DETERMINANTS OF E-TAX SYSTEM ACCEPTANCE BY USERS

This study aims to find out the determinants of the e-tax system acceptance by its users based on theory of planned behavior. As we all know that taxes are the largest national revenue for Indonesia, and expected to increase from year to year. Because of it, the government tries to make it easy for society to pay taxes. One of the ways the government has is to improve e-system taxation. The users in this context were the taxpayers domiciled in Surabaya. The analysis technique used in this study was structural equation modeling. WarpPLS 6.0 software was used to process the data collected through the study. Based on the data collected and processed during the study, some findings were found out. First of all, the users' attitudes were significantly influenced by the perceived useful, beliefs, and compatibility. Secondly, the subjective norms were not affected by the external influence, but rather by interpersonal influence. In addition to those, the perceived behavioral control was completely and significantly influenced by self-efficacy and facilitating conditions. Lastly, the users' intentions in using the e-tax system were influenced by two factors only: the users' attitude and the perceived behavioral control

Taxes are one of the largest sources of national revenues in Indonesia. Most of the national budget funds come from tax revenues, from where national expenditures including public financing and national development are financed. Tax revenues are expected to continue to increase from year to year in order to meet all the country needs. For this reason, various policies have been made by the government in order to increase the national revenues gained from the fiscal sector. One of these policies is the implementation of the self-assessment system. The development of information technology continues to increase time by time. Along with the advancement of the information technology, the society is also required to develop. Because of that reason, the government did not want to miss to take advantage of these technological advances by making changes in the form of modernization of the taxation system. This taxation system modernization was carried out by implementing e-system-based technology. The reformation of e-system taxation modernization by the Directorate General of Taxes was an e-tax system that could be used to conduct eregistration, e-Filing, e-SPT, e-Billing and e-Facture. In February 2015, the Directorate General of Taxes (DGT) issued a new application or information system called DGT ONLINE to make it easier for taxpayers to report their annual tax returns, known as SPT (Surat Pemberitahuan Tahunan). The purpose of renewing the tax administration system with the existence of e-tax was to improve taxpayers' compliance and increase public confidence in tax administration. The online tax system also functioned as an effort to eradicate corruption from the tax sector. Through this online program, taxpayers would not be able to manipulate the amount of tax to be paid. This program was one formula to achieve a good system in Surabaya. The use of e-tax is also believed to be more effective and efficient because the time used by taxpayers to collect the taxes will be shorter than tax payments using manual system (Ibrahim, 2014).
In running or using an information system, software, hardware and humans, as the operator, are needed. These three components must be interconnected to each other so that the data processing activities in SPT reporting can run smoothly. Humans as e-tax system users played an important role in the operation of the e-tax system provided by the government. For that reason, users should fully support this system. However, due to several factors, this system had not been fully accepted by the society as the users. Studies in the US have confirmed that the choices made by taxpayers to use e-tax also depend on whether they use a tax advisor to prepare their taxes or not, and the attitude of their tax advisors towards the use of e-tax in tax development and adoption (Thomas, Manly, & Ritsema, 2004). The role of this tax advisor was confirmed to be influential based on the previous studies carried out by (Hansford, Lymer, & Pilkington, 2006) and Lymer, Hansford, & Pilkington (2005) in the United Kingdom. In addition, navigation facilities and accessibility of a system are important in determining individual perceptions of system quality (Saha, Nath, & Salehi-Sangari, 2012). According to the results of the study by Lymer, Hansford, & Pilkington (2012), it was shown that individuals who accepted this online-based system tended to be individuals who have higher level of enthusiasm about technology and interest in experimenting with new processes.
Nowadays, there are still many people who have not used it yet and do not even understand the e-tax system due to various factors from the users' side. This study aims to determine the factors that influence the acceptance of the e-tax system by the users. This study was conducted to find out whether DGT Online's information system was running as it should and increasing taxpayers' compliance in paying taxes, for which it was necessary to evaluate the performance of the information system. Evaluation was carried out to find out the condition of objects by using instruments such as distributing questionnaires to users of the system and the results are compared with a certain benchmark to obtain conclusions and from there how user acceptance of the performance of the system could be assessed.

LITERATURE REVIEW
E-Tax is a service provided by the government to facilitate taxpayers in making payments and sending tax deposit data through online media to the tax system of the Directorate General of Taxes. E-tax is a form of government programs' embodiment namely e-government which is used to improve relationship between the government and other parties, one of which is society. At the moment, studies on e-government have attracted many people around the world because these studies can provide many reports which are essential for the development of e-government, which can bring about a fundamental transformation of government services (Hung, Chang, & Yu, 2006). As a form of egovernment program, electronic tax is used to improve tax administration efficiently. E-tax is expected to have advantages for Directorate General of Taxes because this system is expected to ease the fraud detection in terms of tax payments, as well as it will benefit the users for which the tax payment will be easier than using the manual system.
The theory of planned behavior is a theory that emphasizes the rationality of human behavior as well as the belief that behavioral targets are under the control of individual consciousness. According to Taylor & Todd (1995) this theory provides a full understanding of behavioral intention by focusing on factors that tend to influence the use of certain systems through design and implementation strategies. This theory was created by Icek Ajzen and later was further developed together with his colleague, Martin Fishbein in 1975. Based on this theory, behavior does not only depend on one's intention, but also on other factors that are not under the control of the individuals, such as the availability of resources and opportunities to display the behavior (Ajzen, 2005). According to Fishbein & Ajzen (1975) intention is a component in the individual who refers to the desire or interest of an individual to do certain behaviors. Bandura (1986) states that intention is a determination to carry out certain activities or produce certain situations in the future. Ajzen (1991), then, reintroduced theory of planned behavior by adding a new component, which is called perceived behavioral control. With this, he expanded the Theory of Planned Behavior to predict behavioral intention and actual behavior. Ajzen (1991) added the concept of behavioral control perception, which came from the theory of self-efficacy. Self-efficacy was first coined by Bandura (1986) from his research on social cognitive theory. According to Bandura (1986), self-efficacy is the belief of an individual that he can succeed in carrying out the behaviors needed to produce results.

METHODS OF RESEARCH
This study used survey method and the population was society in Surabaya. 51 respondents, consisting of 20 males and 31 females, were selected in Surabaya by using purposive and accidental sampling technique. The measurement scale used was a Likert scale with the following criteria: (1) Strongly disagree, (2) Disagree, (3) Neutral, (4) Agree, (5) Strongly agree. The analysis technique used is Structural Equation Modeling. Structural equation modeling is a multivariate statistical analysis technique that is used to analyze structural relationships. This technique is the combination of factor analysis and multiple regression analysis. Structural equation modeling (SEM) is a multivariate statistical framework that is used to model complex relationships between directly and indirectly observed (latent) variables (Stein, 2013). The structural equation model implies a structure for the covariances between the observed variables, which provides the alternative name covariance structure modeling (Hox, 1998). In this paper, WarpPLS 6.0 software was used to process the data.

RESULTS AND DISCUSSION
The results of data processing conducted during the study indicated that the users' attitudes were influenced by perceived usefulness, trust, and compatibility factors. Meanwhile, perceived ease of use, risk, and personal innovativeness did not have significant effects on the users' attitudes. Therefore, from the hypotheses that had been prepared beforehand about the attitude of the users, only H1, H4, and H6 were accepted. There are three important things that can be taken from the results of the data processing, the first of which is that the behavior of taxpayers who adopt e-tax and nonadopters in using e-tax was significantly influenced only by the usefulness factor of the system on the users, the users' trust in the e-tax system, and the compatibility of the system according to the users. Secondly, according to Agarwal & Prasad (1998), personal innovativeness is the nature of individuals in adopting information technology which can help to identify users who tend to adopt innovations on information technology earlier than other people. In addition, based on the results of data processing conducted in this study, it is shown that e-tax adopters do not significantly have a higher level of personal innovativeness than individuals who are not e-tax adopters. This is also supported by previous study conducted by Hung, Chang, & Yu (2006). Hung, Chang, & Yu (2006) stated that the possible cause was that the acceptance or use of e-tax was not seen as a problem but rather an interesting task for e-tax adopters with high personal innovativeness. Lastly, the results of the data processing show that perceived ease of use and risk are not significant determinants of the taxpayers' attitude of both users and non-users of e-tax. Warkentin, Gefen, Pavlou, & Rose (2002) in his study stated that perceived risk is an individual's subjective expectations of losses suffered in pursuit of a desired outcome, so that it will affect the attitude of the taxpayers to e-tax. Whereas perceived ease of use according to Davis (1989) is a level believed by an individual that by using a certain system they will have less effort, therefore the less effort needed in using a system, the greater the acceptance of an individual in the system will be. However, the empirical results of this study state otherwise, which shows that ease of use and risk reduction are not significant factors. The might be because the taxpayers selected as samples for this study feel that these two determinants are not easily perceived or have only a little effects on them (Hung, Chang, & Yu, 2006).
The next hypothesis was about the influences on subjective norms. According to Mathieson (1991) subjective norms reflect the perceived opinions of referent other, which "referent other" means a person or a group who may be an important person to individual. Bhattacherjee (2000) stated that in the absence of first-hand experience with a product or service, individuals may rely on second-hand or vicarious experiences for deciding among behavioral choices, because second-hand experience may be an effective, inexpensive, and convenient way of forming intentions about using new, unproven services such as ecommerce. Based on processed data, subjective norms were only significantly influenced by interpersonal influence but were not significantly influenced by external ones, meaning that H8 was accepted while H7 was rejected. In other words, the users only considered subjective norms based on what each individual felt and not even trustable people for the users can influence their decision whether or not to use the e-tax system. In other hand, perceived behavioral control was all influenced significantly by self-efficacy and facilitating conditions; which mean that H9 and H10 were equally accepted.
Next was about the users' intention to use the e-tax system. Based on all data and hypotheses that had been processed, the users' intention to use the e-tax system was influenced by two factors, namely attitude and perceived behavioral control, while subjective norms did not significantly influence the user's intention to use the e-tax system.

CONCLUSION
Based on the study conducted, some findings were gained. Firstly, the users' attitude was significantly influenced by perceived usefulness, trust, and compatibility. Next, the two subjective norms were not influenced by external influences, but were only influenced by interpersonal influences. Furthermore, it was found that perceived behavioral control was entirely influenced significantly by self-efficacy and facilitating conditions, which meant that both hypotheses regarding perceived behavioral control, namely H9 and H10 were equally accepted. Finally, the users' intention to use the e-tax system was influenced by two factors, namely attitude and perceived behavioral control, while subjective norms did not significantly influence the users' intention to use the e-tax system.

SUGGESTIONS
The government should promote e-tax system usage socialization so that society in Surabaya is well educated about the e-tax system. If it has been well socialized, it will be easier for society to believe and accept the e-tax system.
It will be better if the government does not only provide introductory socialization. Training programs should also be held in using the e-tax system because there are still many Surabaya people who have difficulty in using the e-tax system. In addition, the government must also make the e-tax system more user-friendly so that it will be more easily understood and applied by users.