PUBLIC *INDUSTRY Test of H2t: TIMING
0 1 5 *MI *PYERR 2 *RMM
6 3 *REVENUE
7 4 *TENURE e (3) *PUBLIC *INDUSTRY Test of H2e: EXTENT
0 1 *MI 2 *RMM 3 *REVENUE 4 *TENURE e (4) 5*PYERR 6*PUBLIC 7*INDUSTRY Independent Variables and Expected Signs Intercept MI RMM REVENUE TENURE PYERR PUBLIC INDUSTRY Adjusted R2 ? ? RMM (H1) 1.974 0.002 0.034 0.357 Dependent Variables PERSUASIVENESS TIMING (H2p) (H2t) 1.063 0.001 0.092 0.019 0.136 0.018 0.023 0.207 0.008 0.112 0.271 0.021 0.054 0.255 0.018 0.766 66.39% 118.650 0.001 1.154 0.252 3.153 0.111 2.136 0.005 1.044 0.001 7.827 0.070 1.165 0.370 0.543 0.823 71.06% EXTENT (H2e) 0.023 0.535 0.005 0.175 0.049 0.001 0.002 0.286 0.000 0.244 0.025 0.053 0.016 0.061 0.008 0.233 70.74% 0.045 0.250 0.012 0.204 1.700 0.001 0.123 0.257 0.224 0.082 79.51% Risk of material misstatement (RMM) is a scale from 15 where 1 is low risk and 5 is high risk. PERSUASIVENESS is the average independence score given to audit procedures in the revenue cycle where the independence score is on a three-level scale: 3 for outside information, 2 for outside / inside information, and 1 for inside information. TIMING is the percentage of audit testing conducted at the end of the fiscal year. EXTENT is the total hours spent performing test of controls and substantive testing in the revenue cycle divided by total audit hours. MI is the auditor's assessment of management's integrity measured on a three-level categorical scale of 2, 1, or 0, where 2 represents high levels of integrity. Control variables include natural log of revenue (REVENUE), years as auditor (TENURE), prior year, revenue cycle, audit differences (PYERR), public or private ownership (PUBLIC), and industry (INDUSTRY). Statistical significance for parameter estimates (p-values) are for one-tailed tests with directional expectations and two-tailed tests where there is no directional expectation. The Impact of Management Integrity on Audit Planning and Evidence 63 information is generally more costly to collect) by going outside the client for data verification instead of simply increasing the analysis of the client's information. Equations (1) (4) were also run including two additional control variables YR (the year of the observation) and ROA (return on assets, included as a profitability measure). The results are robust to the inclusion of these variables. The variables are not significant in the regressions and therefore are not included in the primary analysis. Management Integrity and Audit Differences Table 4 presents logistic regression results testing H3. The coefficient on MI is signif0.001) suggesting that MI is associated icant and negative in Equation (5) ( 4.332, p with the likelihood of discovering audit differences. As discussed earlier this finding may be the by-product of more diligent testing for low-integrity assessments, so we attempt to control for the extent of testing to isolate the relationship between misstatement discovery and integrity assessments. Equation (6) includes PERSUASIVENESS, TIMING, and EXTENT as independent variables to control for the potential effect of increased audit effort 0.015). on misstatement discovery. In this equation, MI remains significant ( 9.891, p This evidence provides support for H3. It appears that although the RMM and management integrity assessments are strongly influenced by PYERR, the MI assessment outperforms PYERR as a significant predictor of discovering differences in the current year. Sensitivity Analysis We examine the sensitivity of our results to (1) our use of a combined risk measure, (2) the use of two different metrics for persuasiveness, (3) a scaled measure of the magnitude of audit differences, (4) the inclusion of a fraud risk assessment variable, and (5) the inclusion of an indicator variable for the presence of prior-year misstatements. We rerun all of our empirical tests using either IR or CR in place of RMM. The untabled results of using either IR or CR as the risk measure are consistent with the findings for H1 using RMM. Mock and Wright (1993, 1999) use the Number of Tests conducted (NofTests) to measure the nature of evidence, whereas, we attempt to measure the persuasiveness of evidence. When we use NofTests we obtain similar results to those presented in Table 3 in tests of H2p, MI is marginally significant ( 0.626, p 0.107). It appears that NofTests is a better proxy for the extent of procedures rather than the persuasiveness of evidence. In support of this conjecture, there is a high correlation between the NofTests and the extent measures (0.73, p 0.001). We also use a cumulative independence score as a measure for persuasiveness rather than the average. This approach produces a measure of persuasiveness that captures both the number of procedures and the independence of each procedure (i.e., 1, 2, or 3). Our untabled inferences are unchanged from those reported in Table 3 for tests of H2p. When we change DIFF from an indicator variable to the total documented audit differences scaled by total revenue, the untabled results are consistent with those presented in Table 4. Further, when IR and CR are used in place of RMM as a control with the scaled DIFF as a dependent variable, the results continue to be consistent with Table 4. All but one of the misstatements are overstatements, so our results reflect the association between the management integrity assessment and overstatements. In our hypotheses development, we discussed SAS No. 99 on fraud and linked the MI assessment to the management attitude aspect of the fraud risk. The audit firm included a fraud risk assessment for the audits we examine. Fraud risk is an indicator variable coded 1 if high fraud risk and 0 if low fraud risk. We included this measure in our analysis to
Auditing: A Journal of Practice & Theory, November 2005 64 Kizirian, Mayhew, and Sneathen TABLE 4 Logistic Regression Analysis Tests of H3: DIFF
0 1 5 *MI *PYERR *MI *PYERR 2 *RMM
6 3 *REVENUE
7 4 *TENURE e (5) *PUBLIC
4 DIFF 0 1 5 9 2 *RMM
10 *INDUSTRY e *EXTENT (6) *TIMING or *PERSUASIVENESS Dependent Variable Independent Variables and Expected Signs Intercept MI RMM REVENUE TENURE PYERR PUBLIC INDUSTRY PERSUASIVENESS TIMING EXTENT ? ? ? ? ? DIFF (H3) 4.917 0.493 4.332 0.001 0.540 0.399 0.523 0.246 0.046 0.400 2.277 0.161 3.277 0.109 1.721 0.345 DIFF (H3) 37.471 0.176 9.891 0.015 3.420 0.204 1.087 0.122 0.784 0.094 14.456 0.091 10.655 0.086 5.395 0.519 3.843 0.482 0.448 0.084 125.7 0.074 Dependent variable DIFF equals 1 when the auditor discovered an audit difference during the evidential process and 0 otherwise. All other variables are defined in Table 3. Statistical significance (p-values) for parameter estimates are for one-tailed tests with directional expectations and two-tailed tests where there is no directional expectation. investigate whether it impacted the significance of MI. As expected MI is negatively cor0.66, p .001). When fraud risk is added to our models, it related with fraud risk (r does not have a significant effect on our inferences concerning MI. Fraud risk itself is not significant in any of the models. SAS No. 99 was not in effect during the audits we examine. Future research should revisit the impact on fraud risk assessments on audit planning postSAS No. 99. Auditing: A Journal of Practice & Theory, November 2005 The Impact of Management Integrity on Audit Planning and Evidence 65 SUMMARY AND CONCLUSIONS In light of recent management-induced business failures (e.g., Enron) and the SarbanesOxley-related Section 404 requirement that auditors certify internal controls, the importance of recognizing and considering the integrity of client management has taken on renewed importance. This study examines the influence of the auditor's assessment of management integrity on preliminary risk assessments and the allocation of audit effort. Our results suggest that clients with higher assessed levels of integrity have lower preliminary risk assessments, but prior-year error better explains risk and planning assessments than the management integrity assessment. Nonetheless, our findings indicate that management integrity exhibits incremental explanatory power beyond the risk of material misstatement for the persuasiveness of audit evidence collected. This result suggests that when the client's information is not trustworthy, the auditor seeks external validation of the financial statement information instead of pursuing additional scrutiny of client-supplied evidence. It also suggests that adjustments to the audit risk model are not the only way auditors adjust audit planning when conducting an audit. Auditors appear to adjust the persuasiveness of evidence sought beyond that indicated by their revised audit risk assessment. We also examine the direct linkage between assessed management integrity and misstatements discovered during the course of the audit. Our results indicate that after controlling for overall audit effort and prior-year errors, management integrity is associated with the discovery of current period misstatements. Management integrity goes beyond the ability of prior-year differences to predict the likelihood of discovering misstatements in the current year. Our findings are especially important given that most audit firms assess management's integrity during client acceptance. Generally most firms believe they ``only accept clients with high management integrity.'' Therefore, one can argue that the fact we observe variation in management integrity assessments and that the variation appears to be systematically associated with both audit planning and audit outcomes is particularly interesting. It appears that while severe cases of low management integrity may be weeded out during client acceptance, auditors nonetheless retain clients with a spectrum of management integrity that must be managed within the audit process. That is, management integrity matters even when auditors only accept clients with high integrity. The data utilized in this study brings with it certain limitations. The data was drawn from a single Big 4 firm's practice office that audits predominantly technology-oriented clients and includes only audits of technology firms. Prior research on fraud suggests that high-tech firms have a higher incidence of fraud than other industries (Beasley et al. 2000). This suggests that the variance in MI we observe in this study may not be present more generally in the population. Further, the relationship between management integrity and the risk of material misstatement may be affected by the nature of the audit personnel making the assessments. We were unable to identify the individual who made each assessment, and therefore cannot investigate this issue. The relationships in this study may also be affected by variation in board of director characteristics or management ownership. Data limitations prevent us from addressing these issues as well. The proxy utilized for the persuasiveness of audit testing is unique to this study. Prior research has used the number of audit procedures as a measure of the nature of procedures. We deviate from prior literature in an attempt to capture the true nature or persuasiveness of procedures. Clearly future research can attempt to improve on this variable by developing a multidimensional measure of persuasiveness. The results suggest that management integrity impacts auditors' risk assessment and the actual discovery of misstatements by the auditor. However, the current paper does not
Auditing: A Journal of Practice & Theory, November 2005 66 Kizirian, Mayhew, and Sneathen attempt to identify the determinants of the auditor's assessments of management integrity. Future research can examine these determinants to help better understand what influences auditor assessments of integrity. REFERENCES
Amemiya, T. 1985. Advanced Econometrics. Cambridge, MA: Harvard University Press. American Institute of Certified Public Accountants (AICPA). 1980. Evidential Matter. Statement on Auditing Standards No. 31. New York, NY: AICPA. ------. 1983. Audit Risk and Materiality in Conducting an Audit. Statement on Auditing Standards No. 47. New York, NY: AICPA. ------. 1988. The Auditor's Responsibility to Detect and Report Errors and Irregularities. Statement on Auditing Standards No. 53. New York, NY: AICPA. ------. 1995. Consideration of Internal Control in a Financial Statement Audit: An Amendment to Statement on Auditing Standards No. 55. Statement on Auditing Standards No. 78. New York, NY: AICPA. ------. 1997. Consideration of Fraud in a Financial Statement Audit. Statement on Auditing Standards No. 82. New York, NY: AICPA. ------. 2000. Panel on Audit Effectiveness Oversight Report and Recommendations. Stamford, CT: FASB. ------. 2002a. Audit Documentation. Statement on Auditing Standards No. 96. New York, NY: AICPA. ------. 2002b. Consideration of Fraud in a Financial Statement Audit. Statement on Auditing Standards No. 99. New York, NY: AICPA. ------. 2002c. Exposure Draft (of Seven Statements on Auditing Standards Related to Audit Risk). New York, NY: AICPA. Ashton, A. H. 1991. Experience and error frequency knowledge as potential determinants of audit expertise. The Accounting Review 66 (2): 218239. Ayers, S., and S. Kaplan. 1998. Potential differences between engagement and risk review partners and their effect on client acceptance judgments. Accounting Horizons (2): 139153. Beasley, M. S., J. V. Carcello, D. R. Hermanson, and P. D. Lapides. 2000. Fraudulent financial reporting: Consideration of industry traits and corporate governance mechanisms. Accounting Horizons 14 (4): 441454. Beaulieu, P. R. 1994. Commercial lenders' use of accounting information in interaction with source credibility. Contemporary Accounting Research 11 (1): 557585. ------. 2001. The effects of judgments of new clients' integrity upon risk judgments, audit evidence, and fees. Auditing: A Journal of Practice & Theory 20 (2): 85100. Committee of Sponsoring Organizations (COSO). 1999. Fraudulent Financial Reporting: 1987 1997--An Analysis of U.S. Public Companies. By the National Commission on Fraudulent Financial Reporting. New York, NY: AICPA. Eilifsen, A., and W. F. Messier. 2000. The incidence and detection of misstatements: A review and integration of archival research. Journal of Accounting Literature 19. Greene, W. H. 1997. Econometric Analysis. Third edition. Upper Saddle River, NJ: Prentice Hall. Hirst, D. E. 1994. Auditors' sensitivity to source reliability. Journal of Accounting Research 32 (1): 113126. Johnstone, K. M., and J. C. Bedard. 2003. Risk management in client acceptance decisions. The Accounting Review 78 (4): 10031025. Kinney, W. R. 1979. The predicted power of limited information in preliminary analytical review: An empirical study. Journal of Accounting Research 17 (Supplement): 148165. Krishnamoorthy, G., T. Mock, and M. Washington. 1999. A comparative evaluation of relief revision models in auditing. Auditing: A Journal of Practice & Theory 18 (2): 104126. Auditing: A Journal of Practice & Theory, November 2005 The Impact of Management Integrity on Audit Planning and Evidence 67 Marquandt, D. 1980. A critique of some ridge regression methods. Journal of the American Statistical Association 75 (369): 8791. Mock, T., and A. Wright. 1993. An exploratory study of auditor evidential planning judgments. Auditing: A Journal of Practice & Theory 12 (2): 3961. ------, and ------. 1999. Are audit program plans risk-adjusted? Auditing: A Journal of Practice & Theory 18 (1): 5574. O'Keefe, T. B., D. A. Simunic, and M. T. Stein. 1994. The production of audit services: Evidence from a major public accounting firm. Journal of Accounting Research (2): 241261. St. Pierre, K., and J. A. Anderson. 1984. An analysis of the factors associated with lawsuits against public accountants. The Accounting Review 59 (April): 242263. Schaub, M. K. 1996. Trust and suspicion: The effects of situational and dispositional factors on auditors' trust of clients. Behavioral Research in Accounting: 154. U.S. House of Representatives, Committee on Financial Services. 2002. Sarbanes-Oxley Act of 2002. Public Law No. 107-204. Washington, D.C.: Government Printing Office. Wright, A., and R. H. Ashton. 1989. Identifying audit adjustments with attention-directing procedures. The Accounting Review 64 (October): 710728. Auditing: A Journal of Practice & Theory, November 2005 ...
View Full Document
This note was uploaded on 10/27/2008 for the course TE 17832 taught by Professor Will during the Spring '08 term at American College of Gastroenterology.
- Spring '08