Electronic tax reporting or e-filing is a system that provides important value to taxpayer compliance ((Dakhli, 2022); (Aziz et al., 2017); (Aziz & Idris, 2014)). Therefore, successful reporting with e-filing on the one hand will increase taxpayer confidence and on the other hand increase state revenue from the tax sector while tax itself is very important for national funding ((Abdul-Aziz et al., 2022); (Seno et al., 2022); (Gobena, 2024); (Animashaun & Chitimira, 2023)). Related to this, e-filing is a solution that provides convenience and benefits because it is practical for tax reporting (Mapope et al., 2022). The underlying argument, first: the existence of a trustworthy partner. Therefore, the public in general and taxpayers in particular do not need to worry about the authenticity of the documents and tax returns reported because Klikpajak has been officially registered as a partner of the Directorate General of Taxes. Second: reporting can be done practically, quickly, easily and certainly safely. This means that e-filing is an application that can be used for various tax reports and is certainly integrated in real time online. Third: e-filing can record all report history automatically, neatly, easily accessible again and of course this is guaranteed because of ISO 27001 certification. Fourth: e-filing services are easily accessible on the Directorate General of Taxes website https://djponline.pajak.go.id or application service providers (ASPs) because they are connected to internet access.

Electronic tax reporting or e-filing is a system that provides important value to taxpayer compliance (Dakhli, 2022); (Aziz et al., 2017);(Aziz & Idris, 2014)). Therefore, successful reporting with e-filing on the one hand will increase taxpayer confidence and on the other hand increase state revenue from the tax sector while tax itself is very important for national funding ((Abdul-Aziz et al., 2022); (Animashaun & Chitimira, 2023); (Gobena, 2024); (Seno et al., 2022)). Related to this, e-filing is a solution that provides convenience and benefits because it is practical for tax reporting (Mapope et al., 2022). The underlying argument, first: the existence of a trustworthy partner. Therefore, the public in general and taxpayers in particular do not need to worry about the authenticity of the documents and tax returns reported because Klikpajak has been officially registered as a partner of the Directorate General of Taxes. Second: reporting can be done practically, quickly, easily and certainly safely. This means that e-filing is one application that can be used for various tax reports and is certainly integrated in realtime online. Third: e-filing can record all report history automatically, neatly, easily accessible again and of course this is guaranteed because of ISO 27001 certification. Fourth: e-filing services are easily accessible on the Directorate General of Taxes website https://djponline.pajak.go.id or application service providers (ASPs) because they are connected to internet access.

Technology adoption, including e-filing, is a consequence of technological developments in the current era that promise convenience and usefulness as described by TAM theorists (Naeem et al., 2023). In fact, TAM is one of the robust models to explain a number of technology adoption cases, not only in the field of information systems. Consequently, the TAM model can be applied to a number of aspects and the development of the TAM model is carried out to answer various technology adoption issues referring to various observation settings in various countries. Therefore, the use of the TAM model with various extensions and model developments is expected to answer various problems related to technology adoption. On the one hand, this is a challenge to answer various existing problems and on the other hand, of course, this is an opportunity to explore and exploit technology adoption research, including in the case of e-filing adoption.

Technology adoption cannot be separated from the interests of convenience and usefulness so that this will attract individuals to accept it. Therefore, the synergy between convenience and usefulness is an important attraction in technology adoption, including e-filing. However, the convenience and usefulness determine individual attitudes whether to accept or reject so that the aspects of convenience and usefulness become perceptions that cannot be ignored by individuals. Perceived ease refers more to the understanding that adoption will increase ease of use or operation (Saibon et al., 2016)). In essence, the adoption improves performance through the promised convenience. Perceived usefulness is inseparable from the importance of adoption benefits. This means that the adoption carried out increases benefits compared to not accepting its adoption ((Jouda, 2020); (Ly & Ly, 2022); (Nguyen-Viet & Huynh, 2021); (Rifat et al., 2019); (Aness & Kumar, 2017); (Motwani & Haryani, 2015); (Azmi et al., 2012); (Santhanamery & Ramayah, 2012); (Azmi & Kamarulzaman, 2010)). Therefore, the second and third hypotheses of this research are:

H2 : Perceived usefulness affects consumer attitude

H3 : Perceived ease of use affects consumer attitude

The adoption of technology, including e-filing, cannot be separated from the involvement of the government, not only as a regulator but also to support the formal legal operations. In this regard, government involvement and support are important, including the commitment to secure the implementation of these regulations. This includes the interest in being able to provide legal certainty to the regulations set by the government. Therefore, the stronger the legal certainty supported by the government's involvement and seriousness behind the established regulations, it will affect certainty and this will certainly encourage positive attitudes of individuals to accept technology adoption, including e-filing, and vice versa ((Subedi & Tamang, 2023); (Khan et al., 2021); (Sánchez-Torres et al., 2018); (Aness & Kumar, 2017)). Therefore, the fourth hypothesis of this research is:

H4 : Government support affects consumer attitude

Behavioral theory illustrates that intention is influenced by attitude so that this is a stimulus to create a positive attitude. This means that a positive attitude will have an impact on the intention to take action and conversely a negative attitude will trigger sentiment so that it does not intend to take action. Therefore, the adoption of e-filing cannot be separated from the importance of building a positive attitude, especially associated with guarantees of ease and usefulness of e-filing adoption. So, the synergy of attitude and intention is an important factor of adoption cases, including e-filing so that the stronger a person's positive attitude, the greater the intention to adopt e-filing, and vice versa ((Nguyen-Viet & Huynh, 2021); (Hossain et al., 2020);(Jouda, 2020); (Night & Bananuka, 2020)). Therefore, the fifth hypothesis of this research is:

H5 : Consumer attitude affects behavior intention

From the explanation above, the research model that will be tested in this research is as shown in the following Figure 1.

The time to distribute the questionnaires are once week on November 4^th-9^th, 2024 was 300 respondents collected, but 20 respondents did not fill in completely, which resulted in 280 respondents. Although, there were 35 respondents who filled out the questionnaire incompletely and 45 respondents who showed bias because they had a certain model or pattern so that 80 respondents were excluded from this study and 200 samples were tested next. In this study, 200 respondents were grouped based on certain demographic characteristics, including the gender of respondents dominated by men (60%), age criteria dominated by 41-50 years (46.5%), education criteria dominated by undergraduate (65%), employment criteria dominated by civil servants (83%), marital status criteria dominated by married (57%), bank account ownership criteria dominated by conventional bank owners (66%), income criteria dominated by above Rp 5. 000,000 (87%), criteria for frequency of internet use dominated by more than 5 hours per day (55.5), criteria for experience of internet use dominated by more than 5 years (76.5%), use of online services dominated by experienced (88.5) and ability to use the internet dominated by good ability (73). Respondent demographics are shown in Figure 1.

PLS-SEM testing begins with testing the structural model (outer model). This test is carried out by testing reliability and validity. Reliability testing is done by looking at the Composite Reliability (CR) value, with a threshold value of 0.70 (> 0.70) (Hair et al., 2021). The CR value of this study is at the threshold of 0.768 to 0.855 (> 0.70) and it is concluded that the constructs in this study are reliable. Furthermore, validity testing is carried out by testing convergent validity and discriminant validity. Convergent validity testing is done by looking at the outer loading and Average Variance Extracted (AVE) values. The overall outer loading value is at the threshold of 0.652 to 0.873 (> 0.60) and the AVE value is at the threshold of 0.526 to 0.661 or more than 0.50 (> 0.50) (Hair et al., 2014). Discriminant validity testing is done by looking at the Fornel-Larcker Criterion value (Hair et al., 2019). The test results show that the correlation value between variables has a smaller value than the AVE root. Overall, it can be seen that the indicator constructs of this study are valid. The results are shown in tables 2 and 3.

Testing the inner model begins with looking at symptoms of multicollinearity by looking at the Variance Inflation Factor (VIF) value. The VIF value in this study is at the threshold of 1,000 to 3,028 or below the threshold of 5 (<5) (Hair et al., 2014). So it is concluded that there is no multicollinearity problem. The test continued with the coefficient of determination (R²). Focusing on the behavioral intention construct, the R² value is 0.232. This means that 23.2% of the constructs of this study explain behavioral intention. Furthermore, testing the effect size is done by looking at the Cohen f² value. An effect size of 0.02 means small, a value of 0.15 has a medium effect size, and 0.35 means large (Hair et al., 2014). The results of this research have f² values in the range of 0.092 to 0.308. In testing the predictive relevance (Q²), it is carried out to determine the predictive relevance of the model in this study. The Q² value focusing on the behavioral intention construct is 0.119 and indicates that this research model has predictive relevance because it is above the 0 threshold (Hair et al., 2021). The test results are shown in Table 4.

The final test is hypothesis testing by applying the bootstrapping 5,000 approach resampling procedure for two tailed significance, as suggested by Hair et al (2021). The test results show that all hypotheses are accepted, among others, and the explanation is as follows:

Hypothesis 1, a significant negative effect between perceived risk and consumer attitude (β: 0-0.344; p-value <0.01). Risk is part of technology adoption, including e-filing, while the risk itself can be in the form of human error or technical error. Therefore, reducing risk is important and one of the efforts made is through education and literacy. Related to this, risk is considered an obstacle to technology adoption, although on the other hand this is considered a challenge to support successful technology adoption (Rifat et al., 2019). This means that risk can affect individual attitudes so that when individual attitudes are positive, they will accept and vice versa if the risk has an impact on negative individual attitudes, it will trigger sentiment towards attitudes and ultimately reject adoption (Ly & Ly, 2022) (Jouda, 2020). The results of this research corroborate the findings of (Nguyen-Viet & Huynh, 2021) and (Jouda, 2020), although on the other hand the results of this research contradict the results of(Ly & Ly, 2022).

Hypothesis 2, a significant positive effect between perceived usefulness on consumer attitude (β: 0.539; p-value <0.01) and Hypothesis 3, a significant positive effect between perceived ease of use on consumer attitude (β: 0.303; p-value <0.01). The TAM theoretical model emphasizes the urgency of convenience and usefulness because adoption is expected not only to provide convenience in its operation but also usefulness in its performance. This means that adoption in taxation through e-filing is also expected to combine convenience and usefulness so as to increase the level of taxpayer compliance. This is important because taxes are a source of state revenue. Therefore, convenience and usefulness are expected to have an influence on individual attitudes so that when individual attitudes are more positive, this has an effect on stronger intentions to accept the adoption of e-filing, and vice versa (Ly & Ly, 2022) (Jouda, 2020). Regarding the usefulness that this research strengthens the research results of (Ly & Ly, 2022),(Nguyen-Viet & Huynh, 2021) and(Jouda, 2020) and on the other hand from the aspect of convenience it turns out to trigger dualism because on the one hand it strengthens the research results of (Nguyen-Viet & Huynh, 2021) and(Jouda, 2020) and on the other hand it contradicts the research of (Ly & Ly, 2022).

Hypothesis 4, a significant positive influence between government support on consumer attitude (β: 0.352; p-value <0.01). It cannot be denied that government support is one of the important factors influencing a regulation. In this regard, e-filing is a regulation in the field of taxation that not only provides convenience to taxpayers but also benefits so that it is expected to increase the level of taxpayer participation and compliance. The implication of participation and compliance is an increase in revenue from taxpayer deposits which then benefits development. Government support can take the form of infrastructure development and facilities that provide convenience of services to taxpayers. Government support is very important because e-filing needs internet access so that it can be done real time online. Government support affects individual attitudes which then lead to e-filing adoption intentions ((Subedi & Tamang, 2023); (Khan et al., 2021); (Sánchez-Torres et al., 2018); (Aness & Kumar, 2017)). These results corroborate previous research ((Subedi & Tamang, 2023); (Khan et al., 2021); (Sánchez-Torres et al., 2018)).

Hypothesis 5, a significant positive effect of consumer attitude on behavioral intention (β: 0.485; p-value <0.01). The TAM theoretical model emphasizes the relationship between attitude and intention in adoption and behavioral research. Therefore, the stimulus to attitude is important to support adoption intention. That is, the more positive the individual's attitude, the higher the commitment to accept the adoption, and vice versa. So, the intention to adopt e-filing cannot be separated from individual attitudes so that stimulus to attitudes becomes important to strengthen the intention to adopt e-filing ((Hossain et al., 2020); (Jouda, 2020); (Night & Bananuka, 2020)). The urgency of the synergy between attitude and intention, the results of this research strengthen the results of previous research ((Nguyen-Viet & Huynh, 2021); (Hossain et al., 2020);(Jouda, 2020); (Night & Bananuka, 2020)).

The research results confirm that e-filing adoption intention is influenced by 5 observed variables, namely consumer attitude, perceived ease of use, perceived usefulness, perceived risk and government support. The research results show that all hypotheses are accepted and of the five hypotheses analyzed, it turns out that the effect of perceived usefulness on consumer attitude has the largest influence value marked by a β value of 0.539 while the smallest is the effect of the perceived ease of use variable on consumer attitude (β value is 0.303) and the effect of the perceived risk variable on consumer attitude is negative at 0.344 (Table 5). The results of testing the five hypotheses illustrate that the expansion of the TAM model supports the intention to adopt e-filing and this is shown in Table 5 while the explanation of the interaction can be seen in Figure 2.