How Do Auditors Order Their Tasks, and How Does Task Ordering Affect - - PowerPoint PPT Presentation

How Do Auditors Order Their Tasks, and How Does Task Ordering Affect Performance?

Robert P. Mocadlo College of Business & Public Administration University of North Dakota

Background

• Auditors need to meet deadlines and budgets
• Archival evidence shows that reporting deadlines affect audit quality

– López and Peters 2012, Bryant-Kutcher et al. 2013, Lambert et al. 2015, Glover et al. 2015

• How can auditors reduce time/effort while still completing all of their tasks?

– Some audit tasks have more flexibility than others, so auditors could use this flexibility to ensure that all their tasks are completed

2

Performance Performance range

The size of the performance range depends on the subjectivity of the task’s criteria or guidance

Task Performance

Perfunctory performance Consummate performance Incomplete Inefficient

Objective Tasks

• Performance range: narrow (or zero)
• Low ability to decrease time/effort

– Moving from the upper threshold to the lower threshold saves only a little (or no) time/effort

4

Performance range Performance

Subjective Tasks

• Performance range: wide
• High ability to decrease time/effort

– Moving from the upper threshold to the lower threshold can save a lot of time/effort

5

Performance range Performance

Hypotheses 1 and 2

When auditors are assigned both an objective and subjective task: H1: They are more likely to choose to perform the objective task first regardless of time pressure H2: Performance will decrease more with time pressure for the subjective versus the objective task

6

Mitigating Effects of Criteria Subjectivity

• What would cause auditors to prioritize subjective tasks?

– Subjective accounts (more likely to require subjective audit tasks) tend to be riskier, which may lead to a higher assessed risk for these accounts in the planning phase of the audit – Increased importance of higher-risk tasks – Less likelihood of negative consequences from incomplete work on lower-risk tasks

7

Hypotheses 3 and 4

Regardless of time pressure, when auditors are assigned both an objective and subjective task: H3: Auditors will be more likely to perform the subjective task first if the assessed inherent risk for the account covered by the subjective task is high versus low H4: Performance will be higher on the subjective task relative to the objective task if the assessed inherent risk for the account covered by the subjective task is high versus low

8

Experimental Procedure

9

• 113 auditor participants were given two tasks to complete

– Objective task: matching sales invoices to supporting documentation – Subjective task: forming an expectation for a substantive analytical procedure related to workers’ compensation (adapted from Clor-Proell & Maines 2014)

• Both tasks needed to be completed within an overall time limit
• Auditors chose which task to perform first, worked on that task, then spent any

remaining time on the second task

Independent Variables

• Task Type (within-participants)

– Objective and subjective

• Time Pressure (between-participants)

– In both conditions, participants were told that each task should take 10 minutes (20 minutes total) – Low: Participants had 25 minutes to complete both tasks – High: Participants had 15 minutes to complete both tasks

• Subjective Task Risk (nested, between-participants)

– Low vs. High (provided in the background) – Objective task risk is always low

10

Dependent Variables

• First Task Chosen (objective or subjective)
• Performance

– Objective task: number of seeded errors caught (out of three) – Subjective task: level of disaggregation of data used to form expectation (three possible levels)

11

Results for H1 and H3 (First Task Chosen)

• 67 of 113 participants (59%) performed the objective task first

– Binomial probability test: p = 0.03

10/17/2016 12

Logistic Regression for First Task Chosen df χ2 Two-tailed p-value Constant 1 0.98 0.323 Time Pressure 1 3.64 0.056 Task Presentation Order 1 1.93 0.165 Time Pressure × Task Presentation Order 1 0.76 3.850 Subjective Task Risk 1 6.27 0.012 Time Pressure × Subjective Task Risk 1 9.91 0.002 Task Presentation Order × Subjective Task Risk 1 0.71 0.399 Time Pressure × Task Presentation Order × Subjective Task Risk 1 2.94 0.086

Results for H1 and H3 (First Task Chosen)

• Under High Time Pressure

– First Task Chosen is dependent on Subjective Task Risk (Fisher’s exact test = 0.021) – When Subjective Task Risk is low (e.g. the inherent risk is the same across both tasks):

• 22 of 30 participants (73%) performed the objective task first
• Supports H1

– When Subjective Task Risk is high (e.g. the inherent risk is higher for the subjective task):

• 14 of 33 participants (42%) performed the objective task first
• Supports H3

13

Results for H1 and H3 (First Task Chosen)

• Under Low Time Pressure

– No significant relationship between First Task Chosen and Subjective Task Risk (p = 0.150)

• Therefore, H1 and H3 are conditionally supported (when Time Pressure is High)

10/17/2016 14

Results for H2 and H4 (Performance)

15

.5 1 1.5 2 2.5 3 Performance

Objective Task Performed First Subjective Task Performed First

Low TP High TP Low TP High TP Objective Task Subjective Task

H2: Task Type × Time Pressure interaction when Objective Task performed first (p = 0.002) H2: Task Type × Time Pressure interaction when Subjective Task performed first (p = 0.138) H4: No effect of Subjective Task Risk (smallest p = 0.389)

Supplemental Analyses

• Alternative specifications for robustness
• Reduced observations based on manipulation check failures
• Time spent on tasks
• Likelihood of participants being able to complete both tasks
• Effect of task difficulty

10/17/2016 16

Summary and Contribution

17

• Auditors’ task ordering matters!
• If subjective tasks are necessary to audit subjective accounts, auditors may be taking
• n more risk than they realize
• Standard-setters, regulators, and audit firms should consider how the objective or

subjective language in their guidance may interact with auditors’ incentives

Thank you!

18