Analysis of organizational commitment in determining the success of Accounting Information Systems (AIS) in the banking sector

* Corresponding author, email address: ita.salsalina@gmail.com ABSTRACT This study aims to examine the effect of organizational commitment on the success of accounting information systems (AIS). The unit of analysis in this study was Regional Development Banks (Indonesian: Bank Pembangunan Daerah / BPD) throughout Indonesia. Data collected by using a questionnaire. The respondents were BPD employees as the AIS users. The total research sample used was 275, while the questionnaire return rate was 64% (176). The data were processed and analyzed using the Partial Least Square SEM method. The results proved that organizational commitment has no effect on the success of accounting information system. In general, BPD employees’ commitment to the organization is at the “adequate” level, which means that they are not fully committed to supporting the success of the accounting information system. The results of this study do not support the theory and results of previous studies which state that organizational commitment affects the success of the accounting information systems.


INTRODUCTION
The world is currently entering the fourth industrial revolution, or industry 4.0. in this era, the information technology is developing so rapidly that every organization, from small to large scales, increasingly relies on accounting information systems in order to be able to compete and survive (Bodnar & Hopwood, 2010;O'Brien & Marakas, 2011;Stair & Reynolds, 2016). In other words, the accounting information system (AIS) is designed as a business strategy capable of supporting business processes and operations as well as decision-making processes. As the main system in an organization, the accounting information system aims to provide the information that users need (Romney & Steinbart, 2015). Therefore, it should be measured. This is in line with the statement put forward by Kieso et al. (2013) that the accounting information system is processing transaction data into financial information and distributing it to stakeholders. According to Fortune & Peters (2005), the criteria for measuring the success of a system include the ability to meet user needs, the ability to achieve goals, according to time plans and budgets, the ability to provide user satisfaction, and the ability to meet quality standards. Experts argue that the success of an accounting information system in a company is measured by its ability to achieve organizational goals or meet user needs (Bagranoff et al., 2010;Laudon & Laudon, 2014;Stair & Reynolds, 2016).
This research was conducted based on the information system success model developed by DeLone & McLean (2003), which is to assess the success of a system based on the characteristics of the system, including ease of use, flexibility, reliability, ease of learning, features of intuitiveness, sophistication, and response times. During its development, experts expressed various opinions regarding the characteristics of the success of a system. According to Boczko (2007), the characteristics of the success of a system include reliability, efficiency, integration, accessibility, flexibility, and accuracy. Heidmann (2008) argues that the characteristics of the success of a system include integration, flexibility, accessibility, formalization, and media richness. Baltzan & Phillips (2009) stated that the characteristics of the success of a system include flexibility, scalability, reliability, availability, and performance. In addition, according to Baltzan (2014), the characteristics of the success of a system include accessibility, availability, maintainability, portability, reliability, scalability, and usability. Thus, there are many characteristics of the succes of a system.
In the banking sector, the phenomenon of unsuccessful implementation of accounting information systems is shown by the unfulfilled characteristics of a successful system. The first problem, as pointed out by Budiwiyono (2015), is that the system is not yet integrated (integration). Budiwiyono, as the chairman of ASBANDA, stated that the Financial Services Authority of Indonesia (Indonesian: Otoritas Jasa Keungan or OJK) hoped that the BPD information system would be more integrated, but in reality only 15 out of 26 BPD banks joined the integrated information system (BPD Net Online).
The second problem is the unreliable system (reliability), for example, a fictitious credit case in the name of PT CT which was carried out by employees of Bank Sumsel Babel so that they succeeded in obtaining loan funds of up to IDR 480 billion through false recordkeeping (Adil, 2014). According to the OJK Banking Commissioner (Kristyana, 2015), the first phase of the implementation of the BPD transformation program is in the form of a foundation building phase which aims to build a strong support and capital process along with the quality of human resources, work culture, and reliable information systems through the synergy of the BPD group.
The third problem is an inflexible system (flexibility). For example, Panin Bank Syariah in collaboration with PT Emerio Indonesia applied the Regla RCS (Banking Regulatory & Compliance Suite) Software so that the accounting and reporting system could be precise and flexible according to the needs and character of the company (Sarasidya, 2014). Yusuf (2016) stated that the Vice Governor of East Java even encouraged banks to be more flexible in providing services to the community, especially rural communities, by collaborating the system for a wider reach. The Governor of DKI Jakarta (Purnama, 2016) also expressed a similar opinion that the DKI Bank's information system is considered old or outdated (inflexible).
The fourth problem is a system that is not easy to use (usability). For example, BPD Kaltim launched the Generation 2 of State Revenue Module (MPN-G2) system, or commonly referred to as online tax deposit receipts, in order to make it easier for taxpayers to pay their tax obligations electronically. With the launch of the new system, deposits can be made more practically, quickly, and safely via internet banking, mobile banking, ATMs, or Electronic Data Capture (EDC) machines without the need to queue at teller counters (Hanafiah, 2015). Another example is the virtual account system used by DKI Bank, in which according to the assessment of the Governor of DKI Jakarta (Purnama, 2016), the system is still very difficult for the citizens of Jakarta to use in conducting financial transactions. Therefore, Bank DKI is expected to implement an easy-touse information system.
One of the factors that influence the success of accounting information systems is organizational commitment (O'Brien & Marakas, 2011;Romney & Steinbart, 2015).
A according to O'Brien & Marakas (2011), commitment and involvement of end-users, from lower lever management to the top,is a fundamental requirement for the success of an information system. This is confirmed by Romney & Steinbart's statement that the implementation of a system will fail without the support of all the organization's human resources.
Organizational commitment is basically defined as a strong desire to remain a member of an organization, a willingness to do quality efforts for the benefit of the organization, as well as belief and acceptance of organizational values and goals (Luthans, 2011;Hellriegel & Slocum, 2011;Robbins & Judge. , 2014). Furthermore, Gibson et al. (2012) explain that organizational commitment includes three attitudes: (1) having sense of identifying organizational goals, (2) having willingness to participate in company tasks, and (3) having loyalty to the organization. Basically, organizational commitment is measured in three dimensions: affective commitment, continuance commitment, and normative commitment (Luthans, 2011;Greenberg & Baron, 2011).
Based on the results of a survey conducted by Towers Watson, a global HR consulting company (Awaldi, 2014), most companies in Indonesia fail to understand the importance of employee involvement in the company's business development which results in a lack of employee loyalty. On average (66%) of the employees surveyed were only able to stay in a company for two years, while the rest chose to stay (Awaldi, 2014). According to Tampubolon (2016), the lack of commitment of BPD human resources is considered the cause of not achieving the objectives of the BPD Net Online system. In other words, organizational commitment is considered a factor that determines the success of the accounting information system.

THEORITICAL FRAMEWORK AND HYPOTHESIS Organizational Commitment
Experts define organizational commitment as individual loyalty to the organization (Schemerhorn, 2010), the level of loyalty or involvement of individuals to the organization (Greenberg & Baron, 2011), and the desire of employees to remain members of an organization (Robbins & Judge, 2014;Griffin & Moorhead, 2014). Thus, organizational commitment is related to an attitude in the form of a person's trust/ acceptance of the values and goals of the organization so as to encourage the person to be loyal and participate/ contribute significantly by doing his or her best for the organization so that organizational goals can be achieved (Kondalkar, 2007;Jex & Britt, 2008;Hellriegel & Slocum, 2011;Luthans, 2011;Gibson et al., 2012).
Organizational commitment basically consists of affective commitment, continuance commitment, and normative commitment, or better known as the Three-Component Model of Commitment developed by Meyer & Allen (1997). Experts define affective commitment as one's loyalty to the organization because of emotional attachment and one's desire to provide the best for the organization because of emotional attachment. Continuance commitment is defined as one's loyalty to the organization due to rational/profit-loss or one's desire to provide the best for the organization due to rational/profit-loss considerations. Normative commitment is defined as one's loyalty to the organization due to moral/ethical considerations (Jex & Britt, 2008;Luthans, 2011;Greenberg & Baron, 2011;George & Jones, 2012;Robbins & Judge, 2014).

The Success of Accounting Information Systems
An accounting information system is defined as a subsystem of an information system that aims to collect, process, and report financial information from business transactions (Gelinas & Dull, 2008). The definition of an accounting information system then develops into a collection of data and processing procedures that produce information for users (Bagranoff et al., 2010). Furthermore, Bodnar & Hopwood (2010), state that an accounting information system is a collection of human resources and equipment designed to convert financial data and other data into information used by decision makers. The same thing was also stated by Wild et al. (2011), that the accounting information system collects and processes data from transactions and events, organizes them into useful reports, and then communicates them to decision makers. This understanding is confirmed by the statement of Romney & Steinbart (2015) that accounting formation system is the process of collecting and processing business transaction data into a report that is useful for decision makers.
The success of a system, according to Leitch & Davis (1992), is assessed by the existence of the integration of sub-systems/components. Meanwhile, Romney & Steinbart (2015) argue that the most important component for the success of an accounting information system is the objective of the system. The objective of an accounting information system is to facilitate the procedures of data collection, data maintenance, data management, data control, and presentation of information (Boczko, 2007). Experts argue that the success of a system can be measured by the ability to meet user needs, the ability to achieve goals, the ability to meet user satisfaction, and the ability to meet quality standards (Fortune & Peters, 2005;Stair & Reynolds, 2016).
This study was conducted using a system success model developed by DeLo-ne & McLean (2003) which assesses the success of a system based on the characteristics of the system. The success of the accounting information system in this study is measured based on integration, reliability, flexibility, and usability.
Integration is assessed based on the existence of integration between sub-systems and integration between systems (Wu, 1983;Nash & Roberts, 1984;Leitch & Davis, 1992;Whitten & Bentley, 2007;Boczko, 2007;Heidmann, 2008;Baltzan& Phillips, 2009;Bojic. &Hickie, 2015;Stair & Reynolds, 2016). Reliability of a system is assessed from the ability of the system to function correctly starting from data input, processing, to producing accounting information (output) and the ability of the system to produce accurate accounting information. Flexibility is assessed from the ability of the system to adapt to changing conditions/environment and the ability of the system to adapt to changing needs or business (Nash & Roberts, 1984;Avgerou&Conrnford, 1998;DeLone& McLean, 2003;Boczko, 2007;Baltzan& Phillips, 2009;Bagranoff et al., 2010;Baltzan, 2014). Furthermore, usability of a system is assessed from its ease of use and ease of learning (Avgerou & Conrnford, 1998;Davis et al., 1990;Laudon & Laudon, 2014;Baltzan, 2014).

The Effect of Organizational Commitment on the Success of the Accounting Information System
The success of the implementation of an accounting information system is determined by a full commitment (Yeates & Wakefield, 2004). This is confirmed by the statement of Stair & Reynolds (2016) that the success of the accounting information system is determined by the commitment of the entire team from the lowest to the highest levels of management as well as the commitment of the organization itself. Lack of participation/involvement of AIS users is the main cause of failure of the implementation of the system (Clarke, 2001;Whitten & Bentley, 2007;Laudon & Laudon, 2014).
The results of previous research prove that organizational commitment affects the success of the accounting information system (Basu et al., 2002;Syaifullah, 2014;Nurhayati, 2014;Fitriati & Mulyani, 2015;Indahwati, 2015). Based on the theory that is supported by the results of previous research, the framework that can be put forward in this study is shown in Figure 1.
Based on the framework outlined in Figure , the hypothesis proposed is: H1: Organizational commitment has an effect on the success of the accounting information system.

RESEARCH METHOD
This research is a descriptive study conducted using a survey method, that is, the fact-finding process with the aim of obtaining a description of the phenomenon that occurs (Sekaran & Bougie, 2013). In addition, this research is an explanatory study which is intended to explain causal relationships and hypothesis testing (Effendi & Tukiran, 2012), which means that this study aims to explain the effect of organizational commitment variable on the success of the accounting information system.

Organizational Commitment
The succes of the accounting information system

Figure 1 Framework
The unit of analysis in this study was BPD banks across all provinces in Indonesia, while the observation unit was the accounting and operational divisions of each BPD bank. Furthermore, the distribution of research questionnaires was addressed to BPD employees, such as the division head and staff as users of the accounting information system (AIS).
According to Hair et al. (2014), if a study aims to confirm the theory, the data analysis is conducted using the Covarian-Based SEM (CB SEM) method. However, if the CB SEM assumptions cannot be fulfilled, the best alternative in analyzing data is using Partial Least Square SEM (PLS SEM) for testing the theory.

DATA ANALYSIS AND DISCUSSION
The questionnaires were distributed to 27 BPD banks, including head offices, branch offices and sub-branch offices. The research respondents were division heads and accounting and operational staff. The total number of questionnaires distributed was 275, while the rate of return of questionnaires from 24 BPD banks was 64% (176). This shows that the minimum requirement of 30% has been fulfilled (Sekaran & Bougie, 2013). The same thing is also stated by Cooper & Schindler (2014) that the return rate of the questionnaire is in the "good" category if it exceeds 30%.
The research instrument testing includes validity testing and reliability testing. Validity testing was conducted using product moment correlation, while reliability testing was conducted using the alpha-Cronbach method. The statement item is declared valid if the correlation coefficient value is ≥ 0.30, while the statement item is declared reliable if the reliability coefficient value is ≥ 0.70 (Kaplan & Saccuzzo, 2005). The results of data processing are in Table 1.
Prior to data analysis, the average response score was categorized in a formula where the maximum and minimum score intervals were divided by the number of desired categories:1) the score interval between 1.00-1.99 is in the "bad" category;2) the score interval between 2.00-2.99 is in the "poor" category;3) the score interval between 3.00-3.99 is in the "adequate" category; and4) the score interval between 4.00-5.00 is in the "good" category. Table 2, below presents information on the results of the recapitulation of the average score of respondents' responses for each dimension of organizational commitment.
Based on the data in Table 2, the responses of research respondents related to organizational commitment in terms of affective commitment, continuance commitment, and normative commitment have an average score of 3.70,or in the "adequate" category. In addition, the percentage of the real score compared to the total is 74.03%, indicating a gap of 25.97%. This means that BPD employees are not yet fully committed to the organization, especially affective commitment, a commitment driven Based on data processing, the recapitulation results of the average score of respondents' responses for each dimension of the success of the accounting information system are stated in Table 3.
Referring to the data in Table 3, the responses of research respondents were related to the success of the accounting information system, in terms of integration such as reliability, flexibility, and usability. They have an average score of 3.74, or in the "adequate" category. In addition, the percentage of the real score compared to the total is 74.90%, indicating a gap of 25.10%. This means that in general the success of the accounting information system used by BPD in Indonesia is in the "adequate" category.
In this study, the validity and reliability of each indicator for each dimension that forms the organizational commitment variable were tested using the measurement model of the first order confirmatory factor analysis (CFA) and the second order confirmatory factor analysis (CFA). The results of validity and reliability testing using the first order CFA are described in Table 4.
Referring to the data in Table 4, the loading factor value for each indicator is> 0.50. This shows that all indicators are valid as measurement tools. The composite reliability (CR) value for each dimension is > 0.70, meaning that the indicators have consistency in measuring each dimension. Likewise, the average variance extracted (AVE) value for the dimension of affective commitment is 0.708, meaning that on average 70.8% of the information contained in each indicator is reflected through the dimension of affective commitment. The AVE value for the dimension of continuance commitment is 0.825, meaning  Total 100 Source: Data processing results that on average 82.5% of the information contained in each indicator is reflected through the dimension of continuance commitment. Likewise, the AVE value for the dimension of normative commitment is 0.952, meaning that 95.2% of the information contained in each indicator is reflected through the dimension of normative commitment.
The results of validity and reliability testing using the second order CFA are described in Table 5.
Referring to the data in Table 5, the loading factor value for each dimension is> 0.50. This shows that all dimensions are valid as measuring tools. The composite reliability (CR) value of all dimensions is 0.924, or > 0.70, meaning that the three dimensions have consistency in measuring the latent variable of organizational commitment. Likewise, the average variance extracted (AVE) value is 0.804, meaning that on average 80.4% of the information contained in each dimension can be reflected through the latent variable of organizational commitment. In addition, in Table 5, it can be seen that the largest estimated coefficient of determination (R2) of the latent variable of organizational commitment is expressed by the dimension of continuance commitment (0.928). These data indicate that the dimension of continuance commitment is the strongest dimension in reflecting the latent variable of organizational commitment, whereas the dimension of affective commitment is the weakest dimension in reflecting the latent variable of organizational commitment (0.601).
Organizational commitment is measured using 3 dimensions which are operationalized into 6 indicators. Based on the results of data processing using the second order CFA, it is obtained a path diagram of measurement model for the latent variable of organizational commitment as presented in Figure 2.
The results of testing the hypothesis on the effect of organizational commitment on the success of the accounting information system are presented in Table 6.
Referring to the data in Table 6, the t count value of the organizational commitment variable is 0.386 <t critical of 1.96, at the error level of 5%.So, Ho is accepted. Thus organizational commitment has no effect on the success of the accounting information system.
Based on the summary of the respondents' responses, it is found that most BPD banks in  Indonesia (62.5%) use accounting information system applications developed by parties outside the BPD where the authority related to the implementation and development of the accounting information system lies with the management of the head office. This has resulted in a lack of active participation from BPD employees, especially in branch offices. Organizations, in this case BPD banks, need to actively involve employees in the development of accounting information system (AIS) by accommodating input or suggestions from employees as AIS users. The survey results show that BPD employees in Indonesia, as AIS users, have not fully participated in the development of accounting information systems. In the end, this has an impact on the successful implementation of accounting information systems.

CONCLUSION, IMPLICATION, SUG-GESTION AND LIMITATION
Based on the results of data processing, it can be concluded that organizational commitment has no effect on the success of the accounting information system. The results of this study do not support the theory and the results of previous research because the implementation of the accounting information system at BPD banks in Indonesia is not fully supported by the high organizational commitment.
According to the survey results, the employees of BPD banks in Indonesia generally have a sufficient desire to do their best for the organization because of the encouragement of care and sufficient desire to do the best for the organization through participation (affective commitment). The results of the recapitulation of the average score of respondents' responses indicate a gap of 25.97%, which means that BPD employees have not fully committed themselves to the organization. This is because the organizational commitment is distributed into affective commitment (commitment driven by concern for emotional attachments), continuance commitment (commitment driven by economic considerations), and normative commitment (commitment driven by ethical considerations). The absence of high affective commitment from BPD employees has implications for the success of the accounting information system used. In turn, it has implications for the successful implementation of the accounting information system at each BPD bank.
It is necessary that BPD seek the active participation of BPD employees in the development of the accounting information system by accommodating input or suggestions from employees as AIS users. Therefore, it is expected that there will be an increase in affective commitment from employees that is driven by concern for the organization.  This study has limitations both in the number of research samples and differences in the competence of employees of BPD banks across all provinces in Indonesia in terms of educational background and expertise. These differences have an impact on the objectivity of the respondents in answering the questionnaire. Further research is expected to expand the types of banks and research samples.