The International Journal of Accounting and Business Society Vol. 29, No. 2 August 2021 © Centre for Indonesian Accounting and Management Research Postgraduate Program, Brawijaya University 119 ANALYSIS OF ACCEPTANCE OF ACCOUNTING INFORMATION SYSTEM IMPLEMENTATION BASED ON ELECTRONIC PAYMENT USING THE UTAUT MODEL

Purpose — This study aims to examine and analyze the effect of performance expectancy, effort expectancy, social influences, and facilitating condition on the actual use of electronic payment-based accounting information systems. Design/methodology/approach — Data obtained through online questionnaires filled out by 247 active users of electronic payment systems. The analysis method used is a Structural Equation Modeling/SEM with a Partial Least Square/PLS test tool. Findings — Empirical evidence shows that performance expectancy, effort expectancy, social influences, and facilitating condition significantly affect electronic payment systems' actual use. Practical implications — Electronic payment platforms are recommended to provide an electronic payment application system that can be used easily and usefully. The platform can use these results as a reference regarding customer behavior in using an electronic payment-based accounting record system Originality/value — This study uses UTAUT which is modeled directly on actual use.


INTRODUCTION
Accounting as an information medium must be able to provide reliable and accurate information to meet the needs of internal and external parties. Accounting produces not only financial information, but also operational information and management information (Wahyuni & Hartono, 4: 2019). Therefore, the Accounting Information System (AIS) is essential to be implemented in an entity because of the information it produces. AIS is a structured unification within an entity, including resources and other components that can transform been made according to applicable regulations (Albert, 2018) and lost balances at certain hours due to maintenance (CNN Indonesia, 2018).
These cases show the weaknesses of the electronic payment system that can potentially get negative behavioral responses from customers. Negative behavioral responses are seen when individual behavior tends to reject or refuse to use the information system. Implementing a system cannot develop optimally if humans as users, are unable or reluctant to use the technology, especially when the system is a product of an entity. An entity can experience significant losses if the system product that has been designed does not get a positive response from users. If this is allowed to do so, it is feared that it will have an impact on business stability in the long term. Therefore, it is important to understand the factors that can lead to the acceptance of individual behavior towards information technology through behavioral theory.
UTAUT is one of the acceptance theories developed by Venkatesh, Morris, Davis, & Davis (2003). UTAUT is considered a more perfect model because it can explain the system acceptance factors by 70%. UTAUT have been tested in various countries with different sample and object characteristics, including in Saudi Arabia with the object of e-government (Mansoori, Sarabdeen, & Tchantchane, 2018), conducted in India with the object of mobile payment (Madan & Yadav, 2016), carried out in the UK with the object of mobile payment (Slade, Dwivedi, Piercy, & Williams, 2015), conducted on business school students with the ERP object (Chauhan & Jaiswal, 2016), and carried out in Indonesia with the object of ebudgeting (Andriyanto, Baridwan, & Subekti, 2019a). UTAUT explains that acceptance or use behavior is initially shown by interest behavioral, but this study examines acceptance or usage behavior without being preceded by interest. Empirical studies show that usage behavior is not entirely preceded by interest, including Andriyanto et al. (2019) and Singh, Srivastava, & Sinha (2017). In addition, usage behavior is considered better in explaining the acceptance of an information system better because usage behavior is real use, while interest is a tendency of yes and no to indicate usage.

LITERATURE REVIEW AND HYPOTHESIS DEVELOPMENT UTAUT
The theory of the unification of acceptance and use of technology or the Unified Theory of Acceptance and Use of Technology (UTAUT) is a unification of eight acceptance theories that have been developed previously. The eight theories, including Theory of Reasoned Action (TRA), Technology Acceptance Model (TAM), Theory Figure 1 Venkatesh et al. (2003) UTAUT Model This study refers to the UTAUT model of Venkatesh et al. (2003) by making several adjustments, including 1) This study did not include moderation and intention variables. As previously explained, interest is still a tendency to do something so that real usage behavior is considered better in explaining the acceptance of an information system's implementation. This is supported by the statement of Alshehri, Drew, Alhussain, & Alghamdi (2012) that real usage can directly lead to more accurate conclusions about the acceptance of information system implementation and research results Andriyanto et al. (2019) prove that most UTAUT constructs positive and significant affects towards usage behavior; 2) the study also does not include moderating variables of gender, age, experience, and the voluntariness of use in research so that we examine the construct primary UTAUT the performance expectancy, effort expectancy, social influence, and facilitating condition toward usage (usage behavior/usage).
Empirical studies show that gender does not strengthen most UTAUT constructs in the use of online shopping applications and village government accounting systems (Andriyanto, Baridwan, & Subekti, 2019b;Lian & Yen, 2014). It can be concluded that there is no difference in the use of the system between men and women based on the UTAUT acceptance factors, or it can be interpreted that there is emancipation or gender equality in system acceptance. Age is also proven not to strengthen the main factors of UTAUT (Yuliasari, 2014). This means that there is no difference in the use of information systems based on age, especially since this study uses a system that is voluntary in its use. Therefore, it is believed that young and old individuals will have the same perception regarding the acceptance of the system. The experience variable was not included as a moderating variable because of the different research methods carried out. Venkatesh et al. (2003) conducted longitudinal research at three specific times, namely, immediately after training, a month after implementation, and three months after implementation so that there would be experiences during that time span, while this study was conducted using a one-time survey method with the context of the organization's external users, so there is no difference in the time span. Voluntariness of use variable is not used in this study because the electronic payment system is technology-based information that is voluntary in its use, as well as the voluntariness of use construct, has also been removed in the UTAUT2 research of Venkatesh, Thong, & Xu (2012). After all, it only affects one construct, namely social influence, so that although it is removed, it will not reduce the interaction effect of social influence and other moderating constructs (Morris, Venkatesh, & Ackerman, 2005). The detailed research design used is shown in the figure below. ways to save time and money and can improve the quality of government services and increase equality between all citizens (Alshehri et al., 2012). The research results by Andriyanto et al. (2019a) in Indonesia revealed that the use of the e-village budgeting system could provide utilitarian benefits, such as faster task completion and being able to increase productivity work, so that village officials use it.
Performance expectancy in the use of electronic payment systems is related to consumer perceptions that using electronic payment systems will help make profits and improve job performance compared to transactions using cash. The advantages of using an electronic payment system include the opportunity to get promos/discounts and the availability of a transaction history feature. Getting promo's/discounts can save expenses. In addition, the transaction history feature also makes it easy to record expenses so you can track and control finances. The advantage of improving job performance is that the transaction process is easy, flexible, and fast by simply scanning with a cellphone to save transaction time. Another advantage is that you can pay the nominal you should without worrying about not getting change and can make transactions anywhere and anytime. These advantages can form a perception that using an electronic payment system can get benefits and benefits and improve performance so that consumers can gain confidence and use the electronic payment system to the fullest.
Based on the explanation and research results, it can be concluded that performance expectancy has been tested to predict the use of electronic payment systems. Therefore, it is concluded that the alternative hypothesis is as follows.
H1: Performance expectancy has a positive effect on the actual use of electronic payment systems.

The Effect of Effort Expectancy on Usage
Effort Expectancy (HU) is related to the level of ease of use associated with operating the system. Empirical evidence shows that the level of ease of operation of the cloud computing system will be a consideration for the adoption of MSMEs in Bangladesh; the reason is that the cloud computing system is a new thing for MSMEs in Bangladesh (Khayer et al., 2020). The results of empirical research in the context of an online social network system (Online Social Network) indicate that government employees in the UAE consider the online network system to be easy to use and flexible when they want to communicate long-distance to encourage users to continue to use it . Research by Alshehri et al. (2012) proves that people in Riyadh, Jeddah, and Abha consider easy to learn and easy to use e-government systems, and become more skilled when using the system so that people in Riyadh, Jeddah, and Abha accept the application of e-government systems by using it. The results of research on the application of the system in Indonesia in the context of government also show that the application of the e-village budgeting system is something new in the Banyuwangi government so that the ease of use of the system is essential for the realization of acceptance of the information system (Andriyanto et al., 2019a). Ease of use can encourage users to understand and be skilled in using it more quickly to produce the perception that using the e-village budgeting system does not require high effort.

Vol. 29, No. 2 August 2021 © Centre for Indonesian Accounting and Management Research Postgraduate Program, Brawijaya University
Ease of operation of the system is critical, especially in implementing new systems such as electronic payments. Ease of operation, such as an attractive user interface view, easy to remember, and not complex operation command dialogs, will help form a positive user perception (Resky, Said, & Labbase, 2018). Ease of use will make individuals more quickly understand and master the use of the system so they can use the electronic payment system without excessive effort.
The explanation and empirical results show that effort expectancy, which is one of the constructs in the UTAUT model, has been tested and is a predictor of the use of electronic payment information systems. Therefore, it is concluded that the alternative hypothesis is as follows.
H2: Effort expectancy has a positive effect on the actual use of electronic payment systems.

The Effect of Social Influence on Usage
Social Influence (PS) is an individual's perception of what he believes comes from close people such as family, friends, and social relationships (social networks), which will affect interest in using certain information systems (Venkatesh et al., 2003). Previous research results prove that the online network system is widely recommended or used by supervisors so that colleagues, such as employees, use it too . Social influence is proven to have a significant effect in increasing the use of the mobile payment system (Kar, 2020) the reason is that mobile payment is a new payment method, so that the social environment provides additional confidence as an effort to increase user trust. When the user trusts, the user will be more confident in using mobile payment as a means of payment. This is supported by the statement of Pal, Herath, De, & Rao (2018), which explains that the adoption of mobile payments is greatly influenced by other consumers because the system has systemic and sensitive risks so that the recommendations of others will make users and potential users more confident and trusting in using it.
Social influence is expected to form positive perceptions so that consumers use electronic payment systems. The social environment's role in shaping perceptions includes the perception that using an electronic payment system can increase the degree of prestige and impress others because it is something new. In addition, the social environment also plays a role in finding references to experiences of third parties (other users) that can increase usage confidence.
Based on the explanation and empirical results, it shows that social influence is one of the constructs in the UTAUT model that has been tested and is a predictor of behavior in the use of information systems. Therefore, it is concluded that the alternative hypothesis is as follows.
H3: Social influence has a positive effect on the actual use of electronic payment systems.

The Effect of Facilitating Condition on Usage
Facilitating Condition is the individual's perception of the availability of resources and support available to perform the behavior (Venkatesh et al., 2003). Empirical evidence shows that when organizations provide resources, such as hardware and software, training or use guidance to support the implementation of online network systems, the greater use of online network systems is because individuals already have the knowledge and skills needed to operate the system . Knowledge of use, ease of access, and technical support in the use of e-government systems have proven to be decisive in using these systems in the communities of Riyadh, Jeddah, and Abha (Alshehri et al., 2012). The research results in Indonesia also show that the availability of supporting facilities and the willingness of technical assistance will encourage village officials' behavior to use the e-village budgeting system to the maximum (Andriyanto et al., 2019a). The available supporting facilities, such as hardware, software, and internet networks, are needed by village officials when making village usage reports, realization reports, and accountability reports through the e-village budgeting system. The availability of technical assistance when there is trouble or error also supports the implementation of the system so that village officials can make maximum use of the system (Andriyanto et al., 2019a).

Vol. 29, No. 2 August 2021 © Centre for Indonesian Accounting and Management Research Postgraduate Program, Brawijaya University
Acceptance of electronic payment systems to customers, namely when consumers have technical and non-technical support facilities. Technical support facilities for electronic payment systems, such as smartphones, internet, money, and service compatibility to facilitate access to financial services. Meanwhile, non-technical support facilities such as knowledge of use, regulations regarding transaction security guarantees and consumer protection, and getting help from other people when having difficulties in using the system. When these two facilities are available, it can increase the use of electronic payment systems.
Based on the explanation and empirical results, it shows that facilitating condition is one of the constructs in the UTAUT model that has been tested and is a predictor of behavior in the use of information systems. Therefore, it is concluded that the alternative hypothesis is as follows.
H4: Facilitating condition has a positive effect on the actual use of electronic payment systems.

RESEARCH METHODS Population and Sample
This research is quantitative research with a positive paradigm. The research was conducted in Malang City. The selection of Malang City as a research location is because Malang City has 73% of the productive age population (Badan Pusat Statistik, 2018). Productive age can represent young people who are familiar with the use of technology, such as computers, cell phones, and other technology (Tarhini, El-Masri, Ali, & Serrano, 2016). The population used in this study were all users of the OVO server-based electronic payment system in Malang. The sample and unit of research analysis used were individuals who had transacted using OVO at least once. This is because the data obtained is more representative, namely active OVO users. OVO was chosen as the research object because OVO has added value that makes users interested in using it. This can be seen from the achievements obtained by OVO in 2018 to 2019, including an increase in user base by 400% as of November 2018 (Djumena, 2018), shifting Gopay's position as electronic payment that has the most users (Muslim, 2019), and get a Google Play app rating 0.5 higher than the BUMN (State-owned Enterprise)'s LinkAja payment system application with the same number of downloads.
Sampling was carried out using a non-probability method with a convenience sampling method. The method for determining the minimum sample size uses the PLS sampling technique, which is 10 times the number of causality relationship pathways between latent variables (Hartono & Abdillah, 2016). There are four lines of causality in this study, so that the minimum sample that must be accepted is 4 x 10 = 40 individuals who use OVO electronic payments in Malang City.

Data Collection Methods, Research Instruments, and Measurement
The research data collection was carried out through a questionnaire survey with respondents according to the previously mentioned sample and unit of analysis. The questionnaire was distributed online through (1) social media, such as Twitter, Instagram (IG), and Facebook (FB); (2) message broadcasting Whatapps (WA); and (3) asking for help from real friends of the poor person to post questionnaires on their respective IG / WA accounts. The rate of return obtained was 248 data, but after screening data, there was 0.4% (1 data) of data that did not match the criteria, so that the data that could be processed was 99.6% (247 data).
The research instrument used references to previous research relevant to the criteria, namely using the UTAUT model and using similar objects. Then, the statement from the original reference was translated into Indonesian by an English language expert. After that, a sentence or syllable reconciliation is carried out between the original statement and the translation statement so that it is not mistaken. Furthermore, the statement sentence is adjusted to the object under study. The research instrument is shown in detail in table 1. All variables used are measured with a Likert scale 1-7 with the criteria for point 1 for Strongly Disagree (STS) answers, point 2 for Disagree (TS) answers, point 3 for Slightly Disagree (ATS) answers. 4 points for Neutral (N) answers, 5 points for Slightly Agree (US) answers, 6 for Agree (S) answers, and 7 for Strongly Agree (SS) answers. The seven-point Likert scale was chosen because it has several advantages, including more accurate and more diverse answers because there are many choices to show the true response value.

Data Analysis Performed
The researchers conducted a pre-test on the statements (indicators) used in real research for all variables. The aim is that the research statements are valid and reliable or can be interpreted as fit for the research object used, and the language used can be understood. The data analysis technique used is Structural Equation Modeling regression with the SmartPLS version 3 test tool.
The first stage of the test is evaluating or testing the measurement model (outer model), used to test validity and reliability. The validity test consists of two tests, namely the convergent validity test and the discriminant validity test. The standard values used for the convergent validity test and discriminant validity refer to the standards mentioned by Hartono & Abdillah (2016), namely the convergent validity test standard with an outer loading factor more than 0.7 and an AVE value more than 0.5. Meanwhile, the standard discriminant validity test used the root AVE value greater than the latent variable correlation, and the cross-loading value is more than 0.7 in one variable. The second stage of the test is testing its reliability with a standard test using Cronbach's alpha value with the rule of thumbs having to be > 0.6 and the composite reliability value with the rule of thumbs having to be > 0.7. Then the third stage is the stage of testing the structural model (inner model). Structural models were evaluated using the value of R 2 to construct and use a coefficient dependent on the path (β) for the independent construct. Then, the significance was assessed based on the T-statistic value for each path. This study used a one-tailed hypothesis and used a confidence level of 95% (alpha of 5%); therefore, when the T-statistic value that appears in the results of the analysis tool is higher than the T-table (1.64), then the alternative hypothesis is accepted, and vice versa.  Table 2 shows the composition of respondents involved in the study based on age group, gender, education level, income level, occupation, place of residence in Malang City, the time of electronic payments usage, and frequency of using electronic payment application systems per month. Respondents are OVO users who have made at least one transaction. This is done in order to represent active users of electronic payments. Based on the information in table 1, it is known that the study respondents were predominantly young (68%) and female (59%). The majority of respondents have an Undergraduate/D4 education (59%) and have middle to lower-income (31%) according to established criteria (Indrianawati & Soesatyo, 2015). The majority of respondents work as private/non-civil employees (43%) and reside in the Lowokwaru sub-district (32%). The majority of respondents have used the OVO electronic payment application system for 1 to 2 years (60%) with a frequency of use of 3 to 5 times per month (36%).  Table 3 Information shows that respondents tend to agree that the electronic payment system is useful (HK), easy to use (HU), requires the role of the social environment (PS), requires the support of resources, facilities, and knowledge (KM) for access to its use. Respondents also agreed to use an electronic payment system (PA). In addition, all constructs obtain a standard deviation value greater than the average value, so it can be concluded that there are no outliers data for all constructs   Table 5 show that most of the indicators measuring the variable performance expectations, business expectations, social influence, facilitating conditions, and actual use resulted in a loading factor greater than 0.7 compared to

Vol. 29, No. 2 August 2021 © Centre for Indonesian Accounting and Management Research Postgraduate Program, Brawijaya University
cross-loading on other variables. The KM4 indicator has a value below 0.7, but the KM4 indicator is not removed because the loading factor value is still higher than the loading factor value of other variables. Thus all measuring indicators can be declared valid or have met the standard of discriminant validity testing. All indicators also obtain AVE Root values greater than the latent variable correlation so that it can be stated that the indicators of all constructs are valid.  Table 7 Information shows that all variables have a Cronbach's Alpha value of more than 0.6 and a Composite Reliability value of more than 0.7, so it can be concluded that all indicators measuring the variables are declared reliable.    Table 9 shows that H1 obtained a regression coefficient value of 0.263 and a p-value of less than 0.05. This means that performance expectancy has a positive and significant effect on the use of electronic payment systems (H1 is supported/accepted). H2 shows a regression coefficient of 0.153 and a p-value of less than 0.05. It can be concluded that effort expectancy has a positive and significant effect on the use of the electronic payment system (H2 is supported/accepted). The H3 test results show that the regression coefficient is 0.288 and get a p-value less than 0.05. This means that social influence has a positive and significant effect on the use of electronic payment systems (H3 is supported/accepted). The H4 test shows that the regression coefficient is 0.178, and the p-value is less than 0.05. It can be stated that facilitating condition positively and significantly impacts the use of electronic payment systems (H4 supported/accepted).

Performance Expectancy on the Use of Electronic Payment Systems
Performance expectancy (HK) or proven usefulness are determinants of the actual use of electronic payment systems. The higher the performance expectancy, the higher the system usage. The study results support the studies conducted by Gupta & Arora (2020) and Madan & Yadav (2016). This is because using transactions with mobile wallets can increase productivity and save time because it is possible to do this anywhere and anytime (Gupta & Arora, 2020). The survey results showed that the majority (81.8%) of respondents agreed that electronic payments were useful in everyday life, most (72.9%) of respondents agreed that electronic payments increased the opportunity to done other more important things, partly A large proportion (83.8%) of respondents agree that electronic payments can help get things done faster, and most (62.4%) of respondents also agree that electronic payments can increase productivity.
The benefits that are obtained when using electronic payments, such as being more practical because you no longer need to carry a physical wallet and cash, there are consumer loyalty rewards on platforms certain electronic payment, the opportunity to get promos, more secure, there is transaction history recording, and many other features which can be utilized, such as payments for monthly basic needs, such as Social Security Administrator for Health (BPJS) payments, water, cable TV, credit purchases, insurance payments, alms, ecommerce/marketplace payments, and others. Currently, the electronic payment system features are not only for consumptive needs but are

Vol. 29, No. 2 August 2021 © Centre for Indonesian Accounting and Management Research
Postgraduate Program, Brawijaya University starting to develop for investment needs, such as purchasing gold investments. Based on these features, it will increase productivity because the electronic payment system meets consumptive needs and fulfills the needs of protection, sharing, and investment; all activities can be carried out alone, anytime, anywhere, and guaranteed security. This is supported by the opinion of Madan & Yadav (2016), which states that consumers perceive electronic payments as an easier and faster alternative to paying with cash or transactions at banks because there is no need for physical money transfers or having to come to the bank.

Effort Expectancy on the Use of Electronic Payment Systems
Ease (HU) is a determinant of the use of electronic payments. The research results are in accordance with the study conducted by Andriyanto et al. (2019a) and Arif, Ameen, & Rafiq (2018). A study conducted by Arif et al. (2018) proved that a system that is easy to use, flexible, and easy to understand would increase its use frequency. Ease of use can make users more quickly understand and skilled in using it to form the perception that using the system does not require high effort Andriyanto et al. (2019a). The survey results show that most consumers agree that consumers will use electronic payments when electronic payment systems are easy to learn, easy to operate, and easy to become skilled at when using them.
The electronic payment application system is easy to learn and operate because the electronic payment application has been designed with an attractive user interface view, complete features, clear and concise usage procedures, and easy to remember and simple operation command dialogs, so it will make consumers easy to learn and operate it. The Ipsos survey results also state that consumers believe using electronic payment systems can add new skills and can provide a pleasant experience to use (Ipsos.com, 2020), making it easy to become skilled when using them.

Social Influence on the Use of Electronic Payment Systems
The social environment (PS) is proven to be a determinant of actual use. The results of this study support the study conducted by ; Kar (2020);Pal et al. (2018); dan Tarhini et al. (2016). The social environment can be a determinant of actual use because information obtained from the opinions of friends, colleagues, family, and social media is considered to have more credibility than other sources of information so that it can affect the use of electronic payments (Madan & Yadav, 2016). This is supported by the study results by Ameen et al. (2018), which proves that the more superiors and colleagues who use the

Vol. 29, No. 2 August 2021 © Centre for Indonesian Accounting and Management Research
Postgraduate Program, Brawijaya University online network system, the more likely it will affect users to use and utilize online network systems as a means of communication.
The social environment also plays a role in increasing confidence in electronic payment systems, which are still new (Kar, 2020). Furthermore, Kar (2020) states that the social environment can provide information about previous user experiences about electronic payment systems to convince and shape perceptions about the use of electronic payments. Information from previous consumers can increase usage confidence because electronic payment systems are systemic risks and are sensitive matters because they involve money issues (Pal et al., 2018).
The social media environment can also influence adoption decisions, according to a study conducted by (Tarhini et al., 2016), which states that social media such as FB, Twitter, blogs, SMS, TV, newspapers, radio, and e-mail can form a positive image because they are currently widely used by the public. The survey results showed that the majority (67.2%) of respondents agreed that social media influenced their decision to use electronic payments. Indonesia is also one of the countries with the highest use of social media. Data shows that active social media users as of 2019 reached 56% of Indonesia's 268.2 million total population (Kompas.com, 2019). Social media is a place to share information, such as previous user experiences, promotions, etc., so that it can form a reputation and form perceptions that can influence actual use.

Facilitating Condition on the Use of Electronic Payment Systems
The availability of facilities, resources, and knowledge (KM) is a determinant of interest in using electronic payment systems. The study results are in accordance with studies conducted by Ameen et al. (2018) and Andriyanto et al. (2019a). A study conducted by Arif et al. (2018) proved that the availability of resources (hardware and software), training and usage guidelines to access the network online system, the more likely users are to use it more often. The context of the village e-budgeting accounting system shows that the village as users of the village ebudgeting system needs supporting facilities such as hardware, software, and internet networks to access the system Andriyanto et al. (2019a). Furthermore, Andriyanto et al. (2019a) stated that the availability of technicians who are ready to help at any time could also support the comfort of village officials when experiencing difficulties that can hinder their performance.
The survey results showed that the majority (94.7%) of respondents agreed that accessing electronic payments requires a smartphone, internet network, and money. Most (86.2%) of respondents agreed that using an electronic payment system requires knowledge. Most of the respondents (68.4%) agreed that the need for help from others

Vol. 29, No. 2 August 2021 © Centre for Indonesian Accounting and Management Research
Postgraduate Program, Brawijaya University when experiencing difficulties in operating an electronic payment system. It can be concluded that accessing electronic payment systems requires facilities, resources, and knowledge. Data shows that 93% of respondents are workers, and the majority (79%) of respondents are also highly educated, so they are believed to have resources (such as money, smartphones, and the internet), facilities, skills, and knowledge in accessing electronic payments.
The Indonesian government has also provided supporting facilities to facilitate access to the use of electronic payment systems, such as facilitating the integration of payments between banks and other financial services in one QR Code called the Quick Response Code Indonesian Standard (QRIS) and providing security guarantees for money stored in electronic money instruments in Bank Indonesia Regulation (PBI) Number 20/6/PBI/2018 (Bank Indonesia, 2020b). Besides the government, the service provider of the electronic payment system (platform) also facilitates a lightweight application for all TIPA smartphones and continues to be updated regularly so that the electronic payment application system is easily accessible for all segments of society.

CONCLUSION
Based on the test results prove that the usefulness of the system is a determinant of the use of electronic payments because transactions using electronic payments allow it to be done anywhere and anytime and can help get things done more quickly, thereby increasing performance and productivity. Ease of operation of the system determines the use of electronic payment systems because electronic payment application systems are easy to learn, easy to operate, and easy to become skilled when using them. The social environment is a determinant of the actual use of electronic payments because information obtained from the opinions of friends, colleagues, family, and social media is considered to have more credibility than other sources of information so that it can affect the use of electronic payments. The availability of resources, facilities, and knowledge determines the use of electronic payment systems because when resources, facilities, and knowledge are available, they will facilitate access to electronic payment systems and provide convenience when using them.

Limitations
The limitation of this study is the limited access to distributing the online questionnaire link. The online questionnaire link was distributed

Vol. 29, No. 2 August 2021 © Centre for Indonesian Accounting and Management Research Postgraduate Program, Brawijaya University
with the help of third parties, namely social media accounts with many followers, and asked the help of native Malang friends to post the questionnaire link on their respective IG and WA status. Researchers have sent requests for assistance in distributing questionnaires to several celebrities, influencers, and account admins where the people of Malang City share information through direct messages and/or email/WA listed in the account information. However, many do not respond and are not willing to help, so that researchers only spread the accounts that are willing to help spread it. The distribution of questionnaires was also limited; that is, only one posting was allowed. Researchers are only distributing questionnaires online because of the corona pandemic. Therefore, sample collection is not optimal due to limited space, time, and cost.

Suggestions
Suggestions for further research are to use different research objects and/or populations because electronic payment system operators that have obtained Bank Indonesia permits have increased by 55% from 2009 to March 5, 2020(Bank Indonesia, 2020a. This shows that the digital wallet market in Indonesia is increasingly in demand. Fintech also focuses not only on the payment segment but also on other segments whose user behavior is interesting to research, such as financing, asset management, and another fintech segment. Even more distribution of questionnaires involving third parties should be done, and researchers can pay fees or gifts of appreciation not only to respondents, but also to each Instagram celebrity admin, influencer, and account admin where people share information that has helped distribute questionnaires, so that may post the questionnaire link more than once. Thus, the sample return rate can be maximized.