A STUDY ON AUDIT JUDGMENT PERFORMANCE: THE EFFECT OF

ACCOUNTABILITY, EFFORT AND TASK COMPLEXITY

Ria Nelly Sari1, Diani Mardisar and Rita Anugerah*

*Department of Accounting, University of Riau

Kampud Bina Widya KM 12.5 Simpang Baru, Panam, Pekanbaru – Indonesia
ria_n_sari@yahoo.com

ABSTRACT

This study examines the effect of effort on the relationship between accountability
and audit judgment performance under different levels of task complexity. Data for this study
was obtained from samples of auditors who work in local Accounting Public Firms in
Pekanbaru (Riau Province) and Padang (West Sumatera)- Indonesia. A total of 62 responses
(44,29%) obtained and used in the analysis. Data was then analized using Three Steps of
Mulptiple Regression Analysis recommended by Bonner & Kenny (1986). Result of the
study indicates that the effect of accountability on audit judgment performance is mediated
by effort. In addition, when task complexity is low, auditor with high effort show better
performance than those with low effort. But, when task complexity is high, there are no
differences in audit judgment performance between hight effort and low effort groups.

Keywords: Accountability, Effort, Task Complexity, Audit Judgment Performance, Three

Steps of Mulptiple Regression Analysis

INTRODUCTION

In auditing, the quality of auditor’s work is crucial. Normally the external party will

draw a conclusion and make decisions regarding on the company performances based on the

auditor’s report. While for auditors, their coclusions and judgment will depend on results of

the auditing process they have done. That means, the result of auditor’s work will effect on

auditor’s conclusion and indirectly effects the evaluation or decision made by external party.

In completing audit tasks auditor are always facing difficult challenges (Snead and

Harrel, 1991). Some of the challenges arise from the nature of auditor’s job it self such as

work pressures, inadequate resources or manpower and task uncertainties. All of these

1 Contact author: Ria Nelly Sari, tel 62 815 3751 7072, email address: ria_n_sari@yahoo.com
challenges will lead to lack of consensus among auditors and inaccuracy in audit judgment,

which in turn affects the quality of audit judgment (Trotman,1998).

Some prior studies had found that one of the solutions to increase the quality of audit

judgment performance is by giving performance incentive to auditors either financial or non

financial incentives. The effects of financial/monetary incentive on audit judgment

performance had been studied by Aswathi and Pratt (1990), Baeley et.al (1998) Bonner and

Spilker (2002) and Sanusi and Iskandar 2007). Whereas, non financial incentive like;

justifications (Chang et.al 1997), feedback (Ashton 1990; Sanusi and Iskandar 2007) and

accountability (Mardisar and Sari 2007; Tan and Kao 1999; Cloyd 1997; Kennedy 1993;

Meisier and Quilliam 1992; Chaikan 1980 and Tetclock and Kim 1987) also had significant

effect on audit judgment performance.

In general, the studies above proofed that both financial and nonfinancial incentive

affected the quality of audit judgment performance, but according to Bonner & Spilker

(2002) the studies which analyze the effect of effort as mediation in relationship between

performance incentive and audit judgment performance are still few.

Bonner and Spilker (2002) assembled the performance incentive as a kind of

enviroment variable that could affect individual effort and audit judgment performance. They

also stated that the effect of performance incentive on quality of auditor’s work

(performance) differred depending on the types of incentive given and effort devoted.

Efforts are classified into two, one is a type of effort which lead into the increase of

auditor work’s performance (audit judgement performance) and the second is a type of effort

which lead into the increase of auditor’s learning. The type of effort which lead into an

increase of auditor work’s performance are based on three components namely; effort

direction, effort duration and effort intensity. Whereas, the type of effort which lead toward

the increase in auditor’s learning is known as strategy development. The four types of effort

2
above (effort direction, effort duration, effort intensity and strategy development) has an

interaction with task complexity and give different effects on performances (Bonner and

Spilker 2002).

Accountability is one of the performance incentives that could affect the quality of

audit judgment performance. Accountability has a social-physcology link urging a person to

complete his/her duty and justify to his/her social enviroment. An accountable person will try

to keep a good social image (Meisier and Quiiliam 1992).

Cloyd (1997) found that accountability could increase auditor’s effort (both duration

and intensity). If effort devotion is low, then only those auditors who have high audit

knowledge could increase their audit judgment performance. But the study yet to find

whether task complexity could affect the quality of auditor’s work.

Cloyd’s study was extended by Tan and Kao (1999) and Tan et. Al (2002) by

analyzing the role of task complexity, knowledge and problem solving ability predicted as

variables that also affect quality of auditor’s work. They found that for low complexity task,

accountability will not improve performance; for medium complexity task, higher

accountability subject will result in better performance when auditor have high knowledge;

and for high complexity task, higher accountability will result in better performance when

auditors have both high knowledge and high problem solving ability. But, this study had not

yet analyzed the possibility effect of effort as mediator in relationship between accountability

and quality of auditor’s work.

This study meant as a further research on similar topic previously done by Mardisar

and Sari (2007) which evaluated the effect of accountability on quality of auditor’s work

(audit judgment performance). Later this study will proof whether effort can be mediation

between accountability and audit judgment performance and how it affects the audit

judgment performance when auditors are faced with different level of task complexity (low

3
and high).

The remainder of the paper is organized as follows. The next section develops the

study’s hypotheses related to the effects of accountability on audit judgment performance, the

effects of accountability on effort, the mediation role of effort and the moderator role of task

complexity. The third section details the experimental design and research method. The

fourth section presents the study’s implications and limitations.

LITERATURE REVIEW AND HYPOTHESIS DEVELOPMENT

Audit Judgment Performance

Performance effectiveness refers to how well a particular task is performed in the

relation to a criterion. For auditors, quality of work is appraised by looking at accurate

response given by auditor for every audit task.

Bonner and Spilker (2002) stated that there are three variables that can affect

performance, namely: person variable, task variable and enviroment variable. Person variable

include the attributes that a person posesses prior to perform a task such as knowledge

content, knowledge organization, abilities, confidence, cognitive style, instrisic motivation,

cultural values. Task variable include factors that vary both within and across task, such as

complexity, presentation format, processing mode and response mode. While, enviroment

variable include all conditions, circumtances and influences surrounding a person who is

performing a particular task, such as time pressure, accountability, assigned goal and

feedback.

The quality of auditor’s work can be seen from the quality of judgment and decision

made. According to Edward et.el (1984) in Bedard and Chi (1993), there are two criteria used

to evaluate a decision, namely outcome oriented and process oriented.

4
 Outcome oriented criteria is used when the accuracy of an outcome can be

determined. To evaluate the quality of a chosen decision is done by comparing solution and

criteria of standard outcome. Unlike the outcome oriented criteria, process oriented criteria

are used when the accuracy of an outcome can not be determined. So, to evaluate the quality

of auditor’s decision is to view it from the quality of audit process done by auditor during the

audit work from the beginning to the decision made.

Effect of Accountability on Audit Judgment Performance

According to Tetclock (1985) in Messier and Quilliam (1992) accountability has a

critical role and norm enforcement mechanism: the social psychology link between individual

decision maker on one hand and the social system to which they belong on the other. The fact

that people are accountable for their decision is an implicit or explicit constraint upon all

consequential acts they undertake.

Tan and Kao (1999) stated about three indicators that can be used to evaluate

individual accountability: how motivated they were to perform well on the task; how their

confidence level of their responses would be reviewed by training manager and how amount

of mental effort used. High accountability subjects were more motivated, more likely to think

that they would be reviewed and exert more effort than low accountability subjects.

Meisier and Quilliam (1999) studied the effect of accountability on the individual

cognitive process in working and found that high accountability subjects do more

complicated cognitive process. Similar with Meisier and Quilliam (1999), Tetclock and Kim

(1987) also found that subjects who were informed at the beginning that their task will be

reviewed by training (pre exposure accountability subjects) did more complex cognitive

process, gave more accurate responses and reported more realistic confidence level in their

decisions.

Buchman et.al (1989) in Meisier and Quilliam (1992) found different result that
5
accountability affected performance depends on kind of individual accountability. On their

study, Buchman et.al (1989) classified individual accountability into three categories:

accountable to client, accountable to partner and accountable to no one. They found that

auditor who was accountable to client tended to give unqualified opinion and auditor was

accountable to partner and accountable to no one tended to give qualified opinion. Based on

that study accountability will affect performance depends on the evaluative audience that the

auditors responsive to.

Based on the past study above, we try to reexamine the effect of accountability on the

quality of auditor’s work by using the following hypothesis:

H1: Auditors who have high accountability perform better audit judgment than low

accountability auditors.

Effect of Accountability on Effort

According to Bonner & Spilker (2002), amount of cognitive effort given by subject

to complete a task could be seen from: effort direction, effort duration, effort intensity and

strategy development. Effort direction refers to the task or activity in which the individual

chooses to engage. In laboratory experiment, subject who has high effort direction treated by

only giving one task and they asked to be more focused to the task. Effort duration refers to

the length of time an individual devotes to cognitive and physical resources to particular task

or activity (how long a person works). Effort intensity refers to amount of attention an

individual devotes to a task or activity during a fixed period of time (how hard a person

works). Whereas, effort refers to strategy can be seen from planning, innovation and

conscious problem solving made.

Sanusi and Iskandar (2007) found that performance incentives both monetary

incentive and non monetary incentive (feedback) could increase amount of effort devoted to

complete audit tasks and mediates relationship between performance incentive and quality of
6
auditor’s work. Similar results also found from Baron and Spilker (2002) who studied the

effect of monetary incentive on performance and also found that effort could create an

indirect relationship between monetary incentive-performance.

Accountability is one of a non financial incentive. Cloyd (1997) found that

accountability can increase effort devoted either effort intensity or effort duration. Based on

discussion above, we formally express the hypothesis:

H2: Auditors with higher accountability will devote higher effort than auditor with

lower accountability.

Task Complexity as a Mediator Variable

One of the factors that could influence the audit judgment performance is task

complexity. According to Libby and Lipe (1992) and Kennedy (1993, for a certain level, task

complexity can influence auditor’s effort. Given such effort, auditors can increase their

quality of work. Sanusi and Iskandar (2007) stated that the effect of effort on auditor’s

performance cannot be generalized across different task complexity.

Wood (1986) described task complexity in term of two aspects: First, component

complexity, relating to the number of information to be processed and steps to be executed in

performing a task. A task has more complexity when the level of information and the work

process is high. Second, coordinative complexity referring to the amount of coordination

required to perform the task and it is becoming more complex when steps in one part of the

task has a high dependency on steps taken earlier, when several related steps must be

performed all at once, and when backward/forward reasoning is required.

Cloyd (1997) stated that effort (both of duration and intensity) can affect quality of

auditor’s work. On his study, Cloyd (1997) used assumption that the tasks given to subjects

were of high complexity tasks. However Tan and Kao (1999), Sanusi and Iskandar (2007)

and Bonner and Spilker (2002) found that quality of audit judgment would be affected by the
7
level of task complexity faced by auditors.

Especially for an accountability case, a study done by Tan and Kao (1999) found that

for low complexity task, accountability will not improve performance; for medium

complexity task, higher accountability subject will result a better performance when auditor

have high knowledge; and for high complexity task, higher accountability will result a better

performance when auditors have both high knowledge and high problem solving ability.

Similar study done by Sanusi and Iskandar (2007) found that for a lower task

complexity, high effort would lead to high quality of auditor’s work. However for a higher

task complexity, high effort would not significantly improve quality of work.

According to the discussion above, the following hypothesis is proposed:

H3: The relation between effort and quality of auditor’s work is moderated by task

complexity;

H3a: When task complexity is low, subject with high effort show better quality of

work than subject with low effort

H3b: When task complexity is high, there is no difference in quality of auditor’s work

between high effort and low effort groups.

Model of Study

Based on some prior studies done by Tan and Kao (1999), Tan et.al (2002), Bonner

and Spilker (2002) and Sanusi and Iskandar (2007), we try to integrate the effect of effort as

mediator and task complexity as moderator on the relationship between accountability and

8
audit judgment performance. Figure I below draw that relationship:

FIGURE 1
Model of Study

path a
path b path c

path d

METHOD

Subject

Some accountants who work in public accounting firms in Pekanbaru (Riau

Province) and Padang (West Sumatera Province) Indonesia were taken as subjects. In order to

get more respondents and accurate conclusions, the two locations were chosen because of

their close distance and their similar social economic condition.

Subjects were chosen by non probability method based on purposive sampling. The

respondents chosen for this study have to have the experiences in financial audit. This

criterion was important in order to give accurate data. In fact, not all of public accountants in

audit firms do financial audit. Some of them handle management service, tax matter and

others.

Data Collection

Data was collected by survey method. Questionnaires are sent to respondent

directly. Before distributing the questionnaires, we had classified subjects randomly into two

groups: high accountability subjects and low accountability subject. Subjects who were
9
selected into high accountability group were informed that their task would be reviewed by

partner and they should provide their name. In contrast, low accountability subjects were not

given this instruction. They only asked for their cooperation in fulfilling the questionnaires.

Sixty five (65) auditors from fourteen (14) public accounting firms in Padang and

Pekanbaru had been selected as subjects. Earlier examination found that three questionnaires

from high accountability group could not be used due to incomplete responses. So that, the

number of questionnaires to be analyzed were 62 questionnaires with the final response rate

of 44, 29%.

Variable and Measurement

For this study, we use accountability variable as independent variable; audit

judgment performance as independent variable, effort as mediator and task complexity as

moderator. This study used instrument used by some prior researchers. Some of them are

Libby (1985), Bonner and Lewis (1990), Tan and Kao (1999), Sanusi and Iskandar (2007).

Original instrument was written in English, so for the purpose of this study, the instrument

was translated into Indonesian language. This research instrument was pre tested on some

auditing lecturers and practicing auditors to ensure that information and narratives in the case

were realistic in performing the audit task.

Accountability

Accountability is defined as a link of social-psychology urging of auditor to complete

their duty and respond it to their environment. On this study, accountability was manipulated

as a between-subject variable and was assessed by questions about how motivated they were

to perform the task, their confidence that their work would be reviewed by their partner and

how serious they were in doing the case. Those motivation, confidence and effort are

measured with nine point scales (e.g. 1= not all motivated, 9=extremely motivated).

10
Effort

Effort is measured by amount of time devoted by subject to complete audit task/case

in the questionnaire. Subjects were asked to write the time began and the time they ended the

tasks. The using of time duration as proxy of effort variable had been used in some prior

researchers likes: Jiambalvo (1979); Libby and Lipe (1992); Sanusi and Iskandar (2007).

Task Complexity

Task complexity is a within–subject variable. Subjects were given some audit case

with different level of task complexity, low and high. For low complexity tasks (Part I tasks),

subject were required to complete tasks on: compliance test, to ascertain whether the client’s

control on payables and liabilities were effective; to list substantive test, to search unrecorded

liabilities; and to list financial statement errors from the weakness in the client’s control over

the account payable system. While, for high task complexity (Part II tasks) subjects were

required to complete tasks about ratio analysis and listing errors that could have caused

chances in some financial ratios. For the data analysis, high complexity task, we gave point 1

and for low complexity task we gave point 0.

Audit Judgment Performance

Performance is determined by the number of correct responses compared with

standard criteria. In this case, quality of work was appraised by looking at correct responses

given by auditor for every audit task in the questionnaire. More correct responses given by
11
auditors were interpreted so that the auditor’s work would be more qualified. For compliant

test and substantive test, the maximal score was 7. For listing financial statement errors, the

maximal score was 12 and for ratio analysis case, the maximal score was 15.

Manipulation Procedures

On this study, there are two conditions manipulated or controlled:

a. Task Complexity Manipulation

Before distributing questionnaires, authors had classified task complexity into two

levels: high and low complexity task. To proof that our manipulation done successfully we

rated respondent’s opinion about the level of task complexity they faced. Based on their

answers, we found that they rated the tasks 1 (low complexity tasks) were easier than the

tasks 2 on the questionnaire. An average subjects rated task 1 was 7.38 and the task 2 was

8.95 (1=low complexity task to 9= high complexity task). Based on paired samples test, the

mean score for the manipulation check on the low complexity task was significantly different

from the high complexity task (t=-19.321 p=0.00). These results strongly support that,

manipulation done to task complexity was a success.

b. Accountability Manipulation

In previous part of this paper, we had been discussed that subjects in this study were

divided into two groups: high accountability and low accountability subject. High

accountability subject were informed that the result of their tasks would be reviewed by

manager and they were asked to write their name. Whereas, low accountability subject were

not given this instruction and we only asked their willingness to fulfill every question in the

questionnaire without being reviewed from their manager.

Tan and Kao (1999), Libby and Luft (1993) and Cloyd (1997) found that high

accountability subjects tended to have higher motivation to complete their tasks, have higher

12
confidence level knowing that their manager would review their tasks and they have higher

effort. In contrast, low accountability subjects tended to believe that their work would not be

reviewed by their manager, have low motivation and low effort.

To test whether our manipulation on accountability variable done as predicted, we

had done manipulation check by evaluating three indicators; motivation, confidence level and

effort. This evaluation purposed to examine whether those used indicators were proper to

evaluate the level of individual accountability. The evaluation done in two steps, first we

analyzed the effect of accountability on those three indicators. The results show that

accountability has significant influence on motivation (β = 0.968, p<0.01), confidence (β =

1.335, p<0.01) and effort level ((β = 0.693, p<0.01). It means that all the three indicators are

proper to be used as indicators to assess individual accountability.

The second step taken was to check whether the three indicators: motivation,

confidence level and effort differ between the high accountability subjects and low

accountability subjects. Evaluation on the success of accountability manipulation was done

by comparing the mean value of motivation, confidence and effort between those two groups.

Average value of motivation, confidence and effort between high accountability subject and

low accountability subjects can be seen on the table below.

TABLE 1
Average Value of Motivation, Confidence and Effort
High accountability Low accountability
subjects subjects
Motivation 8.10 4.91
Confidence 8.89 4.00
Seriousness 7.84 5.71

13
 We found that high accountability subjects were more motivated (means = 4.91 and

8.10, respectively, p=0.00), were more likely to think that they would be reviewed (means =

4.00 and 8.89, respectively, p=0.00) and exerted more serious than less accountability

subjects (means 5.71 and 7.84, respectively, p=0.00). These results proof that manipulation

procedure on subject accountability has been done successfully.

Data Analysis

This study examines the effects of mediator and moderator simultaneously. For

mediating effects, Three Steps of Multiple Regression recommended by Baron and Keanny

(1986) and Frazier et.al (2004) was applied. This study also followed the presentation of

data analysis done by Sanusi and Iskandar (2007). According to Frazier (2004), a mediator

variable can be full mediation or partial mediation variable. To Proof it, we can see it by

comparing the value change of unstandardized coefficient (B) of variable independent

(accountability), before and after mediator (effort) was included in analysis. If the relation

between the predictor and the outcome controlling for the mediator is zero, the data are

consistent with a complete mediation. However, if the relation between the predictor and the

outcome is significantly smaller but still greater than zero the data suggest partial mediation.

Since this study examines the effect of mediator and a moderator simultaneously, this

three step technique is analyzed trough a hierarchical regression analysis. For manipulation

check (involving complexity and accountability manipulation), we used the means and

analysis of variance (ANOVA).

ANALYSIS AND RESULTS

Descriptive Analysis

The sample comprised 62 (sixty two) auditors from fourteen (14) public accounting

firm in Padang and Pekanbaru- Indonesia. The average age of subject is 27 years, ranging
14
from 22 to 47 years. Ninety seven percent of the samples hold undergraduate degree and

three percent of them hold master degree. On average subject had 4.3 years of experience

(min 6 months, max 16 years).

The table 2 below present descriptive statistics on effects of two accountability

condition on audit judgment performance under low and high complexity tasks.

TABLE 2
Descriptive Statistic and Mean (Standard Deviation) of Effort and
Audit Judgment Performance
Task 1 (low level
Sample Task 2 (High level
Accountability Effort of task
Size of task complexity)
complexity)
High 66.27
Accountability 29 (12.74) 57.71 (17.22) 14.94 (10.33)
58.93
Low Accountability 33 (13.00) 20.92 (16.68) 4.84 (8.37)

Test of Hypothesis

This table describes the result of regression analysis of accountability as independent

variable, effort as mediator, task complexity as moderator and audit judgment performance as

dependent variable.

TABLE 3
Regression Analysis of Audit Judgment Performance
Testing Steps in Mediation
B SE B ∆ R Square F Change
Model
Testing Step 1 (Path a)
Dependent: Audit Judgment
Performance

15
 Independent: Accountability 6.858** 0.905 0.320 57.372

Testing Step 2 (Path b)

Dependent: Effort
Independent: Accountability 2.408** 0.568 0.128 17.946

Step 3 (Path c & d)

Dependent: Audit Judgment
Performance
Mediator: Effort 0.412* 0.126
Independent:
1. Accountability 6.813 0.621
-20.
2. Task Complexity 453 11.002 0.360 135.153
-
3. Effort X Task Complexity 0.786** 0.173 0.047 20.739

*p<0,05, ** p<0,01

To test mediation effect, we used the three step regression as suggested by Baron

and Keanny (1986). Step 1 examined the direct effect of accountability on audit judgment

performance. From data analysis, we get the unstandardized regression coefficient associated

is 6.85 (p<0.01). It is shown that there are significant effect between accountability and audit

judgment performance. It means, subject with higher accountability give a better audit

judgment performance than subject with lower accountability (H1 is supported).

In step 2, we examined the effect of accountability on effort (H2). The ustandardized

regression coefficient (B) 2.408 associated with these relation were also significant at the p

<0.01 level (path b was significant). Hence subject with high accountability excert more

effort than subject with low accountability (H2 is received).

The step 3 was to examine the effect of accountability (independent), effort

(mediating) silmoutaniously on audit judgment performance. However, in performing step 3,

we have entered not only mediator but also the interaction factor of effort and task

complexity. In this case, the effect of effort as mediator varies across the level of task
16
complexity (moderator).

The result shows that effort influenced the model significantly (0.412, p<0.05). A

significant indirect effect (path c) indicated that a significant amount of independent

variable’s total effect on the dependent variable occurs trough the mediator. This implies that

effort mediates the relationship between accountability and audit judgment performance.

To proof whether effort is partial mediation or full mediation, we see from the

change of B coefficient before and after effort (mediator) which was included in analysis.

Before including effort in analysis, the ustandardized regression coefficient for accountability

was 6.858. But after effort was been included in analysis, ustandardized regression

coefficient for accountability decreased be 6.813, but bigger from zero. This analysis had

proofed that effort has taken a role as a partial mediation variable. To know the significant or

unsignificant of the effort as partial mediation variable, then we can calculate Z statistic. The

result shows that Z value was 7.65 (>1.96). It means that effort has significant effect as

mediation.

In addition to the mediation effect of effort, result in step 3 also demonstrated

evidence on the interaction of effort and task complexity. The ustandardized regression

coefficient for the interaction between effort and task complexity was -0.785 (significant at

the p<0.01). This interaction term explained an additional 4.7% of the variance in audit

judgment performance and 36% expalained by the main effect of effort and task complexity

alone. Result in step 3 show that task complexity interacted with effort to moderate audit

judgment performance (Path d). It means, H3 was supported.

For detailed discussion, the effect of interaction between effort and task complexity

can be seen on following figure:

17
 FIGURE 2
Interaction between Effort and Task Complexity

60.40

Audit Performance 50.40

40.40
Low level of task complexity
30.40
High level of task complexity

20.40

10.40

0.40
SD-1 Mean SD+1
Effort

Figure 2 shows the predicted values for quality of audit judgment at the mean and at

low (-SD from the mean) and high (1 SD fro the mean) values of effort for the two level of

task complexity. Effort exhibited a positive relationship with performance work under low

level of task complexity. It could be summarized that subject with high accountability are

more likely to exert more effort, which in turn lead to better quality of work when task is less

complex (H3a was supported). In contrast, a very modest negative relation between effort and

quality of audit judgment was shown under high level of task complexity. An increase in

effort was less likely to influence better quality of audit judgment under complex task (H3b

was supported). Thus, accountability only improves audit judgment under low task

complexity through the increase in the level of effort.

CONCLUSION, LIMITATION, IMPLICATION AND FUTURE STUDIES

Conclussion

This study purposed to examine the effect of accountability and effort on audit

judgment performance through effort under different level of task complexity. Based on the

result of data analysis, we found that auditors who have high accountability give better
18
performance than auditors with lower accountability. This result was consistent with Sanusi

and Iskandar (2007), Kennedy (1993), Cloyd (1997) who stated that performance incentive

both of financial (monetary) and non financial incentives (such as accountability, feedback,

justification) has significant effect on quality of auditor’s audit judgment.

We also found that effort has role as partial mediator (evaluated by using Z statistic

and the change of ustandardized regression coefficient) in relationship between

accountability and audit judgment performance. Auditor who has high accountability will

exert more effort (duration) than auditor with low accountability. This findings were

consistent with studies done by Sanusi and Iskandar (2007); Kennedy (1993); Cloyd (1997)

and Bonner and Spilker (2002). In addition, the relationship between effort and audit

judgment performance was moderated by task complexity. When task complexity was low,

auditor who has high effort would give a better quality of work. But, when task complexity

was high, there is no difference performance between auditor who has high effort and low

effort. This findings also consistent with those of Sanusi and Iskandar (2007).

Limitations, Implication and Future Study

There are several limitation for this study. First, the external validity of this study is

limited since the case contains less information than the real audit enviroment. In the real

audit enviroment richer information will influence the quality of auditor’s work. Second, the

sample size is relatively small. Future study should attempt to elaborate using larger samples.

This would enhance the external validity of the findings. Third, researcher had less controll

during manipulation and experiment process. Future studies should attempt to get full control

during manipulation and experiment process. It is suggested to use method that applied by

Tan & Kao (1999) and Cloyd (1997) in which all subject gathered in the same place and time,

19
then classifing them to high accountability group and low accountability group and then

giving the same treatment for every subject adjusted with their group. While, on this study,

manipulation procedures both of high accountability group and low accountability group

were done for every partner in every public accounting firm. It might cause dissimilar

treatment for every subject classified to similar group.

Although this research were just done in small scale, we expect that it could serve

as a useful contribution for practicioners. Our research have an implication for practiced

interms of ensuring adequate performance (quality of audit judgment) for task of vaying

complexity. One of solution to result a better quality of auditor’s work is that the accounting

firms should ensure that their auditors have the approriate accountability level. Because

auditors with high accountability will devote more effort to result in a more qualified work.

The result of this study also found that varying level of task complexity assigned to

the auditors may not produce the same quality. It shows that the differences in the task

complexity may interact with other factors. Audit firms should ensure that auditor have a

proper training to increase their skill and knowlege to help auditor to carry out various job

complexities.

Accountability, effort and task complexity are not the only factors that affect quality

audit judgment. Other motivational devices such as financial incentives can be investigated

(Bonner and Sprikle 2002). We also suggest to investigate the effect of other performance

incentives on varying effort (such as effort intensity and direction), because those two kind of

efforts may give different effects on quality of auditor’s work (Bonner and Sprikle 2002).

Furthermore, next researchers can analyze the effect of accountability on audit judgment

performance based on evaluative audience of auditors: whether auditor is accountable to

partner, accountable to client or accountable to no one (Buchman et.al. 1989 in Messier and

Quilliam 1992; Lord 1992)

20
 REFERENCES

Ashton, R. H. 1990. Pressure and performance in accounting decision settings: Paradoxical

effects of incentives, feedback and justification. Journal of Accounting Research 28:
148-180.

Aswathi, V and Pratt, J. 1990. The effect of monetary incentives on effort and decision
performance: The role of cognitive characterictics. The Accounting Review 65: 797-
811.

Baeley. C.D., Brown, L.D. and Coco A. F. The effefct of monetary insentive on worker
leraning and performance in assembly task. Journal of Management Accounting
Research 10:119-131.

Barron, RM and Kenny, D. A. 1986. The moderator mediator variable distiction in social
psychology research: Conceptual, strategic, and statistical considerations. Journal of
Personality and Social Psycology 5 (6): 1173-1182.

Bedard, J. and Chi, M.T.1993. Expertise in auditing. Auditing: Journal of Practice & Theory
12: 21-45.

Bonner, S.E and Sprinkle, G.B. 2002. The effect of monetary incentive on effort & task
performance: Theories, evidence and framework Of research. Accounting,
Organization and Society 27 (4/5): 303-345.

Bonner, S.E and Lewis, B. L.1990. Determinat of Audit Expertise. Journal of Accounting
Research 28: 1-28.

Chaiken, S.1980. Heuristic versus systematic information processing and the use of source
versus message cues in persuation. Journal of Personality and Social Psychology
(November): 752-766.

Chang, C. J., Ho J.L.Y, and Liao, W.M. 1997. The effect of justification, task complexity and
experience/training on problem solving performance. Behaviour Research in
Accounting 9: 98-116.

Cloyd, C. B. 1997. Performance in tax reseach Tasks: The joint effect of knowledge and
accountability. Journal of Accounting Review 72 (1): 111-130.

Frazier, P.A. and Tix A.P. and Barron, K.E. 2004. Testing mederator and mediator effects in
conseling psychology research. Journal of Conseling Psychology 51(1): 115-134.

21
Jiambalvo, J. 1979. Performance evaluation and directed job effort: Model development and
analysis in a CPA firms setting. Journal of Accounting Research 17 (2): 436-455.

Kennedy, J. 1993. Debiasing audit judgment with accountability: A framework and

experimental results. Jurnal of Accounting Research 30(Autumn): 231-245.

Libby, R. and Luft J. 1993. Determinant of judgment performance in accounting setting:

Ability, knowledge, motivation and environment. Accounting, Organization & Society:
425-450.

Lord, A.T. 1992. Pressure: A methodological consideration for behavioral research in

auditing. Auditing: A Journal of Practice & Theory (Fall) 89-108.

Libby, R. and Lipe, M. G. 1992. Incentives, efforts and cognitive processes involved in
accounting- related judgments. Journal of Accounting Research (Autumn): 249-273.

Mardisar, D. and Sari, R.N. 2007. Pengaruh akuntabilitas dan pengetahuan terhadap kualitas
hasil kerja auditor. Paper presented in Simposiun Nasional Akuntansi X (Makassar-
Indonesia).

Messier, W. F Jr. and Quilliam, W.C.1992. The effect of accountability on judgment

develompment of hypothesis for auditing. Auditing: A Journal of Practice & Teory 11:
123-138.

Sanusi, Z. M. and Iskandar, T.M. 2007. Audit judgement performance: Assessing the effect
of performance incentive, effort and task Complexity. Manajerial Auditing Journal 22
(1): 34-52.
Snead, K. and Harrell, A. 1991. The impact of psychological factors on the job satisfaction of
senior auditors. Behavioral Research in Accounting 3: 85-96.

Tan, H.T, Ng, T.B.P and Mak, B.W.Y. 2002. The effect of task complexity on auditor
performance: The impact of accountability and knowledge. Auditing: A Journal of
Practice & Theory 21(2 Suplement): 81-95.

Tan, H. T. and Libby, R.1997. Tacit managerial versus technical knowledge as determinant
of audit expertise in the field. Journal of Accounting Reseach (Spring): 97-113.
Tan, T. H. and Kao, A.1999. Accountability effect on auditor’s performance: The influence
of knowledge, problem solving ability and task complexity. Journal of Accounting
Research 37 (1): 208-223.

Teclock, P.E and Kim, J,L. 1987. Accountability and judgment process in personal prediction
task. Journal of Personality and Social & Psychology (April): 700-709.

22
23