Visual learning methods: strategy for mitigating unethical decisions in accounting education

* Corresponding author, email address: intiyas.utami@uksw.edu ABSTRACT The purpose of this study is to examine the causality between obedience pressures and unethical decisions making and whether the causality can be mitigated by visual-based ethics learning methods. This study employs 2x2 between subjects factorial experiment methods with 114 participants from undergraduate students in accounting. The result of this study shows that there is a causal relationship between obedience pressures and unethical decisions making and that visual-based learning methods have encouraged subjects to make more ethical decisions. In addition, this study also finds that the most ethical decision outcomes are generated by the subjects who are under high obedience pressures and receive visual-based ethics learning methods.


INTRODUCTION
The accounting profession is susceptible to dysfunctional behavior due to obedience pressure. It is also argued that the pressure to obey certain authority causes a person to do negative things that are contrary to normative things (Lord & DeZoort, 2001). In addition, the pressure of obedience causes a person to behave contrary to his personal values and act unethically (Milgram, 2009). According to Craig Smith, Simpson, & Huang (2007), employees commit illegal actions at work because of the pressure exerted by their supervisors. Anand, Ashforth, & Joshi (2004) state that orders from the top leaders of an organization rationalize fraud by management. Davis, DeZoort, & Kopp (2006) also show that management accountants carry out budgetary slack due to pressure received from their seniors. According to Lord & DeZoort (2001), obedience pressure causes auditors to give a fair opinion on financial statements when in fact there are material misstatements. Cahyaningrum & Utami (2015) state that audit decisions become inaccurate2 due to the pressure of obedience in high task complexity.
Experts agree that increasing moral reasoning through professional ethics education is an important component in the effort to reduce malfunctioning behavior in the accounting profession. Previous research has shown that moral reasoning has an important influence on the attainment of ethical awareness (Gaffikin & Lindawati, 2012), resulting in better decisions and ethical behavior (Ponemon, 1992;Abdolmohammadi & Baker, 2006;Thomas, 2012). Shawver (2006) states that professional education programs play an important role in raising awareness of accounting students of the importance of applying ethics in dealing with accounting problems. Williams & Elson (2010) also reveal the importance of ethical emphasis in the educational curriculum. Most business schools in Australia agree that ethics education must be included in the accounting curriculum (Dellaportas, 2006). According to Shaub, Finn, & Munter (1993), ethical education can increase ethical sensitivity. Huss & Patterson, (1993) also state that the moral development of accounting students can be improved through the educational process. Guffey & McCartney (2008) explain that effective ethics teaching is important for developing moral thinking for accounting students.
Because of the importance of ethical education for prospective accounting professionals, ethics education researchers have examined innovative methods in ethics education to better embed ethical values in students and professionals (Liu, Yao, & Hu, 2012). However, although researchers agree on the importance of ethics education in higher education, they have not agreed on the right method for learning ethics. Haywood, McMullen, & Wygal (2004) use the game method in learning ethics and increasing professional responsibility for accounting students. Falkenberg & Woiceshyn, (2008) provide empirical evidence that the case study method can improve ethics.
By observing that an unethical decision on prospective accountant students is important to be mitigated, a study conducted by Ricchiute (1984) provides empirical evidence that visual information can improve audit decisions. It was found that visual learning methods are also thought to be capable of mitigating unethical decisions. This study is conducted to examine the causality between obedience pressure and unethical decisions, and how it can be mitigated by visual-based ethical learning methods. This study is expected to provide further empirical evidence about the use of visual-based ethical learning methods to enhance better ethical understanding.
The results of the study above provide practical contributions to accounting education institutions related to ethics learning methods for prospective accountants in carrying out their professions. Appropriate ethics learning methods are expected to be able to mitigate the impact of obedience pressures that arise when prospective accountants enter the workforce. For the development of behavioral research, this research has contributed in testing the theory of obedience pressure and its interaction with learning theory.

THEORETICAL FRAMEWORK AND HYPOTHESES Obedience Pressure
Obedience pressure is one form of social influence pressure. It is negative and arises from those who have more authority. There are two other forms of social influence pressure: compliance pressure and conformity pressure. Compliance pressure is almost the same as obedience pressure, but compliance pressure arises from the social environment at the same level. Conformity pressure is positive and refers to pressure to conform to existing norms in society. Of the three types of social influence pressure, obedience pressure has a greater force because the pressure from those who have authority is more effective than the pressure exerted by the social environment at the same level (Davis et al., 2006;Baird & Zelin Ii, 2009).

Learning Theory
Learning theory is a conceptual framework that describes how information is absorbed, processed and maintained. The strategy of mitigating unethical decisions can be explained by learning theory. The dimensions of cognitive, emotional, environmental, and past experiences play a role in the process of how understanding is acquired or changed and maintained (Illeris, 2004;Ormrod, 2012). Learning in education is the interaction between students and educators in the process of acquiring knowledge and the formation of attitudes. According to Robbins & Judge (2009), the learning process is a process of forming behavior. The right learning method is able to be a strategy for mitigating unethical decisions on prospective accountants.

Moral Reasoning and Ethical Decisions
Moral reasoning can be defined as an argument about how someone should act or give reasons to justify or criticize certain behaviors (Fox & DeMarco, 2000). Thompson (1998) concludes that moral reasoning consists of three points of view: (a) thinking about what a person must do and why it is necessary to do it; (b) forming ideas to describe and evaluate actions; and (c) assessing certain actions based on general rules that apply. Gaffikin & Lindawati, (2012) define that moral reasoning is a person's argument that aims to explain the process of ethical decision making, or explain a person's process of building behavior or actions based on individual moral judgment (cognitionjudgment action process). Adams, Malone, & James (1995) conclude that the process of one's moral reasoning can be understood as the process of how one internalizes moral standards.

Visual-based Learning Methods
Learning methods are the methods chosen to convey material in a particular learning environment so that goals can be achieved effectively and efficiently (Gerlach, Ely, & Melnick, 1980). Schunk (2012) states that learning has the aim to focus the attention of learners on the processes and methods that help them gain abilities and improve skills. Ability and skills are broad depending on the learning process that is followed by the learners (students). Ethics learning in accounting education has the same goal, to help learners gain abilities and skills in relation to ethical attitudes and behavior.
Visualization refers to visual information presented in the form of graphics or images, while non-visualization refers to information that is presented only in the form of text (Tang, Hess, Valacich, & Sweeney, 2014). More importantly, financial statements that are presented in a visual form allow one to obtain better information and make more accurate decisions. Tang et al. (2014) states that decision makers who receive financial information presented in a visual form enable someone to obtain better information and make more accurate decisions. Likewise with the ethics learning method, visual-based ethics learning methods are expected to provide better information, high ethical understanding, and help students to be able to make ethical decisions.

Relationship between Obedience Pressure and Ethical Decisions
Social pressure is a manifestation of people's preferences towards one's social performance (Baron, 2009). There are three forms of social pressure influence: compliance pressure, obedience pressure, and conformity pressure. Of these three forms of social pressure influence, obedience pressure has the most power to put pressure on someone in the decision making process. Davis et al. (2006) strengthens the results of research conducted by Lord & DeZoort's (2001) that auditors are more likely to make unethical decisions when exposed to obedience pressure from superiors. This shows that obedience pressure influences ethical behavior which means that it also influences moral reasoning. The level of reasoning in individuals who receive high obedience pressure will be different from individuals who do not receive obedience pressure. In conditions of high obedience pressure, individuals will tend to take orders from leaders even if it is unethical to do so. From the arguments above, the first hypothesis is formulated as follows: H1: Subjects without obedience pressure have better ethical decisions than subjects with high obedience pressure.

Relationship between Ethics Learning Methods and Ethical Decisions
Ethical decisions are reflected in moral reasoning which is a cognitive development of an individual. Research on ethics is largely based on Kohlberg's research which builds a theory of moral cognitive development, and also Rest, who builds the development of a model of ethical action that culminates in a neo-Kohlbergian approach (Liu et al., 2012). Thorne (2001) classified research and investigated steps to improve individual cognitive abilities. He found that cognitive development could be improved through educational interventions. According to Özkan (2013), students use their moral cognitive abilities to make decisions, and this conclusion provides an opportunity for educators to find alternative approaches that encourage accounting students to use fundamental opinions in solving ethical accounting dilemmas. Research conducted by Delaney & Coe (2008) provides the result that ethics education is also found to be effective in improving the moral abilities of accounting students. From the results of previous studies it can be concluded that ethics education can improve the moral reasoning of accounting students.
Ethics learning methods in accounting education can vary. The learning that is commonly used is the lecture/ tutorial method. The tutorial method takes the form of a teacher's explanation to students and is usually followed by questions and answers about the contents of the lesson that are not yet clear (Suparman, 2012). This tutorial method is classified in the non-visual learning process (only presented in verbal form). Learning experiences in verbal form are abstract. Ayunanda & Utami (2015) use visualization to mitigate recency effects, in which the use of visualization in presenting information can increase ethical decisions.
The learning method using visual approach is more concrete than using nonvisual approach. Therefore, visual experiences make learners store more information. According to Schunk (2012), the human body is structured in such a way that we can enter more information through the sense of sight than through the other four senses. Furthermore, visual displays help increase attention, learning, and maintaining it, Likewise with ethics learning in accounting education, ethics learning methods with a visual approach can make accounting students pay attention and retain their memories of ethical values and thus can improve their moral reasoning so that the decisions made are more ethical. Based on the arguments and the results of previous research, the second hypothesis can be formulated as follows: H2: Subjects with visual-based ethics learning methods have better ethical decisions than subjects with nonvisual-based ethics learning methods.

Interaction among Obedience Pressure, Learning Methods, and Ethical Decisions
Obedience pressure influences ethical behavior and moral reasoning (Lord & DeZoort, 2001). Educational institution, especially accounting education, has an important role in improving the ethical behavior of prospective accountants. Innovations in ethics education can be developed through learning methods. Ethics learning methods with a visualization approach are expected to be able to improve moral reasoning compared to the non-visualization approach. Visual senses are more dominant and visual experiences make learners store more information (Dilla & Janvrin, 2010;Schunk, 2012). The support of visual-based learning methods will increase the level of reasoning because cognitively acquired knowledge is easier to understand. In the condition that the subjects receive high obedience pressure, visual-based ethics learning methods are expected to help learners achieve better ethical understanding so that they are able to make more ethical decisions. The interaction among obedience pressure, learning methods, and ethical decisions is formulated in the third hypothesis as follows: H3: Under conditions of high obedience pressure, subjects who get visual learning methods will produce the best ethical decisions.

RESEARCH METHOD Design of Experiment
This study uses a laboratory experimental method. The experimental method is chosen because it has a high internal validity to test the causality between the dependent variable and the independent variable. The experimental design is a 2 x 2 factorial design between subjects. The first factor is obedience pressure which consists of two levels: with obedience pressure and without obedience pressure. The second factor is the learning method, which consists of visual-based learning and nonvisual-based learning.
The subjects are undergraduate students of Accounting Study Program, Faculty of Economics and Business, Satya Wacana Christian University, Salatiga. The number of subjects used in this experiment is 114 people. Students are selected from the internal audit class with the argumentation that the students already understand the practices and assignments of internal audits, specifically compliance testing. The selection of students as subjects is in accordance with the purpose of the research designed. This research emphasizes the cognitive aspects of humans in processing information and making general decisions so that the use of students as professional counselors can be accepted scientifically (Nahartyo & Utami, 2015). For example, Fleming, Romanus, & Lightner (2009) examine ethical dilemmas in a professional context by using students as subjects who act as professional accountants.
Experiment is carried out using paper and pen and supported by visualization in the form of an ethics learning video. The experiment module is executed after going through a pilot test with the subject of accounting students who are not the target of the experiment, with the hope that there will be no effect of maturase on experimental subjects. Preliminary test results indicate that the video still needs to be improved in terms of ethical crime cases that need to be clarified, so that improvements are made to the visualization video. The experimental module is developed from the research of Cahyaningrum & Utami (2015) by modifying the learning method. Ethics learning methods are made in the form of nonvisual (narrative) and visual (video).
The independent variables (manipulated variables) are obedience pressure and ethics learning method, while the dependent variable (the measured variable) is an ethical decision. Obedience pressure is defined as negative pressure, that is, to take actions that are unethical and arise from those who have higher authority. The obedience pressure is manipulated in the form of pressure to violate government regulations for environmental preservation. Learning methods are defined as the methods chosen to convey material in a particular learning environment so that goals can be achieved effectively and efficiently.  Subjects are asked to act as members of the organization's internal auditors and tasked with making decisions on the compliance of an organization with environmental maintenance regulations. 4. Next, the subjects face three questions relating to manipulation check for understanding their task and role in the company where they work. 5. Manipulation is given according to the case faced by each subject. High obedience pressure is indicated by orders to violate government regulations on environmental preservation. The company is faced with maintaining the forest because the raw material for furniture is wood. The company leaders ask the subject not to prepare a compliance audit report and advise that it not be submitted when the external audit is carried out by the Public Accounting Firm. The company leaders order management not to comply with government regulations because it has an impact on the costs the company has to issue. To show social pressure, it is presented the conditions of competing companies that are performing well despite violating regulations. The control group is not under obedience pressure as indicated by orders from leaders to management to comply with government regulations on environmental preservation. Subjects in the control group are encouraged to conduct compliance audits as a form of added value to the organization. 6. Manipulation in the form of learning method is given in visual in the form of video and nonvisual in the form of narration. Subjects who receive the visual learning method are given a video on the importance of protecting the environment and the impact of environmental damage. Subjects who receive nonvisual learning methods are given narration on the importance of protecting the environment and its impact on the environment. 7. The stage after giving manipulation is checking the manipulation. 8. The last stage is debriefing, as a stage to explain the aims and objectives of the study.

Analysis Technique
The steps taken in the analysis are as follows: 1. Descriptive presentation of research subject data 2. Testing the effectiveness of randomization with One-Way ANOVA (the dependent variable is ethical decisions and the independent variables are gender, age and student achievement index).

Testing manipulation check is done by
determining the subject's answer score for the three questions given. If the subject answers two or more questions correctly, then the subject passes the manipulation check. 4. The first and second hypothesis testing is done using the independent t-test. The third hypothesis testing is done using the Two-Way ANOVA (Analysis of Variance) Test. The first and second hypotheses are testing the main effect, while the third hypothesis is testing the interaction effect. The hypothesis is supported if the significance value is less than 0.05. Analysis of the data is based on literature review and test results.

DATA ANALYSIS AND DISCUSSION
The subjects of this study consist of 114 students divided into 4 (four) cells: Cell 1, Cell 2, Cell 3, Cell 4. The distribution of subjects in the cells can be seen in table 2. The number of subjects in each cell is relatively comparable. The experiment was conducted on November 23, 2015 in the Internal Audit class at Faculty of Economics and Business, Satya Wacana Christian University, Salatiga. Audit simulations were led by experimenters who acted according to the directions and guidelines in conducting the simulation. The audit simulation process took 50 minutes.
The next step is checking the first stage of manipulation, that is, to test the understanding of the task and the role of the subject in the simulation with 3 questions. The prerequisite for passing the first manipulation check is that the subject correctly answers a minimum of 2  Source: Processed data of the 3 questions presented. All participants passed the first stage of manipulation check.
The second manipulation check is to test the understanding of the situation and conditions faced by the subject. In the second stage of manipulation check, there were 9 or 7.89% of participants who did not pass. Table 4, shows the demographic characteristics of the research participants which include gender, age, and GPA. Participants consist of 30 men and 75 women with a majority of GPA in the range of 3.01 -3.5 (52%). Age of majority participants is 20-21 years. Source: Processed data Subjects have various demographic characteristics. In experiment, a randomization is said to be effective if there is no influence of demographic characteristics on the dependent variable (unethical decision). Statistical testing with One Way ANOVA is conducted to provide support that randomization has been effective. The results of the One Way ANOVA test are presented in Table 5.
ANOVA test results in Table 5 show that there is no influence of demographic characteristics including age, gender, and GPA on ethical decisions (significance above 0.05). Thus, it can be concluded that the results of inter-cell research are the result of manipulation received by each cell and not because of differences in subject characteristics.

Testing the Relationship between Obedience Pressure and Ethical Decisions
The first hypothesis states that subjects without obedience pressure will give better ethical decisions than those with high obedience pressure. The results of the first hypothesis statistical test were performed using an independent t-test by comparing between subjects who received high obedience pressure (cells 1 and 2) and those who did not receive obedience pressure (cells 3 and 4). The results of testing hypothesis 1 can be seen in Table 6. Statistical test results show a significance level of 0.000 <0.05. This means that the obedience pressure influences ethical decision making. Subjects who are not under obedience pressure have better ethical decisions than subjects who are under high obedience pressure. Thus the first hypothesis is supported. The process of ethical decision making is a process by which a person builds a behavior or action based on an individual's moral judgment. This process is influenced not only by internal factors (personal values), but also by external factors (the influence of social pressure, one of which is the pressure of obedience). The results of the statistical test in table 6 show that obedience pressure drives someone to make unethical decisions. This can be seen from the significant difference in the average decision (15.290) between the average decision of the group that receives high obedience pressure and the group that does not receive obedience pressure. The average decision of the subjects who are under obedience pressure is 75.094, while the average decision of the subjects who are not under obedience pressure is 90.384. This shows that there are significant differences in decision making between those who are under obedience pressure and those who are not under obedience pressure. The obedience pressure from those who have more authority gives an influence on the ethical decision making of their subordinates. Better ethical decisions are produced by groups without obedience pressure. The results of this study are consistent with the results of previous research conducted by Davis et al. (2006) which also strengthens the results of research conducted by Lord & DeZoort's (2001) that auditors are more likely to make unethical decisions when exposed to obedience pressure from superiors. The results of this study provide empirical evidence that obedience pressure affects a person in making decisions.

Testing the Relationship between Ethics Learning Method and Ethical Decision
The second hypothesis predicts that subjects who get visual-based ethics learning methods have better ethical decisions than those who get nonvisual-based ethics learning methods. The testing of manipulation of ethics learning methods is carried out using an independent t-test by comparing betweenthe subjects who get visual-based ethics learning methods (cells 1 and 3) and the subjects who get nonvisualbased ethics learning methods (cells 2 and 4). The test results of hypothesis 2 can be seen in table 7.
The result of statistical test of hypothesis 2 shows that the significance level is 0.001 < 0.05. This indicates that ethics learning method affects ethical decisions. Subjects who get visual-based ethics learning methods have better ethical decisions than those who get nonvisual-based ethics learning methods. From the results of statistical test of hypothesis 2, the average group decision for the recipients of manipulation of visual-based ethics learning methods is 96.000, while the average group decision for the recipients of nonvisual-based ethics learning methods is 90,600. This shows that visual-based ethics learning methods affect one's cognitive. Cognitive theory or cognitive psychology explains how humans understand,  Source: Processed data learn, remember, and think about information (Sternberg, 2006). Presentation of information in a visual form is able to build one's cognitive more open, receive information better, process the information, and maintain it so that it pushes someone to reach ethical awareness, and ultimately take more ethical decisions. Thus it can be concluded that the second hypothesis is also supported. The results of this study are consistent with the statement of Tang et al. (2014) that decision makers who receive financial information presented in a visual form enable one to obtain better information and make more accurate decisions. Likewise with the ethics learning method, the ethical values conveyed in visual form help learners gain a better understanding of ethics so as to provide more ethical decisions.

Testing the Effect of Obedience Pressure and Learning Methods on Ethical Decision
Hypothesis 3 states that the subjects who are under high obedience pressure and get visual-based ethics learning methods will produce the best ethical decisions. Hypothesis 3 examines the effect of different conditions of observational obedience and various ethical learning methods on ethical decision.
The statistical test results of the relationship between obedience pressure, ethics learning methods, and ethical decisions support the third hypothesis with a significance value of 0.026 (<0.05). Table 8 presents the test results of hypothesis 3. In the relationship between obedience pressure and ethical decisions, the statistical test using Two Way ANOVA in Table 8 shows a significance value of 0.022 <0.05, which means that obedience pressure has a significant effect on ethical decisions. Subjects who are under obedience pressure provide ethical decisions that are different from subjects who are not under obedience pressure. This is consistent with the test results of Hypothesis 2 using the Independent t-test. Likewise with the relationship between ethics learning methods and ethical decisions, it is consistent with the results of the statistical test of hypothesis 2 that there is a significant effect of 0.000 < 0.05. Subjects who receive visual-based ethics learning methods have better ethical decisions than those who receive no visual-based ethics learning methods. Hypothesis 3 predicts that the subjects who are under conditions of high obedience pressure and get visual-based

CONCLUSION, IMPLICATION, SUG-GESTION, AND LIMITATIONS
This study provides empirical evidence that obedience pressure influences someone in making ethical decisions. The results of this research show that subjects who are not under obedience pressure make better ethical decisions than those who are under obedience pressure. The obedience pressure makes someone set aside personal values and produces unethical decisions. This negative pressure of obedience cannot be avoided by the actors of the organization, especially accountants, in carrying out their roles and profession. The impact of this obedience pressure can be mitigated through the education process using appropriate professional ethics learning methods.
The results of this study also provide empirical evidence that subjects who receive visual-based ethics learning methods have better ethical decisions than those who receive nonvisual-based ethics learning methods. In addition, the results of this study state that there is a relationship between obedience pressure, ethics learning methods, and ethical decisions. Visual-based ethics learning methods produce a better understanding of ethics for someone. Through a good understanding of ethics, a person is expected to be able to make more ethical decisions and ultimately produce ethical actions. Thus, visual-based ethics learning method can be one of the strategies to mitigate the impact of obedience pressure.

Limitation and Suggestion
This study has several limitations. First, the experiments were conducted in two internal audit classes at different times. The experiments which were conducted at the different time have the potential for information seepage. Initially the experiments would be carried out in only one class, but due to insufficient number of subjects, they were conducted in two internal audit classes. However, this threat was minimized with a lag that was not too long. The first wave experiment involving 59 participants was carried out at 10:00 while the second wave experiment involving 55 participants was carried out at 14.00. Secondly, this study was conducted by different experimenters in groups who received visualbased ethics learning methods and groups who received nonvisual-based ethics learning methods. The personality of experimenters, which is an innate factor, has the potential to influence the results of the experiment. To overcome this, the researchers provided guidance for the experimenters in carrying out the experiment so that the procedure carried out remained standard. Researchers did not act as experimenters to avoid the demand effects that could affect the subject's response. Future research could consider the design of internet-based experiments. Other research developments that need to be considered are ethical decision making by participants in group settings and discussions.

Implication
The results of this study provide empirical evidence that visual-based ethics learning methods make the subjects produce better ethics understanding. The results of this study have the implication that high obedience pressure from those who have authority makes someone produce bad ethical decisions. The results of this study contribute to the development of behavioral research in testing the theory of obedience. The results of this study support previous research conducted by Davis et al., (2006) and Baird & Zelin II (2009) that obedience pressure has a greater force because the pressure from the authorities is more effective than the pressure exerted by the social environment in same level.
These findings contribute to the accounting education institutions related to ethics learning methods especially the ethics of the accountant profession as a strategy to mitigate the obedience pressures that arises when prospective accountants enter the workforce. Visual-based ethics learning methods are able to build sensitivity development processes and improve ethics understanding better. A good understanding of ethics can help someone to make more ethical decisions so that in the end take ethical actions.