From Offline to Online Audit in the Corona
Crisis: Workload and Job Complexity on
Auditor’s Professional Skepticism
Master Thesis, MSc Accountancy
University of Groningen, Faculty of Economics and Business
Marlieske van der Velde
S3862526
Supervisor: Dennis Veltrop Word Count: 8.126 Groningen, January 18, 2021
ABSTRACT
The excessive workload among auditors is a massive problem for the last decades. COVID-19 arguably reinforces this problem. The working environment and audits changed for all auditors. It is essential for the audit quality that the auditor exhibit a certain level of professional skepticism. I expect that excessive workload has a negative effect on the level of professional skepticism exhibited. The impact of the corona crisis is examined since the corona crisis can make the relationship more negative. Above, the effect of job complexity is introduced since auditors’ tasks are diverse.
Regression analyses were used to analyze survey data of 295 Dutch auditors. The results indicate that there is a positive association between workload and professional skepticism. Moreover, the corona crisis moderates this effect in such a way that the relationship is even more positive during the corona crisis. However, no support was found for the predicted moderating effect of job complexity.
Altogether, this study provides insight into how, why, and when workload impacts professional skepticism.
2
TABLE OF CONTENTS
I. Introduction ... 3
II. Theoretical Framework ... 6
2.1 Professional Skepticism ... 6
2.2 Professional skepticism and workload ... 7
2.3 The moderating role of COVID-19 ... 8
2.4 The moderating role of job complexity ... 9
III. Method ... 12 3.1 Data collection ... 12 3.2 Measures ... 12 3.2.1 Dependent variable ... 12 3.2.2 Independent variable ... 13 3.2.3 Moderating variables ... 13 3.2.4 Control variables ... 13 3.3 Data analysis... 14 IV. Results ... 15 4.1 Descriptive statistics ... 15 4.2 Hypothesis testing ... 16 4.3 Robustness check ... 18
V. Conclusion and Discussion ... 21
5.1 Findings ... 21
5.2 Theoretical & practical implications ... 22
5.3 Limitations & future research ... 23
References ... 25
3
I. INTRODUCTION
arch 2020, COVID-19 set foot in the Netherlands. As a result, the country went into an intelligent lockdown on the 12th of March. The corona crisis at the beginning of 2020
considerably changed a lot for schools and organizations. Rapidly, physical lectures shift to online lectures and working at the office shift to working from home. Employees from all different sectors must work from home, likewise auditors. Due to the coronavirus, audit teams cannot do audit activities at the client, it takes longer to get information from the client, and there is no face-to-face communication (NBA, 2020). As a result, auditors interact solely via technology; e-mail, electronic sharing through the internet, phone, Skype or, Zoom – rather than seeing their fellow auditors and clients face-to-face. Direct contact between colleagues is decreasing, and hence a leak of social support arises. Social support enhances work motivation and employees’ mental health (Lu, 1999). Due to the restrictions forced by COVID-19 – social distancing and isolation – people experience a high sense of loneliness. Social support is a significant factor in preventing depression and anxiety (Bergin & Pakenham, 2015) and its positive effect on employees’ motivation. Auditors’ well-being is vital in the way they perform their job.
Auditors generally experience high workload and stress (López & Peters, 2012). Before the corona crisis, the Dutch Committee on the future of the accountancy sector (Commissie Toekomst
Accountancysector, 2020) already came with a report in which they criticize auditors’ workload. A higher workload induces fatigue, which causes a lack of motivation and focus (Commissie Toekomst Accountancysector, 2020). They recommend a culture change “to do the right thing” with a focus on soft skills, workload, and work-life balance. The workload in the audit profession has been extensively examined. A high workload could cause less commitment, and eventually, burnout (Sweeney & Summers, 2002; Fogarty et al., 2000). An increased workload decreases the capacity to discover and report misstatements (López & Peters, 2012). Auditors are lesser skeptical when they face a high workload.
Professional skepticism can be defined as an auditor’s decisions and judgments to evaluate risks of material misstatements (Nelson, 2009). Previous researches, McDaniel (1990) and Nelson (2009), show a negative relationship between workload and professional skepticism. The need for efficiency is high with a higher workload, which affects an auditor's professional skepticism (Nelson, 2009). A more increased workload encourages auditors to ignore audit information, whether relevant or irrelevant. Due to the time pressure arising from the high workload, auditors are willing to accept weak client explanations (López & Peters, 2012). The number of hours auditors work, and the willingness to finish the audit are reasons, according to Hurtt et al. (2013), for accepting weak client explanations. During the corona crisis, all auditors worked from home. Workload dynamics can be
M
4
more demanding due to working from home and using Computer-Mediated Communication (CMC) as the primary working method.
The working way for auditors changed because of the corona crisis. Suddenly, one needs to work from home together with other family members, all working activities happen to be online, and there is less interaction between colleagues. These factors could lead to more distractions, focusing less on the job, and tasks are not fulfilled. In such times, the quality of self-control demands is rising. It takes auditors more effort to control their job. If one has self-control problems at home, one takes more or prolonged breaks or is busy with irrelevant tasks (Wehrt et al., 2020). Motivation is an essential matter for self-control. If one is motivated, they can perform enormous work. However, less motivation results in less completed work, which has a rising effect on an employee’s workload. Since all operational activities are performed online, the NBA (2020) mentions that it is crucial to stay professional skeptical, but that is not always achievable through online communication. Research by Bennet & Hatfield (2018) shows that auditors are less skeptical when Computer-Mediated Communication is being used. Auditors ask fewer follow-up questions, and there is less interaction than by face-to-face communication. COVID-19 reinforces the relationship between professional skepticism and workload. This leads to the following research question:
Is the negative impact of workload on auditor’s professional skepticism more severe during the corona crisis?
Although professional skepticism is expected to be negatively impacted by workload, especially during the corona crisis, some auditors may be more susceptible to pressure. In this vein, research indicates that individuals involved in complex tasks are generally more likely to suffer from the negative consequences of high workload. An important reason for this is that individuals’ cognitive resources are more likely to be constrained by the combination of high workload and complex tasks in these situations. Auditors perform a wide range of tasks while performing an audit, from simple routine tasks to very complex tasks where more professional skepticism is asked. According to the cognitive load theory, there is a limited working memory capacity (Plass et al., 2010). Complex tasks decrease an employee’s working memory capacity (Haji et al., 2015). With no working memory capacity left, employees find it hard to make decisions (Bolisani et al., 2018). Auditors with a reduced working memory capacity are more susceptible to the negative consequences of workload. Simon (1960) mentions that complex, unstructured tasks require more judgment than simple, structured tasks. According to Abdolmohammadi & Wright (1987), complex tasks require more professional
skepticism. Complex tasks are more time-consuming than simple tasks because it requires more knowledge and cognitive effort from the employee (Asare & McDaniel, 1996). Hurtt et al. (2013) mention that when an employee is not well skilled for tasks, this results in a decline in professional skepticism.
5
This paper contributes to the literature for several reasons. First, because professional skepticism is a fundamental foundation of auditing (Nolder & Kadous, 2018). Research on this topic is interesting for the audit profession. Second, the research focuses on the effects of a changing working environment due to COVID-19. Since COVID-19 was introduced at the end of 2019, there is little to no research on COVID-19’s impact on the audit profession. Third, the effect of the corona crisis is interesting for the prospects of the future. The economy is changing to a digital economy. Characteristics of the digital economy are no specific working location, critical roles of online platforms, and the use of big data (Valenduc & Vendramin, 2016). This industrial revolution will change the way of working. Flexible hours and flexible workplaces will be the new standard, and more online platforms will be used. Offices will not fulfill their original goal due to people choosing their workplace, e.g., at home. COVID-19 is accelerating the process towards a digital economy. If a higher workload and more complex tasks are consequences of the corona crisis, auditors’ skeptical behavior can be questioned. Moreover, the input for this research is high because all employees experienced working from home. Usually, working from home is voluntary, but now it is a restriction by the government. In previous research, there was no broad experience among auditors.
This paper is organized as follows. In section 2, literature related to the mentioned topics will be discussed, whereas hypotheses and the conceptual model will be presented. Followed-up, the obtained data and methodology are highlighted in section 3. The findings will be visualized in section 4. Last, there will be a discussion section about the findings and limitations of the paper.
6
II. THEORETICAL FRAMEWORK
2.1 Professional SkepticismA skeptical professional attitude is expected from an auditor. The International Standard of Auditing (2009) defines professional skepticism as: “An attitude that includes a questioning mind, being alert to conditions which may indicate possible misstatement due to error or fraud, and a critical
assessment of audit evidence.” Hurtt (2010) expresses professional skepticism through six characteristics: a) questioning mind, b) suspension of judgment, c) search for knowledge, d) interpersonal understanding, e) autonomy, and f) self-esteem. Nelson (2009) mentions that professional skepticism is the auditor's assessment of risk that an assertion is incorrect. There are multiple definitions of professional skepticism, and in line with these definitions, I define professional skepticism as the auditor’s ability and attitude to detect risks of material misstatements.
Professional skepticism is an essential characteristic of high-quality audits (Nolder & Kadous, 2018). According to Nolder & Kadous (2018), the skeptical professional attitude should recognize errors and misstatements, and an appropriate level of professional skepticism leads to a high-quality audit. Auditors who possess a high professional skeptical attitude are more likely to confront a client if irregularities arise (Knechel et al., 2013). Hurtt (2010) proposes that a skeptical professional attitude is a multi-dimensional individual characteristic. More skeptical judgments concern higher skepticism levels and lower trust levels (Hurtt et al., 2013). Factors like motivation and personality affect the level of professional skepticism exhibited. To each engagement, auditors bring different reasons and individual characteristics. Moreover, experience results in a greater understanding of the client; thus, senior personnel are better equipped to exhibit a skeptical professional attitude (Knechel et al., 2013; Hurtt et al., 2013; Nelson, 2009). Above all, multiple researchers (Hurtt et al., 2013; Nelson, 2009; Quadackers et al., 2014) found that professional skepticism can be trained. Smaller differences between high and low professional skepticism among auditors can be shrunk through fraud training.
Professional skepticism has been extensively examined over the last years. Two components of professional skepticism, according to Nelson (2009), are skeptical judgment and skeptical action. The skeptical judgment appears when an auditor identifies potential material misstatement and more work effort is essential. Skeptical action is the behavior of the auditor and is determined by the skeptical judgment. A distinction between trait- and state skepticism is made by Nelson (2009). Trait skepticism is the stable, enduring aspect of an individual, and state skepticism is a temporary condition awakened by situational factors (Hurtt, 2010). Based on the state of the auditor, the level of professional
skepticism varies for each engagement. State skepticism assumes that engagements with a higher risk of material misstatement desire auditors to exhibit a higher level of professional skepticism
7
(Robinson et al., 2018). According to Robinson et al. (2018) and Hurtt (2010), trait skepticism involves three characteristics: a) questioning mind (QM), b) search for knowledge (SK), and c) suspension of judgment (SJ). QM is the ability of ongoing questioning of information from the client to indicate risks of material misstatement. SK differs from QM because SK is about the general curiosity of interest. Skeptical auditors are interested in knowledge in general. SJ is that an auditor makes decisions based on an appropriate level of evidence. Situational factors that could influence state skepticism, according to Robinson et al. (2018), are incentives, time pressure, inconsistencies in client evidence, client incentives, and the client control environment (Quadackers et al., 2014). In this paper, I study professional skepticism as state skepticism following Hurtt (2010) and Robinson et al. (2018).
2.2 Professional skepticism and workload
The CTA (2020) came this year with a report in which they criticize the Dutch auditing sector's workload. In this report, the CTA advises a cultural change in the audit profession. Over the last two decades, there has been no change in auditors’ workload. Both Sweeney & Summers (2002) and Fogarty et al. (2000) state auditors experience a high workload during the busy season. In the busy season, most year-end audits take place. There are many audit engagements to complete with limited audit recourses (López & Peters, 2012). In the busy season, auditors experience a high workload and for the rest of the year. Accounting firms face a high turnover ratio and face staff shortages (Persellin, Schmid, et al., 2019). Additionally, auditors are working in different teams for different clients (Hurtt et al., 2013; Knechel et al., 2013). Moreover, work pressures are rising due to time budgets and time deadlines (Nelson, 2009). Accounting firms have become more aware of the workload problem, but no significant changes have been made.
A high workload is associated with negative consequences for the employee. High workload causes emotional exhaustion and presenteeism (Wirtz et al., 2017). Moreover, high workload results in sleep disturbance and fatigue (Huyghebaert et al., 2018). Auditors' commitment decreases due to an increased workload (Law, 2010; Huyghebaert et al., 2018). Sweeney & Summers (2002) found that auditors develop a depersonalized approach due to emotional exhaustion. Emotional exhaustion appears at the beginning when an employee faces burnout. Ten Bummelhuis et al. (2011) state that when employees face an increase in job demands and decreased job resources, an employee's
motivation also decreases. When an employee is in a bad mood, it makes irrelevant decisions (Forgas, 1989) and harms the judgment process (Albert & Steinberg, 2011). In Nelson’s professional
skepticism model, this link is recognized in the link between incentives and traits (Nelson, 2009). Motivation and personality differences impact auditors’ skeptical attitudes, according to Hurtt et al. (2013). This demotivating work attitude leads to a decrease in professional skepticism. Workload reduces auditors’ ability to detect risks of material misstatements (Suryandari & Yuesti, 2017).
8
Supposing an employee is underperforming, its motivation and incentives decrease (Mohd-Sanusi & Mohd-Iskandar, 2007). Many researchers (Libby & Lipe, 1992; Ashton, 1990; Awasthi & Pratt, 1990) found a positive relationship between incentives and judgment. If the employee is motivated, it has high financial and non-financial incentives, whereas auditors are better at making judgments.
Based on the theoretical background, a negative relation between workload and professional skepticism is assumed. This leads to the following hypothesis:
H1: Workload is negatively associated with professional skepticism
2.3 The moderating role of COVID-19
Since COVID-19, auditors are interacting solely via technology on the audit – there is no direct
interaction between colleagues themselves or the client. This reduced contact gives a decrease in social support. Social support reduces stress factors and strain (Lu, 1999; Cooper, 1981; Henderson & Argyle, 1985). For that reason, in high-stress jobs, more social support is needed. Stress is rising in these times, and there is a higher risk of losing your job. Ravn & Sterk (2017) found that job uncertainty is higher among employees in times of crisis. Job uncertainty leads to a higher degree of stress and strain and decreases job satisfaction (Paulsen, et al., 2005). Besides that, social support reduces stress, but it increases motivation (Lu, 1999). Motivation is essential for job efficiency and satisfaction. It can be hard to find motivation in these times. A lack of motivation results in self-control failure (Wehrt, Casper, & Sonnentag, 2020). Self-self-control failure ends up in employees performing irrelevant tasks, prolonged lunch breaks, less effort, and more distractions. In the end, a lack of motivation leads to an undesired behavior that harms the quality of work (Wehrt, Casper, & Sonnentag, 2020). The effects of COVID-19 harm the well-being of auditors.
COVID-19 made all auditors work from home. Whereas typically, audit teams visit the client during an engagement, meetings need to be done through phone calls, e-mails, Skype, or Google Teams. Meetings with the client shifted from offline to online. The difference between face-to-face communication and computer-mediated communication (CMC) has been studied before by several researchers (Bennet & Hatfield, 2018; Saiewitz, 2018; Mehra, 2010). Following the media richness theory, face-to-face communication is the richest in information (Daft & Lengel, 1986). The media richness theory is in line with the social presence theory by Short et al. (1976). This theory suggests that other communication methods than face-to-face have a lower social presence, which implies that online communication is not as effective as face-to-face communication. Because auditors are not capable of visiting the client, other ways of communication are employed. Since face-to-face
communication is impossible, it becomes harder for auditors to gain rich information. The workload is getting more severe in times of COVID-19 since it becomes harder for auditors to gain the right and richest information from the client.
9
Based on the arguments mentioned above, the corona crisis positively moderates the relationship of workload on professional skepticism. This leads to the following hypothesis:
H2: Corona crisis moderates the relationship between workload and skepticism, such that the relationship is more negative during the corona crisis
2.4 The moderating role of job complexity
Morgeson & Humphrey (2006) define job complexity as tasks that are hard and complex to perform. Work involving complex tasks requires more cognitive effort and skills than simple tasks (Morgeson & Humphrey, 2006; Asare & McDaniel, 1996). Complex tasks require more time than simple tasks because they need more skills and are mentally demanding and challenging (Asare & McDaniel, 1996). Auditors perform a wide range of tasks. Performing complex tasks requires more employee’s working memory. According to Plass et al. (2010), there is a limited capacity and duration of employee’s working memory. The working memory can be divided into different loads, which are visualized in Figure 1 below.
FIGURE 1
10
Intrinsic cognitive load is the amount of effort that is exhibited. Extraneous cognitive load is the way information is presented. The total cognitive load is based on element interactivity (Sweller, 1994). Element interactivity is the expertise to connect different types of information elements that are based on prior knowledge. A high element interactive goes along with a high cognitive load and reverse. Intrinsic and extraneous load together determine the total cognitive load. If the total cognitive load exceeds, the total working memory capacity cannot process information and learn. Intrinsic cognitive load is determined by task complexity (Haji et al., 2015), and extraneous cognitive load is determined by workload (Sweller, 1994). A combination of the high intrinsic and high extraneous cognitive load exceeds the cognitive resources, which results in failure (Sweller, 1994; Plass et al., 2010; Galy et al., 2012). Exceeding the cognitive resources is also known as cognitive overload. Bolsani et al. (2018) mention cognitive overload's consequences: employees cannot solve problems or make decisions. The decision-making process is influenced by the task's complexity and psychological factors (Bolisani et al., 2018).
A mental work overload occurs when the job demands exceed the worker’s capacity (Rubio-Valdehita et al., 2012). On the other hand, if the worker’s ability exceeding the job demand, there is a residual capacity that can be used for additional tasks, complex tasks. Multiple studies (Rubio-Valdehita et al., 2012; Wickens, 2008; Boles, 2007) show that the worker's mental workload increases as the task's complexity increases. The increasing mental workload harms the employee’s performance, especially by performing complex tasks (Rubio-Valdehita et al., 2012). According to the multiple resource theory, performance fails when an employee experiences work overload imposed by multiple, complex tasks (Wickens, 2008). Decreased performance results in a lower level of professional skepticism exhibited by auditors (Huey & Wickens, 1993; López & Peters, 2012). The workload is getting more severe by complex tasks since complex tasks are time-consuming and need cognitive effort, which harms the level of professional skepticism exhibited.
Mental workload involves two factors: job complexity and individual drivers (Rubio-Valdehita et al., 2012). Individual factors are stress, motivation, effort, and strain. Self-control failure is the result of the negative consequences of the corona crisis. Self-control failure is strongly related to the overall cognitive load, which implies that self-control fails due to a high cognitive load (Plass et al., 2010). High intrinsic load, caused by task complexity, results in self-control failure (Plass et al., 2010). Auditors already perceive self-control failure due to the negative consequences of the corona crisis. The combination of complex tasks and self-control failure exceeds employee’s cognitive resources, which harms the decision-making process, and the level of professional skepticism exhibited.
Based on the theoretical background, job complexity positively moderates the relationship of workload on professional skepticism. This leads to the following hypothesis:
11
H3a: Job complexity moderates the relationship between workload and skepticism, such that the relationship is more negative for auditors with complex tasks.
H3b: Job complexity strengthens the moderating effect of the corona crisis on the relationship between workload and skepticism, such that the relationship between workload and professional
skepticism is more negative during the corona crisis for auditors with complex tasks.
The hypotheses, as mentioned above, result in the following conceptual model illustrated in Figure 2.
FIGURE 2 Conceptual Model Workload Job Complexity Professional Skepticism Corona Crisis
12
III. METHOD
3.1 Data collectionData was collected via two surveys at different times from various audit firms in the Netherlands. The first survey was distributed in March 2020, and the second survey was distributed in November. Data is obtained together with other master students. In total, 295 auditors participated. Respondents were personally asked to participate in the survey. The sample consists of auditors with at least two years of working experience. This distinction made in years of experience is in line with previous research (Ohlsen, 2017; Quadackers, Groot, & Wright, 2014; Hurtt, 2010). The survey was administered in Dutch. By using the back and forward translation method, validity can be assured. The distribution of the survey has been done through mailings. To stimulate the respondents, respondents can receive a benchmark report and the survey results. Moreover, respondents are made aware of the fact that the results are anonymized. All questions derived their answers from a seven-point Likert scale. Preston & Colman (2000) found that using a seven-point scale leads to high reliability, internal consistency, and validity and is preferred by respondents. Moreover, Andrews (1984) mentions that a seven-point scale contributes to fewer measurement errors. The survey started with a cover letter informing the
participants about the subject, type of questions, and time needed to finish the survey. The invitations were sent personally, which increases the response rate (Dillman, 2007). Two reminders were sent in the following two weeks to minimize the non-response bias.
3.2 Measures
3.2.1 Dependent variable
Professional skepticism was measured using the scale of Robinson et al. (2018) based on the Hurtt scale (Hurtt, 2010). To measure professional skepticism, the focus, according to Robinson et al. (2018), should be on questioning mind (QM), suspension of judgment (SJ), and search for knowledge (SK). This distribution is in line with Hurtt (2010), who mentions that professional skepticism should include a questioning mind, requires the ability to ask questions to gather information and evidence, and the ability to judge if enough evidence is gathered. In total, there are twelve questions to measure professional skepticism: three questions regarding questioning mind, five regarding the suspension of judgment, and four regarding the search for knowledge. For answering, a Likert scale has been used from (1) strongly disagree to (7) strongly agree. In line with Robinson et al. (2018), the survey questions are modified to fit the audit sector. Auditors were asked to keep one specific client in mind for whom they performed audit activities in the last three months. One condition was that the
engagement is completed. This boundary condition is in line with Robinson et al. (2018), who recommends that the item contains a specific client employee. Auditors can answer the questions based on their experiences during that engagement. This results in that answers are more reliable
13
because no desired answers or behavior is given. See the Appendix for the survey questions. The Cronbach’s alpha is 0,81 and therefore reliable.
3.2.2 Independent variable
The scale Quantitative Workload Inventory (QWI) of Spector & Jex (1998) is used to measure the independent variable workload. The scale is developed to measure the speed and volume of the work. The scale consists of five items concerning the quantitative workload. In total, there are six statements in the survey. These statements are: how often a) does your job require you to work very hard; b) does your job require you to work very fast; c) does your job require you with little time to get things done; d) is there a great deal to be done; e) do you have to do more work than you can do well. On these statements, the auditor needs to indicate how often these items occur – from (1) less than once per month or never to (5) several times per day. A total end score represents a high or low workload. In the survey, auditors need to answer a 7-point scale, where (1) never, and (7) very often indicates. The Cronbach’s alpha is 0.78 and therefore reliable.
3.2.3 Moderating variables
COVID-19 is an exogenous variable. COVID-19 is measured by comparing the results of this survey with the survey results before the corona crisis. The survey upfront of the corona crisis was distributed in March. Even that this was the period COVID-19 set foot in the Netherlands, auditors were asked to reflect on previous audit engagements. COVID-19 will be measured using a dummy variable, where (0) is pre corona crisis and (1) during the corona crisis.
Job complexity is measured by the scale of Zacher & Frese (2011), which consists of four questions: 1) do you receive tasks that are extraordinary and particularly difficult; 2) do you often have to make very complicated decisions in your work; 3) can you use all your knowledge and skills in your work; 4) can you learn new things in your work. The items correspond to the definition of job complexity of Morgeson & Humphrey (2006). Instead of using the measurement from (1) very little to (5) very much as Zacher & Frese (2011) did, we use in the survey (1) a very small to (7) a very large extent. The Cronbach’s alpha is 0.72 and therefore reliable.
3.2.4 Control variables
Prior research has shown a consistent relation between experience and the level of professional skepticism exhibited (Abdolmohammadi & Wright, 1987; Nelson, 2009; Hurtt et al., 2013; Knecher et al., 2013). Senior staff has more knowledge of the client and its operating industry, whereas a better risk identification can be made, and individuals are better at making skeptical judgments. I control for experience by asking the auditors how long they work as an auditor.
14
3.3 Data analysis
Data analyses are used to measure the relationships and the extent between professional skepticism, workload, job complexity, and COVID-19. Multiple regressions and moderation analyses have been used to examine the data and to test for hypotheses. First, the dataset is optimized by checking all variables on outliers with univariate analysis. Outliers are values that are higher or lower than three standard deviations from the mean. These values have been revised. Second, an overview is displayed on the characteristics of the survey sample. Third, a reliability analysis in the sense of the Cronbach’s alpha was done to guarantee internal consistency. Before performing a multiple regression, a
15
IV. RESULTS
4.1 Descriptive statisticsThe percentages of frequencies of the twelve professional skepticism categories are provided in Table 1 below. The mean value of professional skepticism (Mean = 4.89) can be considered average, which means auditors possess a skeptical professional attitude. However, Table 1 shows that not all elements of professional skepticism are as essential. The answers to the question are either symmetrical
distributed (PS1, PS, PS3, PS7) or negative skewed (PS4, PS5, PS6, PS8, PS9-PS12). Moreover, Table 1 shows that auditors possess and use the skills to search for knowledge (PS9-PS12). The most prevalent element of professional skepticism is PS9; almost 80% think reading the client materials gives a better chance to arrive at correct assessments.
TABLE 1
Percentages of Frequencies, Mean and Standard Deviation of each Element of Professional Skepticism (N=295) PS1 PS2 PS3 PS4 PS5 PS6 PS7 PS8 PS9 PS10 PS11 PS12 1 “strongly disagree” 6.4 3.7 2.4 0.0 0 1.7 0 1.4 0 0.3 0.3 0.3 2 15.3 14.9 8.1 0.7 0.7 8.8 0.3 5.4 0.3 0.3 1.0 1.0 3 26.1 17.3 13.9 4.7 6.8 13.9 4.1 10.8 1.0 5.1 5.4 3.7 4 22.0 16.6 20.3 12.2 11.2 23.4 9.8 16.3 9.8 15.9 13.2 14.9 5 18.6 28.1 31.9 39.7 32.2 27.8 37.3 31.2 36.9 36.3 35.6 32.9 6 9.2 16.3 18.6 34.2 37.3 20.3 39.7 28.8 42.7 36.3 35.3 38.0 7 “strongly agree” 2.4 3.1 4.7 8.5 11.9 4.1 8.8 6.1 9.2 5.8 9.2 9.2 Mean 3.68 4.12 4.46 5.27 5.34 4.44 5.38 4.81 5.48 5.19 5.25 5.29 SD 1.46 1.52 1.40 0.99 1.09 1.38 0.95 1.34 0.86 1.00 1.07 1.05 Notes:
PS1: Questioning the statements from the client PS2: Questioning things that are seen or read PS3: Rejecting statements unless proof that it was true PS4: Take time to make decisions
PS5: Making decisions based on all reviewed, available information PS6: Making decisions quickly
PS7: Considering all information before making decisions PS8: Waiting on information to make decisions
PS9: Reading client materials for a better assessment PS10: Comparing different client’s documentation carefully PS11: Searching active for information
PS12: Using all resources available to get information
Table 2 provides the descriptive statistics and Pearson’s two-tailed correlations of all variables of interest. Outliers are replaced by the mean plus or minus three times the standard deviation. This limits the influence of outliers. For the variable experience, ten outliers are replaced. Above, missing values in this variable are replaced by the mean of the variable. A correlation and regression analysis show no substantial other results between using and not involving missing values; therefore, the mean is used to replace the missing values. Outliers are not replaced for other variables since there is no substantial difference between controlling and not controlling for outliers. Therefore, the outliers are not replaced.
16
The average experience of an auditor is 7.26 years. Due to anonymity, the experience variable is standardized, and consequently, no standard deviation is given. The mean of job complexity (Mean = 5.19) is high, which indicates that auditors experience their tasks as complicated and challenging. Above, it has a significant relation with professional skepticism. From the correlations in Table 2, it can be derived that the control variable experience is significantly correlated with professional skepticism and workload, which affirms controlling for this variable. Interesting is that experience is negatively related to workload, which would indicate that more experienced staff experiences a lower workload.
TABLE 2
Descriptive Statistics and Correlations
Variables M SD 1 2 3 4 5 1. Professional Skepticism 4.89 0.68 - 2. Workload 4.93 0.88 0.06 - 3. Job Complexity 5.19 0.90 0.28** 0.10 - 4. COVID-19 0.43 0.50 -0.06 -0.06 -0.09 - 5. Experience 7.26 n/a 0.13* -0.15** 0.07 -0.10 - N=295, *P<0.05, **P<0.01 (two-tailed) 4.2 Hypothesis testing
Multiple regression analyses on professional skepticism are performed to test for hypotheses. The results of this regression are presented in Table 3 below. The regression analysis also includes the variance inflation factors (VIF). VIF is an indicator of multicollinearity between variables. If the VIF values are considerably higher than 1, this indicates multicollinearity (Alin, 2010). None of the VIF values in the regression analyses significantly vary from one; therefore, there is no multicollinearity indication.
Model 1 is the standard model, which includes the control variable experience. Experience has a significant effect on professional skepticism (β=0.09, p<0.05). This relation is almost constant in all other models. This relation indicates that the more experience an auditor has, the higher the level of professional skepticism.
Hypothesis 1 predicted that workload is negatively associated with professional skepticism. The findings of model 2 show a significant positive association between workload and professional skepticism (β=0.06, p<0.10). This association implies that a higher workload increases the skeptical attitude of auditors. Hence, the results do not confirm hypothesis 1.
17
Hypothesis 2 stated that the corona crisis moderates the relationship between workload and
skepticism, such that the relationship is more negative during the corona crisis. The findings of model 6 show a significant positive interaction effect (β=0.06, p<0.05). The interaction effect is visualized in Figure 3. From this graph, it can be derived that auditors who experience a high workload exhibit a higher level of professional skepticism, and this relationship is stronger during the corona crisis.
In hypothesis 3a, it is assumed that job complexity moderates the relationship between workload and skepticism, such that the relationship is more negative for auditors with complex tasks. The results of model 5 demonstrate no effect (β=0.00, p>0.10).
Hypothesis 3b predicts that job complexity strengthens the corona crisis's moderating effect on the relationship between workload and skepticism. The results of model 8 show an adverse effect, but it is not significant (β=-0.01, P>0.10). The findings do not confirm hypothesis 3b.
TABLE 3
Regression Analysis on Professional Skepticism (N=295)
Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 Model 7 Model 8 Intercept 4.89*** (0.04) 4.58*** (0.23) 3.66*** (0.30) 3.65*** (0.30) 3.69*** (0.31) 3.91*** (0.33) 3.79*** (0.34) 3.80*** (0.35) Controls Experience 0.09** (0.04) 0.10** (0.05) 0.09** (0.04) 0.08** (0.04) 0.08** (0.04) 0.09** (0.04) 0.09** (0.04) 0.09** (0.04) Main effects Workload 0.06* (0.05) 0.04 (0.04) 0.04 (0.04) 0.04 (0.05) 0.01 (0.05) 0.01 (0.05) 0.01 (0.05) Job Complexity 0.20*** (0.04) 0,20*** (0.04) 0.20*** (0.04) 0.19*** (0.04) 0.21*** (0.05) 0.20*** (0.05) COVID-19 -0.03 (0.08) -0.03 (0.08) -0.03 (0.08) -0.03 (0.08) -0.03 (0.08) Two-way interaction
Workload x Job Complexity 0.00
(0.04) 0.00 (0.04) 0.00 (0.04) 0.00 (0.04) Workload x COVID-19 0.06** (0.03) 0.07** (0.03) 0.07** (0.03)
Job Complexity x COVID-19 -0.04
(0.03)
-0.04 (0.04)
Three-way interaction
Workload x Job Complexity x COVID-19 -0.01 (0.03) R2 0.02 0.02 0.09 0.09 0.09 0.10 0.11 0.11 Adjusted R2 0.01 0.02 0.08 0.08 0.08 0.08 0.08 0.08 F-value 4.70** 3.34** 9.59*** 7.21*** 5.75*** 5.43*** 4.85*** 4.24*** VIFⁱ 1.000 1.024 1.029 1.033 1.048 1.281 1.146 1.161 ***P<0.01, **P<0.05, *P<0.10
Two-tailed test for the control variables, one-tailed test for the hypothesized variables
18
FIGURE 3
The Relationship Between Workload and Professional Skepticism Before and After the Corona Crisis
4.3 Robustness check
In this section, two robustness checks are performed. The first robustness check is performed for hypothesis 1 to investigate a negative association between workload and professional skepticism using the three different dimensions of professional skepticism. In line with Robinson et al. (2018), the three dimensions questioning mind (QM), suspension of judgment (SJ), and search for knowledge (SK) are used. The Cronbach’s alphas for these dimensions are 0.87 (QM), 0.68 (SJ), and 0.80 (SK). Hence, therefore the different dimensions of professional skepticism are reliable. For each dimension, a regression analysis is examined with experience as the control variable and with workload as the main effect; see Table 4 below. All VIF values are around 1, which indicates no multicollinearity. For dimension QM, model 2 shows a significant positive association between workload and professional skepticism. For the other dimensions, SJ and SK, no significant associations are found. Noteworthy, all findings indicate a positive association between workload and the dimensions of professional skepticism. This finding is in line with the previous results but not in line with hypothesis 1. This additional analysis indicates that the questioning mind is the underlying driver for the positive relationship between professional skepticism and workload.
19
TABLE 4
Regression Analysis on Different Dimensions of Professional Skepticism
QM SJ SK
Model 1 Model 2 Model 1 Model 2 Model 1 Model 2 Intercept 4.09*** (0.08) 3.33*** (0.08) 5.05*** (0.04) 4.83*** (0.26) 5.30*** (0.05) 5.20*** (0.05) Controls Experience -0.00 (0.08) 0.02 (0.08) 0.15*** (0.05) 0.15*** (0.05) 0.10** (0.05) 0.11** (0.05) Main effect Workload 0.15** (0.09) 0.04 (0.05) 0.02 (0.05) R2 0.00 0.10 0.03 0.03 0.01 0.01 Adjusted R2 -0.003 0.004 0.03 0.03 0.01 0.01 F-value 0.002 1.54 9.08*** 4.90*** 4.50** 2.10 VIFⁱ 1.00 1.02 1.00 1.02 1.00 1.02 ***P<0.01, **P<0.05, *P<0.10
Two-tailed test for the control variables, one-tailed test for the hypothesized variables
ⁱ Mean of multiple VIF values
The second robustness check was performed to investigate if the central relationship between
professional skepticism and workload would change if another workload measurement would be used. Workload can be measured with different measurements. In the primary analysis, the workload is measured in line with Spector & Jex (1998). Another way to measure workload is the average amount of hours worked in a week. The same regression analysis is performed in the primary analysis, but instead of using workload, work hours are used as the independent variable. This regression analysis is presented in Table 5. All VIF values are around 1, which indicates no multicollinearity.
The findings of model 2 show a small positive association between work hours and professional skepticism, but it is not significant (β=0.01, p>0.10). However, in all other models, the effect is zero. A difference between the primary analysis is that the moderating effect of COVID-19 has a negative association (β=-0.02, p>0.10), see model 6, instead of a significant positive association. This association is visualized in Figure 4, which implies that auditors perform lower professional skepticism levels during the corona crisis, but no significant alterations on professional skepticism arise between many work hours or few work hours.
20
TABLE 5
Regression Analysis on Professional Skepticism by Work Hours
Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 Model 7 Model 8 Intercept 4.89*** (0.04) 4.51*** (0.41) 3.87*** (0.42) 3.92*** (0.44) 3.914*** (0.44) 3.82*** (0.49) 3.77*** (0.50) 3.77*** (0.50) Controls Experience 0.09** (0.04) 0.09* (0.04) 0.08* (0.04) 0.08* (0.04) 0.08* (0.04) 0.08* (0.04) 0.08* (0.04) 0.08* (0.04) Main effects Work Hours 0.01 (0.01) -0.00 (0.01) -0.00 (0.01) -0.00 (0.01) 0.00 (0.01) 0.00 (0.01) 0.00 (0.01) Job Complexity 0.21*** (0.04) 0.20*** (0.04) 0.20*** (0.04) 0.21*** (0.04) 0.22*** (0.04) 0.22*** (0.05) COVID-19 -0.04 (0.08) -0.04 (0.08) -0.04 (0.08) -0.04 (0.08) -0.04 (0.08) Two-way interaction
Work Hours x Job Complexity 0.01 (0.04) 0.00 (0.04) -0.00 (0.04) -0.00 (0.04)
Work Hours x COVID-19 -0.02
(0.03) -0.01 (0.04) -0.01 (0.04) Job Complexity x COVID-19 -0.02 (0.04) -0.02 (0.04) Three-way interaction
Work Hours x Job Complexity x COVID-19 -0.00 (0.03) R2 0.02 0.02 0.09 0.09 0.09 0.09 0.09 0.09 Adjusted R2 0.01 0.01 0.08 0.08 0.07 0.07 0.07 0.07 F-value 4.70** 2.77* 9.29*** 7.01*** 5.60*** 4.69*** 4.07*** 3.55*** VIFⁱ 1.00 1.05 1.07 1.07 1.07 1.15 1.22 1.29 ***P<0.01, **P<0.05, *P<0.10
Two-tailed test for the control variables, one-tailed test for the hypothesized variables ⁱ Mean of multiple VIF values
FIGURE 4
21
V. CONCLUSION AND DISCUSSION
5.1 FindingsA skeptical professional attitude is essential for a high-quality audit (Nolder & Kadous, 2018; Knechel et al., 2013; IAASB, 2009). However, auditors experiencing excessive workload (Fogarty et al., 2000; López & Peters, 2012; Persellin et al., 2019) are more emotionally exhausted and less motivated. These factors cause a lower level of professional skepticism exhibited by the auditor. Therefore, the purpose of this study was to examine if workload undermines professional skepticism and what moderating affect the corona crisis and job complexity plays in this respect. I predicted a negative relation between professional skepticism and workload that there would be a positive moderating effect of job complexity and the corona crisis, in such a way that both relationships are more harmful. Based on survey responses from 295 auditors, this study finds a direct positive relationship between professional skepticism and workload. The moderating effect of the corona crisis is confirmed.
First, I found a positive effect of workload on professional skepticism. The presence of workload among auditors increases the level of professional skepticism exhibited. This result is not supported by previous studies (López & Peters, 2012; Hurtt, 2010). Literature is mixed about the relationship between workload and performance. There is no clear evidence of this relation. According to Huey & Wickens (1993), an employee’s performance is most reliable under a moderate workload. This workload should not suddenly or unpredictably change. When the workload is not exceeding limits, it will not harm the performance. According to Brüggen (2015), employees need a certain level of stress and workload to perform at their best. The relationship between performance and workload is like an inverted U-curve (Brüggen, 2015). This relation indicates that when the workload increases,
performance increases until a particular turning. This inverted U-curve is personalized and differs per person. If performance does not harm workload, then it has no adverse effect on professional
skepticism. Workload affects motivation. If the employee complies with the workload, motivation will not be harmed by the workload. In line with previous literature, motivation positively affects
professional skepticism (Nelson, 2009; Forgas, 1989; Hurtt et al., 2013). As long as the employee is motivated in combination with a moderate workload, the auditor's professional skeptical attitude will not be harmed.
Second, I found a significant positive moderation effect of the corona crisis on the relationship between workload and professional skepticism. Since the relationship between workload and
professional skepticism is positive, the corona crisis moderates how this relationship is more positive in the corona crisis. Despite that workload is increasing during the corona crisis, auditors have become more aware of their skeptical professional attitude. The corona crisis is an exceptional situation, which increases the likelihood of risks. In high-risk contexts, the audit effort is increased by greater budget
22
hours and additional audit procedures (Quadackers, Groot, & Wright, 2014). An auditor should assess and identify the potential risks of an audit. Audits went from offline to online during the corona crisis. An example of considerable risk in the corona crisis is the Wirecard scandal. The previous year's Wirecard scandal showed that auditors cannot rely on screenshots from the client and need more advanced checks, physically or by screen sharing. Another example of a risk in this scandal was that Wirecard possibly used actors for video conferences. Wirecard employees could not be identified. Since more significant risks arise from the corona crisis, and additional controls at the client’s office are not possible, a more professional skeptical attitude is needed from the auditor. As the NBA (2020) also mentions in their report, it is vital to stay professional skeptical in these times.
Last, I found no significant moderation effect of job complexity on the relationship between workload and professional skepticism. Beyond, I found no significant three-way interaction between workload, job complexity, and the corona crisis on professional skepticism. Surprisingly, a significant, direct relation between job complexity and professional skepticism has been found. This relation suggests that auditors performing more complex tasks exhibit a higher level of professional skepticism, which is in line with the previous findings. Libby and Lipe (1992) mention that performance is not only affected by incentives, but also by task complexity. More effort is needed to understand the problem for complex tasks, resulting in productive search strategies to identify relevant information.
5.2 Theoretical & practical implications
This study provides several important implications for professional skepticism research and practice. This paper contributes to the academic, professional skepticism studies that attempt to understand the workload on professional skepticism since this has not directly been studied. This study shows a positive, significant relationship between workload and professional skepticism. The additional analysis indicates that the questioning mind is the underlying driver for the positive relationship between professional skepticism and workload. Workload has been examined a lot, but not the direct relation of workload on professional skepticism. Having a further look into the direct relationship between workload and professional skepticism could be interesting. Since workload has been extensively examined, more moderators and mediators can be controlled for in further research.
Workload is an important subject these days in the audit profession. Research shows that even when auditors experience excessive workload, they still have a skeptical professional attitude. The audit profession needs to have a moderate workload since it keeps auditors motivated, enhancing a skeptical professional attitude. As CTA (2020) mentioned, the audit profession needs to make a cultural change to mitigate the auditor’s excessive workload. Research shows that the corona crisis harms auditors’ workload. Nevertheless, in these uncertain times, auditors will get more skeptical and identify more risks. The corona crisis involved a new working way, and all audit activities were executed from home. By using CMC, auditors performed high-quality audits. In the future, this means that auditors
23
could work more from home, and unnecessary client visits can be withdrawn. A right balance between working from home and working at the office could enlighten the auditors' workload, which affects auditors' job satisfaction. Above, the change to a digital economy will be accelerated based on the knowledge that auditors can work from home using online audits, and the skeptical professional attitude will not be harmed.
5.3 Limitations & future research
As with any study, some limitations should be acknowledged. First of all, it has the usual limitations in survey research. Survey research can be used when a study regards attitudes and past actions.
However, biases arise when using survey research, and these biases need to be controlled. A critical bias that arises is conformity bias. The conformity bias is that people think in groups and answer what socially desirable is, instead of using their judgment (Padalia, 2014). Supposing respondents are given desirable answers; this questions the truthfulness of the outcome of professional skepticism. By carefully managing and designing the survey, the biases can be prevented. However, future research may extend this study with experimental research.
Second, the sample size only consists of Dutch auditors. This distribution makes it impossible to generalize the results to other countries and cultures. Excessive workload among auditors is not only a huge problem in the Netherlands (Commissie Toekomst Accountancysector, 2020) but worldwide (López & Peters, 2012; Sweeney & Summers, 2002). Excessive workload is also gaining interest by international boards like the Public Company Accounting Oversight Board (PCAOB) and the International Auditing and Assurance Standards Boards (IAASB), because it harms the audit quality and performance. Future research could expand this study by involving multiple countries since workload and professional skepticism are global concerns. Countries can be compared to the
relationship between excessive workload and professional skepticism. Moreover, the sample size only consists of auditors from the Big Four. For this reason, there can be no distinction made between Big Four and the non-Big Four. According to Persellin et al. (2019), non-Big Four auditors are more satisfied with their jobs, resulting in a higher level of professional skepticism and higher audit quality. Future research could involve both Big Four and non-Big four to enable a comparison between these two groups.
Finally, a limitation of this study is that the first survey was distributed during the busy season. During the busy season, auditors already perceive a high workload. The findings could be biased because respondents experienced an excessive workload when completing the survey. Despite perceiving an increased workload, respondents have little time to complete the survey; either they fill it in in a hurry or not at all. This excessive workload could also harm the non-response bias. I was not able to test for the non-response bias since the results of the respondents are anonymized. The second survey was sent in November, which is not during the busy season. Nevertheless, auditors were in the middle of the
24
corona restrictions, which increases the workload. During the second survey, auditors also perceived excessive workload due to the corona crisis. For the longitudinal study, the survey could be distributed again in April and November 2021 to better compare with the previous surveys better understand the coronavirus's impact.
25
REFERENCES
Abdolmohammadi, M., & Wright, A. (1987). An Examination of the Effects of Experience and Task Complexity on Audit Judgments. The Accounting Review, 62(1), 1-13.
Albert, D., & Steinberg, L. (2011). Judgment and Decision Making in Adolescence. Journal of Research on Adolescence, 21(1), 211-224.
Alin, A. (2010). Multicollinearity. Wiley Interdisciplinary Reviews: Computational Statistics, 2(3), 370-374.
Andrews, F. M. (1984). Construct Validity and Error Components of Survey Measures: A Structural Modeling Approach. Public Opinion Quarterly, 48(2), 409-442.
Asare, S. K., & McDaniel, L. S. (1996). The Effects of Familiarity with the Preparer and Task Complexity on the Effectiveness. The Accounting Review, 71(2), 139-159.
Ashton, R. (1990). Pressure and performance in accounting decision settings: paradoxical effects of incentives, feedback and justification. Journal of Accounting Research, 28, 148-180.
Awasthi, V., & Pratt, J. (1990). The effects of monetary incentives on effort and decision
performance: the role of cognitive characteristics. The Accounting Review, 65(4), 797-811.
Bennet, G. B., & Hatfield, R. C. (2018). Staff Auditors' Proclivity for Computer-Mediated
Communication with Clients and Its Effect on Skeptical Behavior. Accounting, Organizations and Society, 68-69, 42-57.
Bergin, A., & Pakenham, K. (2015). Law Student Stress: Relationships Between Academic Demands, Social Isolation, Career Pressure, Study/Life Imbalance and Adjustment Outcomes in Law Students. Psychiatry, Psychology, and Law, 22(3), 388-406.
Boles, D. (2007). Predicting dual-task performance with the multiple resources questionnaire (MRQ). Human Factors, 49(1), 32-45.
Bolisani, E., Scarso, E., & Padova, A. (2018). Cognitive overload in organizational knowledge management: Casy study research. Knowledge and Process Management, 25(4), 223-231.
Brüggen, A. (2015). An empirical investigation of the relationship between workload and performance. Management Decision, 53(10), 2377-2389.
Commissie Toekomst Accountancysector. (2020, Januari 30). Vertrouwen op Controle Eindrapport van de Commissie toekomst accountancysector. Retrieved from Rijksoverheid:
26
https://www.rijksoverheid.nl/documenten/kamerstukken/2020/01/30/vertrouwen-op-controle-eindrapport-van-de-commissie-toekomst-accountancysector
Cooper, C. L. (1981). Social Support at Work and Stress Management. Small Group Behavior, 12(3), 285-297.
Daft, R. L., & Lengel, R. H. (1986). Organizational Information Requirements, Media Richness and Structural Design. Management Science, 5(32), 554-571.
Dillman, D. (2007). Mail and internet surveys: The tailored design. New York: Wiley.
Fogarty, T., Singh, J., Rhoads, G., & Moore, R. (2000). Antecedents and consequences of burnout in accounting; Beyond the role stress model. Behavioral Research in Accounting, 12, 31-67.
Forgas, J. P. (1989). Mood effects on decision making process. Australian Journal of Psychology, 41(2), 197-214.
Galy, E., Cariou, M., & Mélan, C. (2012). What is the relationship between mental workload factors and cognitive load types? International Journal of Psychology, 83, 269-275.
Haji, F. A., Rojas, D., Childs, R., de Ribaupierre, S., & Dubrowski, A. (2015). Measuring cognitive load: performance, mental effort and simulation task complexity. Medical Education, 49, 815-827.
Henderson, M., & Argyle, M. (1985). Social support by four categories of work colleagues:
Relationships between activities, stress and satisfaction. Journal of Occupational Behaviour, 6, 229-239.
Huey, B. M., & Wickens, C. D. (1993). Workload Transition: Implications for Individuals and Team Performance . Washington DC, US: National Academy Press .
Hurtt, R. K. (2010). Development of a Scale to Measure Professional Skepticism. Auditing: A Journal of Practice & Theory, 29(1), 149-171.
Hurtt, R. K., Brown-Liburd, H., Earley, C. E., & Krishnamoorthy, G. (2013). Research on Auditor Professional Skepticism: Literature Synthesis and Opportunities for Future Research. Auditing: A Journal of Practice & Theory, 32(1), 45-97.
Huyghebaert, T., Gillet, N., Beltou, N., Tellier, F., & Fouquereau, E. (2018). Effects of workload on teachers' functioning: A moderated mediation model including sleeping problems and overcommitment. Stress and Health, 34, 601-611.
27
International Auditing and Assurance Standards Boards. (2009). ISA 220 Overall objectives of the independent auditor and the conduct of an audit in accordance with international standards on auditing. New York.
Knechel, W. R., Krishnan, G. V., Pevzner, M., Shefchik, L. B., & Velury, U. K. (2013). Audit Quality: Insights from the Academic Literature. Auditing: A Journal of Practice & Theory, 32(1), 385-421.
Law, P. (2010). Examination of the actual turnover decisions of female auditors in public accounting: Evidence from Hong Kong. Managerial Auditing Journal, 25(5), 484-502.
Libby, R., & Lipe, M. G. (1992). Incentives, Effort, and the Cognitive Processes Involved in Accounting-Related Judgments. Journal of Accounting Research, 30(2), 249-273.
López, D. M., & Peters, G. F. (2012). The Effect of Workload Compression on Audit Quality. Auditing: A Journal of Practice & Theory, 31(4), 139-165.
Lu, L. (1999). Work motivation, job stress and employees' well-being. Journal of Applied Management Studies, 8(1), 61-72.
McDaniel, L. S. (1990). The effects of time pressure and audit program structure on audit performance. Journal of Accounting Research, 28(2), 267-285.
Mehra, P. (2010). Impact of Task Types and CMC Technology on Exchange Quality: An Empirical Investigation. Vikalpa, 35(1), 31-52.
Mohd-Sanusi, Z., & Mohd-Iskandar, T. (2007). Audit judgment performance: assessing the effect of performance incentives, effort and task complexity. Managerial Auditing Journal, 22(1), 34-52.
Morgeson, F. P., & Humphrey, S. E. (2006). The Work Design Questionnaire (WDQ): Developing and Validating a Comprehensive Measure for Assessing Job Design and the Nature of Work. Journal of Applied Psychology, 9(6), 1321-1339.
NBA. (2020, April 6). NBA Alert - 42: Impact Coronavirus on Professional Services Provided by Professional Accountants. Retrieved from NBA: https://www.nba.nl/globalassets/wet--en- regelgeving/nba-alerts/alert-42/english-version-nba-alert-42-translation-of-the-dutch-alert-impact-coronavirus-on-accountants-procedures-april-6-2020.pdf
Nelson, M. W. (2009). A Model and Literature Review of Professional Skepticism in Auditing. Auditing: A Journal of Practice & Theory, 28(2), 1-34.
28
Nolder, C. J., & Kadous, K. (2018). Grounding the professional skepticism construct in mindset and attitude theory: A way forward. Accounting, Organization and Society, 67, 1-14.
Ohlsen, C. (2017). A Study of Professional Skepticism. Bergen, Norway: Springer.
Padalia, D. (2014). Conformity Bias: A Fact or an Experimental Artifact? Psychological Studies, 59(3), 223-230.
Paulsen, N., Callan, V., Grice, T., Rooney, D., Gallois, C., Jones, E., . . . Bordia, P. (2005). Job uncertainty and personal control during downsizing: A comparison of survivors and victims. Human Relations, 58(4), 463-496.
Persellin, J. S., Schmidt, J. J., Vandervelde, S. D., & Wilkins, M. S. (2019). Auditor Perceptions of Audit Workloads, Audit Quality, and. Journal of Accounting Horizons, 33(4), 95-117.
Plass, J. L., Moreno, R., & Brünken, R. (2010). Cognitive Load Theory. Cambridge: Cambridge University Press. doi:10.1017/CBO9780511844744
Preston, C., & Colman, A. (2000). Optimal number of response categories in rating scales: reliability, validity, discriminating power, and respondent preferences. Acta Psychologica, 104(1), 1-15.
Quadackers, L., Groot, T., & Wright, A. (2014). Auditor's Professional Skepticism: Neutrality versus Presumptive Doubt. Contemporary Accounting Research, 31(3), 639-657.
Ravn, M. E., & Sterk, V. (2017). Job uncertainty and deep recessions. Journal of Monetary Economics, 90, 125-141.
Robinson, S. N., Curtis, M. B., & Robertson, J. C. (2018). 'Disentangling the trait and state components of professional skepticism: Specifying a process for state scale development'. Auditing: A Journal of Practice & Theory, 37(1), 215-235.
Rubio-Valdehita, S., Díaz-Ramiro, E. M., López-Higes, R., & Martín-García, J. (2012). Effects of task load and cognitive abilities on performance and subjective mental workload in a tracking task. Annals of Psychology, 28(3), 986-995.
Saiewitz, A. (2018). Email versus In-Person Audit Inquiry:. Current Issues in Auditing, 12(2), A36-A44.
Sandrin, É., Gillet, N., Fernet, C., Leloup, M., & Depin-Rouault, C. (2019). Effects of motivation and workload on firefighters' perceived health, stress and performance. Stress and Health, 35, 447-456.
29
Short, J., Williams, E., & Christie, B. (1976). The Social Psychology of Telecommunications. London: John Wiley & Sons.
Simon, H. A. (1960). The New Science of Management Decision. New York, United States : Harper & Brothers.
Spector, P., & Jex, S. (1998). Development of four self-report measures of job stressors and strain: Interpersonal conflict at work scale, organizational constraints scale, quantitative workload inventory, and physical symptoms inventory. Journal of Occupational Health Psychology, 3(4), 356-367.
Suryandari, N. N., & Yuesti, A. (2017). Professional Skepticism and Auditors Ability to Detect Fraud Based on Workload and Characteristics of Auditors. Scientific Research Journal, 5(9), 109-115.
Sweeney, J., & Summers, S. (2002). The effect of the busy season workload on public accounts' job burnout. Behavioral Research in Accounting, 14(1), 223-245.
Sweller, J. (1994). Cognitive load theory, learning difficulty, and instructional design. Learning and Instruction, 4(4), 295-312.
Ten Bummelhuis, L. L., Ter Hoeven, C. L., & Bakker, A. B. (2011). Breaking through the loss cycle of burnout: The role of motivation. Journal of Occupational and Organizational Psychology, 84, 268-287.
Valenduc, G., & Vendramin, P. (2016). Work in the digital economy: sorting the old from the new. Working paper. Retrieved from http://ftu-namur.org/fichiers/Work_in_the_digital_economy-ETUI2016-3-EN.pdf
Wehrt, W., Casper, A., & Sonnentag, S. (2020). Beyond depletion: Daily self-control motivation as an explanation of self-control failure at work. Journal of Organizational Behavior , 1-17.
Wickens, C. D. (2008). Multiple resources and mental workload. Human Factors, 50(3), 449-455.
Wirtz, N., Rigotti, T., Otto, K., & Loeb, C. (2017). What about the leader? Crossover of emotional exhaustion and work engagement from followers to leaders. Journal of Occupational Health Psychology, 22, 86-97.
Zacher, H., & Frese, M. (2011). Maintaining a focus on opportunities at work: The interplay between age, job complexity, and the use of selection, optimization, and compensation strategies. Journal of Organizational Behavior, 32, 291-318.
30
APPENDIX
Survey Questions
Professional Skepticism
Before answering the questions to professional skepticism, the auditor needed to take one client in mind by answering the following questions.
Letter/initials of the client you have in mind: ____________ - How long do you work for this client?
- How many hours have you worked for this client?
After they had taken a client in mind, the following questions were asked:
1. Overall, I tended to question the statement that I read from [name client].
2. While working on this engagement, I frequently questions things that I saw or read. 3. While working on this engagement, I had a tendency to reject statements unless I had proof
that they were true.
4. While working on this engagement, I took my time when making decisions.
5. During this engagement, I did not like deciding until I had a chance to look at all of the available information.
6. I did not like having to make decisions quickly while working on this engagement.
7. While working on this engagement, I tried to ensure that I had considered the most available information before making a decision.
8. While completing this engagement, I waited to make decisions until I could get more information.
9. I felt that reading client materials would give me a better chance to arrive at correct assessments.
10. I tended to read closely and compare between different sections of client documentation in order to improve my chances of making a correct statement.
11. I actively sought out all of the information that I could while completing this engagement. 12. I used all resources available to me to get all of the information that I could, in this
engagement.
Note questions 1-3 about QM, questions 4-8 about SJ, questions 9-12 about SK.
Workload
1. How often does your job require you to work very fast? 2. How often does your job require you to work very hard?
3. How often does your job leave you with little time to get things done? 4. How often is there a great deal to be done?
5. How often do you have to do more work than you can do well?
Work Hours
During the last period, how many hours did you work in an average week?
Experience
