Internal Control, Organizational Culture, and Quality of Information Accounting to Prevent Fraud: Case Study From Indonesia's Agriculture Industry

This research wants to find out how far internal control, organization culture and the quality of accounting information system will help the small-medium enterprises (SMEs) to prevent fraud. by applying the case study approach in achieving its aims and objectives. This study is done by a used case study from SMEs in the agriculture industry in Lampung province, Indonesia. The data were collected through observations and semi-structured interviews with employed and managerial staff. This research applied a mixed method in collecting and analyzing data, which were document analyses and interviews. Applying more than a single method in collecting data enables the researcher to compare and to verify the information accuracy (Brewer and Hunter 2006). This method can increase the credibility and validity of the findings because the final bias will depend on one method which later can be avoided (Yin 2012). This type of research is quantitative descriptive research. The purpose of this descriptive research is to provide a descriptive, systematic, factual and accurate description of the facts, properties, and relationships between the phenomena investigated. All data that will be used in this study is sourced from the results of respondents' answers to the questionnaire given to employees at PT. XYZ as many as 70 respondents with the unit of analysis are part of Business Control, Human Capital, Finance, Marketing, and Operations. The sampling technique that uses saturated sampling, which is a sampling technique where all members of the population will be used as samples. The results of the study show that some weaknesses of the internal controls have been identified as one of the factors of fraud. The results show that Internal Control Organizational Culture and Quality of Information Accounting have a positive significant effect to prevent fraud.


Introduction
generally define fraud as any purposeful action that deceives, manipulates, or conceals facts to create a false impression). According to the ACFE (Association of Certified Fraud Examiners), fraud can be classified into three main blocks: fraud in financial statements, embezzlement or misappropriation of assets, and corruption, with fraud in the financial statements generating the highest costs, although it is less common than the misappropriation of assets (ACFE, 2014). Poor internal controls have been identified as one of the causes of fraud (Siregar and Tenoyo 2015;Zakari, Nawawi, and Salin 2016). A company must possess effective internal controls (KPMG 2004) to prevent fraud which can lead to a big loss. Internal controls can be described as policies or procedures regulated to convince that a certain purposed entity will be achieved. The main purpose of internal controls is to support the entity in administering the risks to achieve the purposed entity being built and to maintain the work ethics. Fraud is a form is intentionally done so it can cause losses unnoticed by the injured party and cause benefits for the perpetrators of fraud. Fraud generally occurs because of pressure to commit fraud or encouragement to take advantage of opportunities. A working system that is not transparent is an opportunity for perpetrators of fraud. Fraud by tricking the system is precisely possible because the culprit is an "insider" or involving people who have authority over the system. A non-transparent system closes the opportunity for many people to supervise and provide input on the current system. Research-related ISSN 1923-4023 E-ISSN 1923-4031 Chairun Nisak et al (2013, Sukadwilinda and Ratnawati (2013) concluded that internal control has a positive effect on fraud prevention. Internal control is applied in the organization to provide certainty in achieving organizational goals. Research also conducted by Zulkarnain (2013) shows that there is a negative influence between the effectiveness of the internal control system and fraud in the government sector. This means that the effectiveness of the employee's high internal control system in an agency can prevent fraud in the government sector. Mirinaviciene (2014) mentions strong internal control is a factor that affects fair financial reporting and fraud prevention and detection. Research by Dimitrijevic et al (2015) concluded that the main task of internal control is not to find the cause of fraud, but rather to detect and stop further expansion of fraud. An internal control system, if designed and functioning properly, will ensure that fraud will not occur. For this goal to be achieved, internal controls must cooperate with other controls such as external audit and forensic accounting in applying new methods and techniques for fraud prevention. the weaknesses of internal controls lead to fraud (Lokanan 2014; Skaife, Veenman, and Wangerin 2013), cause a tendency of increasing credit loss (Cho and Chung 2016), decrease the sales (Su, Zhao, and Zhou 2014), reduce future work ethics (Weiss 2014), improve the association which creates higher benefits of individual controls (Gong, Ke, and Yu 2013), reduce company's market values and the relevance of accounting information values (Hu et al. 2013), create low-quality of financial reports (Ghosh and Lee 2013), cause higher loans, loss of backups and provisions in the banking company (Cho and Chung 2016), and trigger negative.
Corporate culture represents the personality of an organization and its shared beliefs, values, behaviors, the way things are done and it's explicit and implicit rules (Bouwman, 2013). Organizational cultures are both visible and invisible, formal and informal (Weiss, 2014). Culture is often expressed informally through passing comments, gestures, and behaviors of especially top management. KPMG International (2017) also explains that corporate culture is not aspirational value ... but the unwritten rules that drive the thousands of decisions employees make (p.4). Culture can be an antidote to many organizational problems and can significantly affect how an organization is run every day throughout the company. (Warrick, 2017).
Research by Awadh and Saad (2013) states that organizational values and norms are based on employee relations. A positive relationship between culture and performance will help in increasing organizational results. Organizational work performance has a strong impact on a strong organizational culture because it leads to increased productivity. Organizational norms and values based on different cultures affect workforce management. A strong cultural organization enables effective and efficient employee management. Net profit in an organization helps in improving employee performance. Organizational culture is very helpful in enhancing and providing a competitive advantage. Employee commitment and group efficiency help in improving performance based on organizational sustainability. The nature and strength of organizational culture influence the organization's sustainability and effectiveness. In research by Surjandari dan Martaningtyas (2015) mentions that there is a negative correlation between work culture and fraud. Types of fraud committed by government officials are counterfeiting and/or stamping, taking office equipment, taking small money boxes, not recording transactions and taking money, recording incorrectly or inaccurately and manipulating the amount and price of materials purchased, making purchases that do not conform with specifications. And the most common types of fraud are Counterfeiting by Muhammad et al (2017) in his research concluded that the implementation of the accounting information system harms fraud on the BPR (Bank Perkreditan Rakyat) of Banda Aceh City. This means that the higher the application of its accounting information system, the more it will minimize fraud at BPR Kota Banda Aceh.

Internal Control
According to Arens et al (2015: 340), internal control is a policy and procedure that is designed to provide management with reasonable certainty that the company has achieved its goals and objectives, namely the reliability of financial reporting, efficiency and effectiveness of operations and compliance with laws and regulations. " According to the Indonesian Institute of Accountants (2015: 38), internal control is a process implemented to produce an adequate level of confidence so that the following control objectives can be met, namely asset protection, maintaining detailed records to report company assets accurately and fairly, providing accurate information accurate and reliable, compile financial reports following the criteria/standards required, support and improve operating efficiency, encourage compliance with established management policies, and comply with applicable laws and regulations.
1. Accurate, meaning that information must reflect the real situation, meaning that information must be free from unbiased or misleading mistakes. Accurately can be interpreted that the information can reflect its purpose.
2. On-time, information must be available when the information is needed. Information that comes to the recipient must not be too late. If the information is late, it will cause the decision to be late and it can be fatal for the company.
3. Relevant, meaning that the information provided must be by what is needed. The information submitted must be related to the problem to be discussed with the information and must also be useful for the wearer.
4. Complete, meaning that the information provided must be complete as a whole in the sense that there are no things that are reduced in conveying the information.

Fraud Prevention
There are 4 main pillars in combating fraud, namely fraud prevention, early fraud detection, fraud investigation and law enforcement or follow-up legal action. According to Arens (2014: 398), three conditions cause fraud called the fraud triangle, which are pressure, opportunity and attitude/rationalization. Based on the literature review above, the thought framework can be described as follows:

Hypothesis
Research on the effect of internal control on fraud prevention conducted by Hermiyeti (2010), Soleman (2013), Chairun Nisak et al (2013), Sukadwilinda and Ratnawati (2013) concluded that internal control has a positive effect on fraud prevention. Internal controls are applied in the organization to provide certainty in achieving organizational goals. Research also conducted by Zulkarnain (2013) shows that there is a negative influence between the effectiveness of the internal control system and fraud in the government sector. This means that the effectiveness of the employee's high internal control system in an agency can prevent fraud in the government sector. Mirinaviciene (2014) mentions strong internal control is a factor that affects fair financial reporting and fraud prevention and detection. In the journals, Hamdani and Albar (2016) and Alfian et al (2017) mention that weak internal control is one reason for fraud. The application of good internal control is expected to prevent fraud. Management can strengthen the role of internal control by adding to its employees, and there are not several important positions that are filled by the same employee. Albar and Fitri (2018), Akhsani (2018), Siregar and Hamdani (2018) in their research stated that the internal control system harmed fraud. This means that increasing the internal control system will reduce fraud. Through internal control, it can be seen whether the employees have done the job according to their duties and functions efficiently and effectively. Good internal control can increase employee compliance with applicable rules and reduce cheating practices, as in the fraud triangle theory, namely the opportunity for anyone to commit fraud. However, by reducing opportunities (opportunity) can reduce the tendency of cheating practices and can be easier to detect fraud early on.
Research on the influence of organizational culture on fraud conducted by Siregar et al (2018) and Albar and Fitri (2018) results in the conclusion that organizational culture harms fraud. This means that the higher the value of ethical culture that is applied to the work environment will affect the decline in the occurrence of fraud in the organization. However, the results of this research are not in line with research conducted by Akhsani (2018) that organizational culture does not affect the tendency of cheating practices. This means that the higher the ethics in organizational culture may not necessarily reduce the tendency to practice fraud, because fraud can be committed by anyone even by employees who initially had good norms and ethics. There is an opportunity (opportunity) to commit fraud that encourages someone's actions to do so. In the research of Surjandari and Martaningtyas (2015), there is a negative correlation between work culture and fraud. Types of fraud committed by government officials are counterfeiting and/or stamping, taking office equipment, taking small money boxes, not recording transactions and taking money, recording incorrectly or inaccurately and manipulating the amount and price of materials purchased, making purchases that do not conform with specifications. And the most common types of fraud are Counterfeiting and / or Cap. Muhammad et al (2017) in their research concluded that the implementation of the accounting information system harmed fraud on the BPR of Banda Aceh City. This means that the higher the application of its accounting information system, the more it will minimize fraud at BPR Kota Banda Aceh. While research Animah (2018) states that the application of accounting information systems has a positive effect on trends in accounting fraud. This is caused by the low competence of employees, resulting in a lack of knowledge and understanding of the use of accounting and bookkeeping systems. Due to the low level of knowledge and understanding of employees of the system, it will be difficult for employees to find loopholes to cheat even if there is an opportunity.
Based on the above, the hypothesis is taken as follows: H1: Internal Control has a positive effect on fraud prevention.
H2: Organizational Culture has a positive effect on fraud prevention.
H3: The Quality of Accounting Information Systems has a positive effect on fraud prevention.

Research Method
This type of research is quantitative descriptive research. The purpose of this descriptive research is to provide a descriptive, systematic, factual and accurate description of the facts, properties, and relationships between the phenomena investigated.
All data that will be used in this study is sourced from the results of respondents' answers to the questionnaire given to employees at PT. XYZ as many as 70 respondents with the unit of analysis are part of Business Control, Human Capital, Finance, Marketing, and Operations. The sampling technique that uses saturated sampling, which is a sampling technique where all members of the population will be used as samples. The research period is from May to September 2019.

Analysis of Statistical Descriptions
Before the data from the research questionnaire results are further analyzed, the validity and reliability tests of the research measuring instruments are first performed. A validity test is conducted to determine the ability of research instruments to measure what should be measured (Cooper et al, 2013: 257). So that the process of testing and processing of data can be done quickly and precisely, the data processing uses the help of SPSS version 25. The method is to compare the amount of r count with r table. If r count is greater than r table then the statement is declared valid. With the number of respondents (n) of 70 and α = 0.05 then using the distribution table r table values will get r tables of 0.235. The results of testing the validity of the internal control variables are presented in Table 1 below. be carried out on a reliability test that aims to see whether the questionnaire has consistency if the measuring instrument is used repeatedly. The condition is that the reliability coefficient > 0.7 then overall the statement is declared reliable. The reliability coefficient is seen from the Cronbach-alpha value.  (1)  The accounting information system quality variable has 3 (three) dimensions, namely the service system dimension consisting of 7 indicators, the quality system dimension consisting of 4 indicators and the information quality dimension consisting of 8 indicators. So the total number of indicators used to measure the variable quality of the accounting information system is 19 indicators, which are represented by 19 questions on the questionnaire.

Measurement Model (Outer Model)
Evaluation of the measurement model or outer model is carried out to assess the validity and reliability of the model. Validity testing aims to ensure that each item in the research instrument can measure the variables specified in the study. There are 2 validity tests, namely convergent validity, and discriminant validity.
The convergent validity of the measurement model with the reflective indicator is assessed based on the loading factor of each indicator forming latent constructs. A latent contract is considered to have a good convergent validity of the loading factor value is more than 0.7 and is significant. However, for the initial stages of bookkeeping scale research development, loading values of 0.5 to 0.6 are considered sufficient (Chin, 1998 in Ghozali andKusumadewi, 2016: 37). Invalid statements will be dropped from the table. The final results of the convergent validity test can be seen in the following table. Based on the table above, the indicator already shows the loading factor value> 0.50, this means the indicator is valid.
The way to test discriminant validity with reflective indicators is by comparing the square root of AVE for each construct with the correlation value between constructs in the model. Good discriminant validity is shown from the square root AVE for each construct greater than the correlation between constructs in the model (Fornell and Larcker in Ghozali, 2016: 21). For variables that only have 1 (one) or a single indicator in the construct, it can be said as a formative indicator so that it does not require a validity and reliability test. Evaluation of measurements is carried out with a significant weight (Ghozali, 2016: 24).
While reliability testing is done to prove the accuracy, consistency, and accuracy of the instrument in measuring the construct of the reflective indicator construct, it is declared to be reliable if it has a Cronbach's Alpha and D.G Rho value if the value is> 0.60. (Ghozali, 2016: 87). The results of discriminant validity testing are presented in Table 4 below.  Based on Tables 4, 5, 6 and 7 above it can be seen that the AVE value of each construct is greater than the square value of the correlation between constructs. Thus fulfilling good discriminant validity.

Structural Model (Inner Model)
The structural model or inner model aims to predict the relationship between latent variables that are hypothesized.
To assess the structural model, look at the value of R-Squares for each endogenous latent variable as an interpretation of OLS Regression (Ghozali, 2016: 25). Based on Table 10, obtained an R 2 of 0.721, this means that the fraud prevention construct is influenced by 72.1% by the internal control construct, organizational culture and the quality of the accounting information system while 27.9% is influenced by other constructs not contained in the study. To see the significance of the model can be seen from the t-statistic value compared to the t table value. With a confidence level of 95%, df = 67 we get at a table of 1.67. Ho will be rejected if the t-statistic value > 1.67 and Ho will be accepted if the t-statistic value < 1.67. For the direction of the influence of exogenous variables to endogenous variables can be seen from negative or positive signs on the t-statistic value. Based on table 12. the calculated value of the internal control variable is 1.983 and is positive. This value is above the t-table value of 1.67, meaning the hypothesis (Ho) which states that internal control does not affect fraud prevention is rejected and instead accepts the alternative hypothesis (Ha) which states that the internal control variable has a positive and significant effect on the fraud prevention variable.
This study is in line with research conducted by Hermiyeti (2010)

Organizational Culture Has a Positive Effect on Fraud Prevention
Based on Table 12, the calculated t value for the organizational culture variable is 1.982. This value is above the t-table value of 1.67, meaning the hypothesis (Ho) which states that organizational culture does not affect fraud prevention is rejected and instead accepts an alternative hypothesis (H2) which states that organizational culture variables have a positive effect on fraud prevention. This research is not in line with research conducted by Akhsani (2018) that organizational culture does not influence the tendency of cheating practices. This means that the higher the ethics in organizational culture may not necessarily reduce the tendency to practice fraud, because fraud can be committed by anyone even by employees who initially had good norms and ethics. There is an opportunity (opportunity) to commit fraud that encourages someone's actions to do so.
The results of this study are in line with research conducted by Siregar et al (2018) and Albar and Fitri (2018), Surjandari and Martaningtyas (2015) which concluded that the higher the organizational culture, the lower the fraud in the organization.

Quality of Accounting Information System Has a Positive Effect on Fraud Prevention
Based on Table 12, the calculated t value for the variable quality of the accounting information system is 4.29. This value is above the t-table value of 1.67, meaning the hypothesis (Ho) which states the quality of the accounting information system does not affect the prevention of fraud is rejected and instead accepts the alternative hypothesis (H1) which states that the quality of the accounting information system affects fraud prevention.
The results of this study are in line with research conducted by Muhammad et al (2017), Animah (2018).