AN INTEGRATED MODEL ON ONLINE TAX ADOPTION IN MALAYSIA

Transcription

1 AN INTEGRATED MODEL ON ONLINE TAX ADOPTION IN MALAYSIA *Ramlah Hussein, Faculty of ICT, International Islamic University Malaysia, Malaysia, Norshidah Mohamed, Faculty of ICT, International Islamic University Malaysia, Malaysia, Abdul Rahman Ahlan, Faculty of ICT, International Islamic University Malaysia, Malaysia, Murni Mahmud, Faculty of ICT, International Islamic University Malaysia, Malaysia, Umar Aditiawarman, Faculty of ICT, International Islamic University Malaysia, Malaysia Abstract As one of the MSC Malaysia Flagship Applications, Online Tax System or e-filing has gained its popularity since it was launched in But the obstacles remain concern taxpayers who are really anxious about the technology, lacking of technical skills, or do not trust of any online transaction. This has led to the investigation of the factors that might contribute to the taxpayers adoption of the online tax or e-filing system. To fulfill this aim, a model of online tax adoption was developed by integrating the Technology Acceptance Model (TAM) theory, Diffusion and Innovation (DOI) theory, Perceived Characteristics of Innovating (PCI) and other relevant factors including social influence, web-based service quality, trust of the internet, trust of the government political self efficacy, and perceived risk. Thirteen main hypotheses were developed and tested. The results demonstrate that trust of the government and web-based service quality are found to be a significant factors influencing citizens intention to use e-filing. Key implications for theory and practice are also discussed. Keywords: G2C adoption, TAM, perceived risk, trust 1 INTRODUCTION Electronic government or e-government has long been sounding as a breakthrough of communication and transaction between the government and citizens or government and industries. Simply, e-government is meant not only to help government to accomplish daily administrative activities but also to provide more easy way to communicate with external entities such as citizens and businesses throughout the utilization of ICT. According to Gronlund and Horan (2004), e-government can be defined as the use by government agencies of information technologies that have the ability to transform relations with citizens, businesses, and other arms of governments. These technologies can serve a variety of different ends: better delivery of government services to citizens, improved interactions with business and industry, citizen empowerment through access to information, or more efficient government management. The resulting benefits can be less corruption, increased transparency, greater convenience, revenue growth, and/or cost reductions. (pp ). One of the most important reasons of e-government implementation is to embrace citizens and businesses closer to their governments. Consequently, the interactions and transactions can be established at anywhere and anytime. To date, there are many literatures that have been conducted to identify the challenges and opportunities of e-government system in developing countries, not only by international organizations (eg. UN, WorldBank, OECD, etc) but also by practitioners and researchers. Some studies focused on how to adopt the successful plan and the strategy of e-government implementation from developed countries, which in most cases full 1

2 considerations need to be taken because of many major differences from both sides (eg. cultural issues, IT infrastructure, government system, etc). This study attempts to explain the status of G2C adoption in Malaysia by adopting TAM as basic research model. Other supporting theories to explain the adoption of new technology such as DOI and PCI were also adopted in the study. Thus, an integrated research model was developed based on these theories. The determinants of TAM in the context of G2C adoption identified in the past research such as perceived risk, trust and social influence were also tested in this study. 2 BACKGROUND OF STUDY In 1996, the government of Malaysia initiated Multimedia Super Corridor Malaysia (MSC-Malaysia) as a platform to build a competitive market for the ICT companies and industries. MSC Malaysia has hosted more than 900 multinational and local companies that focused on ICT and multimedia products, services, solutions, and research and development. The projects under the e-government flagship have been started since ten years ago aimed at building a more effective and efficient way to communicate and transact with the citizens and industries. One of the projects under e-government flagship is Online Tax System or e-filing. In 2006, the Inland Revenue Board Malaysia (IRBM) on behalf of the government introduced the e-filing system. The objectives of e-filing are to facilitate tax compliance and to provide taxpayers service through the use of Internet technologies and WWW. By using e-filing system, taxpayers are able to preparing, reporting and paying their tax online. Administratively, e-filing may offer a potential benefit to the government because the process of tax return by the citizens can be managed effectively via the enabling technologies. It is reported that the number of e-filing users has outnumbered the expectation of 500,000 users become 874,814 users in 2007; an increase of percent since it was first introduced (Bernama, 2008). But the obstacles remain concerns the taxpayers who are really anxious about the technology, lacking of technical skills, or do not trust of any online transaction. This has led the researchers to investigate the factors that might contribute to taxpayer s intention to use e-filing system. 3 LITERATURE REVIEW With the implementation of e-government in Malaysia, a study about Government to Citizens (G2C) adoption could be significantly important in reflecting its progress. Apparently, most of the established studies on e-government adoption are rooted from the theoretical model of e-commerce adoption. Some of the theories that constitute user s intention to use technology such as, Technology Acceptance Model (TAM), Diffusion of Innovation (DOI), Perceived Characteristics of Innovating (PCI), Trust and Perceived Risk, were empirically tested in e-commerce adoption research (Pavlou, 2003; Gefen et al., 2003; McKnight et al., 2002). Since both e-commerce and e-government share the same characteristics, there is no doubt that such studies of technology adoption in e-commerce could also be applicable in e-government context. Therefore, this study aims to adapt the constructs from established e-commerce adoption models that are mainly based on TAM and web trust. In addition, other factors such as, perceived risk, and political self-efficacy, social influence, and service quality are believed influencing of e-government adoption. 2

3 3.1 Technology Acceptance Model (TAM) Studying the acceptance and use of ICT has been the focus of many researchers in Information System (IS) research for the last two decades. Adopted from the Theory of Reasoned Action (TRA) originated by Fishbein and Ajzen (1975), in his study, Davis (1989) developed Technology Acceptance Model (TAM), which is specifically meant to explain computer usage behaviour. TAM is one of the most theories applied in IS research and has established as a parsimonious powerful model for explaining and predicting technology acceptance (Davis and Venkatesh, 1996). The purpose of TAM is to provide an explanation of the determinants of computer acceptance in general as well as a basis for tracing the impact of external variables on internal beliefs, attitudes and intentions (Davis et al., 1989). Perceived Ease of Use External Variables Attitude Intention to Use Actual Use Perceived Usefulness Figure 1. Technology Acceptance Model (TAM) by Davis (1989) TAM posits that perceived usefulness and perceived ease of use influence computer user s intention and actual usage behaviour. According to Davis (1989), perceived usefulness is defined as the degree to which a person believes that using a particular system would enhance his or her job performance. Meanwhile the perceived ease of use refers to the degree to which a person believes that using a particular system would free of effort. The proliferation use of TAM theory as a research model in predicting usage behaviour of new technologies in recent years has also led researchers and practitioners apply this theory in the Internet and WWW context (Gardner and Amoroso, 2004; Klopping and McKinney, 2004; Chen et al., 2001). However, only few researchers extended the TAM theory by associating with other variables or theories to study e-government adoption. In most of their studies Belanger & Carter (2008) and Phang et al. (2005) integrated the theory of Diffusion of Innovation (DOI) with TAM to provide a comprehensive model of e-government adoption. Other studies conducted by Al-Adawi et al. (2005) and Warkentin et al. (2002) also combined TAM with Trust and Perceived Risk. The study of tax administration in worldwide e-government has received immense attention in recent years. Wang (2002) used TAM to explain electronic tax-filing system in Taiwan based on telephone interviews of 260 participants. He found that the extended TAM can be used to predict users intention to adopt electronic taxfiling system. In Singapore, Tan and Pan (2003) applied in-depth case research and qualitative method to study the transformation of tax administration. The case study was conducted over a period of six months. Before the creation of the Inland Revenue Authority of Singapore (IRAS), tax administration in Singapore was described as classic hierarchical bureaucracy. In 1998, the IRAS introduced an Internet filing system. After three years, the IRAS conducted a survey on taxpayers to determine the outcome of the system introduced earlier. It was found that the majority of individual taxpayers, corporate taxpayers and goods and services taxpayers expressed satisfaction with IRAS services. Tan and Pan (2003) concluded that the IRAS demonstrated success of its ITenabled transformation. In the case study, citizens satisfaction appeared to be the measure of success. Following previous study, this study also attempts to explain user s behavioral intention to use e-filing by adopting TAM 3

4 theory. Further, the authors are also interested in confirming the relationship among the TAM constructs in online tax adoption context, thus we hypothesized: H3: Perceived ease of use will positively influence perceived usefulness of e-filing system. H4: Perceived ease of use will positively influence intention to use e-filing system. H5: Perceived usefulness will positively influence intention to use e-filing system. 3.2 Diffusion of Innovation (DOI) and Perceived Characteristics of Innovating (PCI) Beside TAM, Diffusion of Innovation (DOI) initiated by Rogers (1995), is another popular theory used in IS research to explain user adoption of new technology. Diffusion is defined as the process by which an innovation is communicated through certain channels over time among the members of social society, while innovation refers to an idea or object that is perceived to be new. According to Rogers (1995), the rate of diffusion is affected by an innovation s relative advantage, compatibility, complexity, trialability and observability. It has been demonstrated in his study that compatibility is particularly important in explaining an innovation s rate of adoption. Compatibility is defined as the degree to which an innovation is perceived as being consistent with the existing values, past experiences, and needs of potential adopters. In the context of e- government, Carter and Belanger (2005) implied that citizens are willing to use e-government technologies if these services are congruent with the way they like to interact with others. Their findings indicated that compatibility was found as the most significant factor influencing e-government adoption. Moore and Benbasat (1991) modified Rogers constructs by eliminating complexity from the study and introduced four other new constructs: image, result demonstrability, visibility and voluntariness that were believed influencing the acceptance and use of an innovation. This extended model is known as Perceived Characteristics of Innovating (PCI). Image refers to one s perception of an innovation as a status symbol. Carter and Belanger (2005) associated image with subjective norm of Theory reasoned Action (TRA), developed by Fishbein and Ajzen (1975). For those citizens who view the adoption of e-government as prestigious are likely engaging further and continue to use the system in the future. To utilize or not to utilize the e-government service is a matter of how individual takes the opportunity to communicate and transact with the government effectively and efficiently. It is believed for those who adopt the e-government system will create their own status in the community as savvy and progressive people. Furthermore, Moore and Benbasat (1991) argued that Davis Perceived Usefulness (PU) and Perceived Ease of Use (PEOU) are Similar to Roger s Perceived Relative Advantage and Perceived Complexity respectively. Since the reliability of both constructs PEOU and PU has been well tested by various studies in different contexts (Davis, 1989; Davis et al., 1989; Davis and Venkatesh, 1996) hence, this study prefers to use these constructs over Relative Advantage and Complexity. Past studies indicated that observability construct was quite complex to be applied. Observability is defined as the degree to which the results of an innovation are visible and communicable (Rogers, 1995). Rogers posits that software dominant innovation has less observability and usually has slower rates of adoption than hardware innovation. Thus, the more a potential adopter can see an innovation as visible, the more likely he/she adopts it. Moreover, such visibility has led individual to share and communicate his/her experience to others. With this argument, Moore and Benbasat (1991) formed a new construct called as Result Demonstrability that combined the observability and communicability. This construct is believed to give better understanding on how the users view innovation as something that gives them advantages to their life s practice and how they share and communicate their experience of using it to others. Based on the arguments above, we proposed the following hypotheses: H8: Compatibility will positively influence intention to use of e-filing system. H9: Image will positively influence intention to use of e-filing system. H10: Result demonstrability will positively influence intention to use of e-filing system. 4

5 3.3 Social Influence Studying new technology adoption, social influence has been acknowledged as an important factor influencing behavioral intention. Previous studies assessed the social influence factor on one s behavior through various measurements (Malhotra & Galletta, 1999; Venkatesh & Davis, 2000; Venkatesh & Morris, 2000; Lee et al. 2006). The most popular theory about social influence was originated from psychology discipline by Fishbein and Ajzen (1975) known as subjective norm referred to one s perception that most people who are important to him think he/she should or should not perform the behavior in question. Subjective norm is believed to influence intention to use because people may choose to perform behavior, even if they are not themselves favorable toward the behavior or the consequences. If they believe one or more important referents think they should, and they are sufficiently motivated to comply with the referents (Venkatesh & Davis, 2000). Another study by Kelman (1958) viewed social influence from different aspects as he distinguished between the three different processes of social influence toward individual behavior they are compliance, identification, and internalization. By distinguishing these processes, one could ascertain if usage behavior is caused by the influence of referents on one s intent or by one s own attitude. Past studies have examined the direct and indirect effect of subjective norm on technology adoption, but the results have come out with different implications. Davis et al. (1989) and Mathieson (1991) found no significant effect of subjective norm on intention, while Taylor & Todd (1995) did find significant effect. Venkatesh & Davis (2000) have justified these various results based on the user s experience of using the new technology and the usage contexts, mandatory or voluntary. They argued that in voluntary situation, subjective norm significantly influenced perceived usefulness as the users gained direct experience with a system over time, thus they relied less on social information. While in mandatory situation for inexperienced users, subjective norm has a direct effect on intention to use as a result of information circulation by referents. In e-filing adoption context, reporting and paying taxes are obliged for those who have income. However, the method to settle the transaction using e-filing is voluntary. For those who have used e-filing might be benefited by the effectiveness and efficiency of doing transaction online. But the initiative behind to start using the e-filing is in quest. Hence, subjective norm is considered to be important factor to explain one s behavioral intention to use new technology. Thus, this study interested in examining the effect of subjective norm on intention to use e-filing. With this argument we hypothesized: H1: Social influence will positively influence perceived usefulness of e-filing system. H2: Social influence will positively influence intention to use e-filing system. 3.4 Service Quality Service quality is one of the focuses in marketing research as well as IS research recently. The study about service quality by Parasuraman et al. (1985) has critically identified some dimensions that reflect service attributes provided by businesses. Their study has carried out a well-known scale called SERVQUAL containing 10 basic dimensions to measure the quality of service. Extensive studies by other researchers to evaluate various services provided by businesses using SERVQUAL have also been conducted in the past 20 years. Later the SERVQUAL was refined into five distinct dimensions namely, tangibles (physical facilities, equipment, and appearance of personnel), reliability (ability to perform the promised service dependably and accurately), responsiveness (willingness to help customers and provide prompt service), assurance (knowledge and courtesy of employees and their ability to inspire trust and confidence) and empathy (caring, individualized attention the firm provides its customer). 5

6 Based on the multi-item scales of SERVQUAL, researchers and practitioners have conducted the research to examine customer s perceptions of service quality on the internet environment, either to test the scale on specific study (e.g. e-commerce, web services, etc) or to extend the scale which then fit with overall online services (Zeithaml et al., 2002, Li et al., 2002; Parasuraman, 2005; Kuo et al, 2005). Several aspects need to be taken into account before adopting the SERVQUAL scale considering the interactions and transactions through online are different from the traditional way. Thus, some researchers have developed their own model to study about online service quality. Li et al., (2002) conducted the study on electronic shopping using the SERVQUAL scale. The study involved international participants to test the proposed model for measuring the quality of web-based service. The result demonstrated that two dimensions of SERVQUAL (reliability and assurance) have fallen into a single dimension due to the broad definition of these two scales. The study developed a 28-items scale reflecting six dimensions namely, responsiveness, competence, information quality, empathy, web assistance and call back systems. As the first who initiated SERVQUAL scale, Parasuraman et al. (2005) have also conducted the same study on online shopping by adopting the original SERVQUAL scale. They conceptualized two different scales to capture user s perceptions of electronic service quality. The first scale which they named as E-S-QUAL contains of 22 items reflecting four dimensions: efficiency, fulfilment, system availability, and privacy. The second scale was labeled as E-RecS-QUAL (e-recovery service quality) consisting of 11 items representing three dimensions: responsiveness, compensation, and contact. Both scales have been tested empirically in different research setting concerning online service quality. In e-government context, service quality might be an important factor to explain citizen s acceptance of e- services. As it is believed that user perceived service quality through online differs from conventional. Citizens might be satisfied with the current conventional service when they are dealing with the tax administration. In online environment, the situation is totally different, where citizens will gain new experience using the online system with the service offered that they may not find in conventional. Hence, examining the quality of e-service could determine whether users tend to continue to use the system or not. It is believed also by offering the best service will entice the citizens to use online tax and gain the advantages from it. These arguments have led us to postulate the following hypotheses: H6: Service quality will positively influence perceived usefulness of e-filing system. H7: Service quality will positively influence intention to use of e-filing system. 3.5 Trust and Perceived Risk The dimensions of trust and perceived risk could be the factors that contribute to citizen s intention to use e-government system. It cannot be denied that trust become a central issue in all daily interactions, communications, transactions and practices, especially when it s done remotely, through the internet. Although there is little agreement in the literature about how to define citizen trust in government context or how it is gained and lost, most researchers agree that it is an important determinant of public action and cooperation (Thomas, 1998; Welch and Hinnant, 2003; Pavlou, 2003). In his study, Rotter s (1971) defined trust as an expectancy that the promise of an individual or group can be relied upon. This definition has led to some of the definitions of citizen s trust in e-government context. Zucker (1986) divided trust into three modes, these include; characteristics-based trust, process-based trust, and institution-based trust. Characteristics-based trust is associated with personal characteristics such as, gender, age, race, nationality, etc. Researchers (Welch and Hinnant, 2003; Warkentin et al., 2002) agreed that this mode of trust may not feasible and difficult to apply in e-government environment. Since characteristics-based trust is related to social issues, it may explicitly contradict the legal statutes if it is applied in the study. Process-based trust is a trust based on individual s prior experience when dealing with government. This mode of trust can be created in e-government context if the government could ensure the e-government technologies are well prepared 6

7 and maintained with high security system before it is implemented. Furthermore, encouraging and convincing the citizens that they would have same experience when they use e-government compared to traditional system are also one of the factors in building up citizen s trust. Lastly, institution-based trust is associated with individual perceptions of the institutional environment such as; the structure, regulation and legislation that make an environment feel safe and trustworthy. Belanger and Carter (2008) suggested two types of trusts in e-government context. Firstly, trust of the internet, is identified as a key predictor of e-service adoption (Carter and Belanger, 2005; McKnight et al., 2002; Pavlou, 2003). This type of trust is associated with the institution-based trust because of the communities that exist on the internet are varied according to their norms, expectations, and values (Pavlou, 2003). Secondly, trust of the government, is related to citizen confidence in the ability of an agency to provide online services. The information and services provided by the government through the internet are not automatically building up people trust on how they are delivered. Privacy, security, and fraud on the internet are still surrounding people s concern to adopt e-government technologies. Nelson (1997) also added that user trusts in web-based technologies are affected by such issues in networks security, confidentiality, reliability of information, and jurisdiction. Thus citizen must have confident in both the government and the technologies (Carter and Belanger, 2005). Trust is essentially needed only in uncertain situations since trust effectively means to assume risks and become vulnerable to trusted parties (Hosmer, 1995). Trust is found to be a significant antecedent of perceived risk. If there was no risk and actions could be taken with complete certainty no trust would be required (Pavlou, 2003; Al-adawi et al., 2005). Prior studies have discussed the relationship between trust and perceived risk (Pavlou, 2003; Belanger and Carter, 2008; Al-Adawi et al., 2005; Jarvenpaa and Tractinsky, 1999). It was found that perceived risk decreases when trust occurs. However, since risk itself is difficult to measure objectively, established research has predominately defined perceived risk as the citizen s subjective expectation of suffering a loss in pursuit of a desired outcome (Warkentin et al., 2002; Belanger and Carter, 2008). In association with TAM constructs, previous studies have shown that trust has a relationship with perceived usefulness, perceived ease of use and intention to use (Pavlou, 2003; Gefen & Straub, 2002, Gefen et al., 2003; Belanger ad Carter, 2008). In the context of e-commerce, Pavlou (2003) found trust to be a significant determinant of intention to use whether directly or indirectly thorough perceived ease of use and perceived usefulness. Meanwhile, Belanger and Carter (2008) reported that trust has a direct impact to the intention to use e- government system. In general online tax system, there are two main processes involved during the online transaction, the report and payment. Normally, citizens perceived less risk when they report the tax, but when it comes to the online payment, the level of perceived risk could be higher resulting less intention to use the system in the future. Hence, trust of the internet and trust of the government should be examined accordingly with perceived risk to find out their impact on intention to use. This argument has led us to set the following hypotheses: H11: Perceived risk will negatively influence intention to use e-filing system. H12a: Trust of the Internet will positively influence intention to use e-filing system. H12b: Trust of the government will positively influence intention to use e-filing system. H13a: Higher trust of the internet will reduce the perceived risk of using e-filing system. H13b: Higher trust of the government will reduce the perceived risk of using e-filing system. 3.6 Political Self-Efficacy Citizen s prior experience when dealing with the government agencies remains to be an important issue in G2C adoption. The propensity to use the e-government systems not only depends on trust of the government and the internet, but also the disposition of trust itself as a reflection of one s psychological perceptions based on his or her past experience with the government. Zimmerman (1989) asserted that political self-efficacy is one 7

8 construct that can be used to understand the factors involved in initiating and maintaining the citizen involvement in sociopolitical environment. Highly political self-efficacy could lead one s perception to participate in government s programmes or activities. Conversely, low political self-efficacy might lead one s perception to withdraw from such activities that could create distrustful on the government. In this study, political self-efficacy is believed to be a significant factor that contributes to citizen s trust in dealing with government. There are few studies have been conducted to measure determinants of trust in e- government context (Parent et al., 2004; Belanger & Carter, 2008). As citizen s political self-efficacy has been exist long before the e-government was developed, it is necessary to involve this construct in the study. According to Compeau and Higgins (1995), based on original definition of self-efficacy first coined by Bandura (1986), selfefficacy was defined as the belief that one has the capability to perform a particular behavior. This leads to the definition of political self-efficacy as defined by Parent et al. (2004) as, the citizen s sense that they have an impact on political developments and their perception of government responsiveness. Parent et al. (2004) divided political self-efficacy into two dimensions. Firstly, internal political selfefficacy is where the citizen feels that they have a contribution and an impact to the political development. Secondly, external self-political efficacy refers to the citizen s perception of overall responsiveness of the government in fulfilling their needs. The findings reported that political self-efficacy significantly influenced and explained the citizen s overall trust in e-government context, in turn; the government would be responsive to the citizen s needs. Although e-government initiatives are growing in popularity, citizens are just beginning to acquire more meaningful information about the benefits and consequences of completing transactions with the government online. Hence, one s general propensity to trust will have an impact on e-government adoption through its influence on trust of the internet and trust of the government (Belanger & Carter, 2008). It is suggested that government not only focus on how to improve the e-government technology, but also to stress on how to encourage citizens with high pre-existing level of trust to use e-government system. Based on above discussion, we hypothesized: H14a: Internal political self-efficacy will positively influence trust of the government. H14b: External political self-efficacy will positively influence trust of the government. 4 RESEARCH FRAMEWORK A number of well established research models have contributed to the development of the conceptual framework of technology adoption and the determinants. Among them are Theory of Reasoned Action (TRA), Theory of Planned Behavior (TPB), Diffusion of Innovation (DOI), Technology Adoption Model (TAM), Unified Theory of Acceptance and Use of technology (UTAUT) and Perceived of Innovating (PCI). Based on previous research, TAM is seen as the most parsimonious model to explain user s intention to use new technology. The TAM model has been tested in various setting including, office application, , e-learning, e-banking, e- commerce and e-government. Following previous studies about e-government adoption, current study adopts TAM as the main research model. Furthermore, factors that were believed to be the predictors of e-government adoption are also included and tested in the proposed research model. Figure 2 exhibits the proposed model of this study. The hypotheses are then formulated based on the association among the constructs. 8

9 Compatibility H8 Image H9 H10 Result Demonstrability Perceived Ease of Use H3 Perceived Usefulness H1 Social Influence H2 H4 H5 H6 Intention to Use e-filing H7 Service Quality H12a & H12b H11 H13a & H13b Trust of e-government Trust of the Internet Perceived Risk H14a & H14b Trust of the Government Disposition to Trust Internal Political Self-efficacy External Political Self-efficacy Figure 2. Proposed Research Model 5 RESEARCH METHODOLOGY This study is shaped with quantitative approach and survey design. At the initial stage, the study tends to explore the phenomenon of e-government implementation and its adoption by citizen. In order to obtain a clearer picture about the current implementation of e-government and how it is being used by the citizen, several steps have been undertaken in study. The first step is by critically reviewing the past literatures which mainly related to e-government adoption in Malaysia. Thus, the research problem can be defined and the objectives and research framework can be conceptualized based on the previous works. In addition, the instruments reflecting all the factors involved in the study from G2C adoption literature were adopted and adapted. Based on the proposed model, there are eleven main constructs involved in the study. To measure Social Influence, this study adapts the subjective norms instrument developed by Fishbein and Ajzen (1975). The instrument for Service Quality was adapted from two previous study on e-service quality or web-based service quality by Parasuraman et al. (2005) and Li et al. (2002). Out of fifty items provided by this combination, only twelve items were selected that fit with the study. Instruments for Compatibility, Image and Result Demonstrability were adapted from two previous studies that applied PCI. These three constructs were originally developed by Moore and Benbasat (1992), then Carter and Belanger (2005) adapted the constructs in e- government context. The Trust construct comprises of two dimensions, trust of the internet and trust of the government. There are seven items representing this construct which were adopted from the study by Belanger and Carter (2008). The Perceived Risk construct consists of two items derived from the study by Pavlou (2003). Meanwhile, the Political Self-efficacy construct contains of Internal and External political self-efficacy factors. There are nine items reflecting these dimensions which were adopted from the study by Niemi et al. (1991). The TAM comprises three constructs, Perceived Ease of Use, Perceived Usefulness and Intention to Use. All these constructs were adapted from the previous studies by Davis (1989) and Belanger and Carter (2008). A seven point-likert scale is applied for each item of the constructs except for the citizen s demographic section. The likert-scale is ranging from 1=strongly disagree to 7=strongly agree. 9

10 The next step was to conduct a pilot study. The pilot study was performed to reduce biases in format and content of the instruments. Twenty two participants comprised of professors and lecturers were voluntarily participated in the pilot study to evaluate the e-filing system. The reason of choosing these people because the authors believed they have a better understanding about the system and experiences in using the system that would enable them to evaluate the e-filing more accurately. The reliability test was performed using SPSS (release 16) for Windows to evaluate the instruments. For this purpose, the Cronbach s alpha technique was adopted. According to Hair et al. (1998), if the factor scores above 0.7 of the Cronbach alpha values, then it is considered as reliable. For the actual study, lecturers and administrative staffs from five different public universities in Malaysia have been chosen as the target sample. These universities include Universiti Teknologi Malaysia (UTM), Universiti Teknologi Mara (UiTM), Universiti Malaya (UM), Universiti Putra Malaysia (UPM), Universiti Islam Antrabangsa Malaysia (UIAM). A stratified sampling was employed in the study. Prior to the data collection, an invitational letter was sent to the selected faculties in each university. After the approval, 500 questionnaires were distributed targeting those participants who have experienced the e-filing system. 6 FINDINGS Out of 500 questionnaires collected, there are 411 usable responses which can be used for further data analysis. The rest were unusable due to their incompleteness and other inconsistencies such as typical answers from strongly disagree or strongly agree in all questions. The feedback from respondents yielded a high response rate of 82.2%. To further analyze the data, several steps of analysis were performed. The descriptive analysis was used to portray the data about the respondents and the variables involved in the study. The characteristics of the respondents were exposed descriptively through specific parameters, these include age, gender, race, educational level, computer skill, computer use experience, Internet use experience, and daily Internet usage. Both male and female respondents made up of 47% and 53% respectively. About 29.4% of the respondents held Master degree followed by Doctorate (25.5%), Bachelor (21.7%) and Diploma (16.3%). Besides capturing the general profile of the respondents, information about respondent s computer and internet experience are also reported. In term of computer skill, majority of the respondents (61.1%) are intermediate users followed by 31.9% and 7.1% are expert and beginner users respectively. Due to the activities on web environment where e- Filing take place, it is mandatory to ask the respondents about their general internet experience and usage. About 44% of the respondents have been using the internet for 6 to 10 years and more than half (53.7%) of the respondents were using the internet from 1 to 5 hours a day. The reliability test was conducted based on the collected data. Cronbach Alpha was used to assess the reliability of the construct. According to Nunnaly (1978), a Cronbach s score 0.7 or higher is considered reliable. Each item was assessed by examining the loadings of the items on their respective constructs. As depicted in Table 1, all the factors scored relatively high ranging between and 0.965, with Perceived Usefulness (PU) scored highest. The reliability test indicates that all the constructs are reliable and suitable to measure the concepts employed in the study. No. Construct Cronbach α N of Items 1. Social Influence Trust of the Internet Trust of the Government Perceived Risk Service Quality Compatibility Image

11 Table Result Demonstrability Internal Political Self-efficacy External Political Self-efficacy Perceived Ease of Use Perceived Usefulness Intention to Use Reliability Test A principal component method analysis was used with varimax rotation to test the correlation among the factors. The appropriate cut-off significant loading based on the sample size of this study is 0.4 (Hair et al., 1998). Table 2 summarizes the results of factor analysis suggesting that the measurement exhibited somewhat suitable for the context of this study. Component SQUA SQUA SQUA SQUA SQUA SQUA SQUA SQUA SQUA SQUA SQUA SQUA PUSE PUSE PUSE PUSE IPOS IPOS IPOS IPOS IPOS IMG IMG IMG IMG TRSG TRSG TRSG TRSG EPOS EPOS EPOS EPOS COMP COMP COMP PEOU

12 PEOU PEOU PEOU TRSI TRSI TRSI DEMO DEMO DEMO IUSE IUSE IUSE IUSE PRSK PRSK SOCI SOCI Table 2. Factor Analysis To test the hypotheses underlying the proposed model, correlation analysis was conducted. Correlation analysis was used to determine the association between the predictor factors; social influence, image, result demonstrability, compatibility, service quality, perceived ease of use, perceived usefulness, perceived risk, trust, and political self-efficacy; and criterion factor, intention to use. As can be seen in Table 3, all the predictors except perceived risk were found to have positive relationship with intention to use. It was also discovered that internal and external political self-efficacy had positive relationship with trust. The result exhibits that trust of the internet (r=0.524) and trust of the government (r=0.493) are moderately related to intention to use. This indicates that the citizen believe e-filing system is safe and secured enough to perform online transaction. Besides, the citizen is also assured by the ability of the government to carry out online transaction faithfully. The more people lean their trustworthiness on the internet environment and government agencies the more they intent to use e-filing system. Thus, hypotheses H5a and H5b are supported in the study. Surprisingly, perceived risk (r=-0.002) was not related to intention to use. This result contradicts with some previous studies (Pavlou, 2003; Belanger and Carter, 2008). Past studies have shown a relationship between perceived risk and intention to use, although it was not that strong. In the context of e-commerce, Pavlou (2003) reported that perceived risk was related to intention to use in negative direction. The higher customer s perceived risk of online transaction, the lower consumer s intention to use internet for transaction. On the contrary, Belanger and Carter (2008) found a positive relationship between perceived risk and intention to use in e-government context. Based on these studies, the role of perceived risk in e-government adoption was inconsistent compared to e-commerce. This might due to the nature and the differences underlie these two technologies, e-commerce and e- government. Unlike e-commerce where the commercial view is the main purpose of adoption, e-government was developed as part of the role of the government agency to provide the information and services to the citizen more effectively and efficiently (Warkentin et al., 2002; Carter and Belanger, 2005). It is implied that perceived risk might not become a predictor of intention to use in mandatory situation like e-filing, where law-abiding citizen have no choice but to file taxes. Another insignificant relationship was found between trust of the internet and perceived risk (r=-0.037). On the other hand, trust of the government alone indicated a positive relationship on perceived risk, although it was very weak (r=0.142). These findings have refuted both hypotheses H6a and H6b. While, the disposition to trust, political self-efficacy, was found to be related to both trust of the internet (r=0.233) and trust of the government (r=0.337), validating H7a and H7b. For the TAM constructs, the findings of current study have confirmed several past studies in e-government context (Pavlou, 2003; Wu and Chen, 2005; Carter, 2008). Perceived ease of use was highly related to the intention to use compared to other constructs with r=

13 followed by perceived usefulness with r= It implies that the citizens discover the e-filing system to be easy to use and beneficial, thus they will continue to perform the online tax in the future. Thus H1, H2, and H3 are supported in this study. SOCI COMP DEMO IMG SQUA PRSK TRSI TRSG IPOS EPOS PEOU PUSE IUSE SOCI COMP.255 ** DEMO.374 **.487 ** IMG.442 **.250 **.377 ** SQUA.261 **.533 **.602 **.342 ** PRSK ** TRSI.179 **.460 **.499 **.168 **.552 ** TRSG.135 **.328 **.345 **.356 **.404 **.142 **.319 ** IPOS.196 **.249 **.237 **.258 **.253 **.202 **.210 **.296 ** EPOS.317 **.220 **.266 **.364 **.235 **.169 **.118 *.179 ** PEOU.285 **.560 **.642 **.293 **.617 ** **.370 **.333 **.213 ** PUSE.298 **.508 **.561 **.280 **.564 ** **.363 **.257 **.192 **.718 ** IUSE.333 **.545 **.589 **.429 **.599 ** **.493 **.362 **.258 **.702 **.687 ** **. Correlation is significant at the 0.01 level (2-tailed) Table 3. Correlation Analysis Dependent Variable Adj. R 2 Predictors Std. Beta IUSE.642 PEOU.282 PUSE.275 TRSG.161 IMG.152 COMP.105 SQUA.097 t value p Value Table 4. Regression Analysis using Stepwise Method To determine which predictor that has a strongest impact on intention to use, a multiple regression analysis using a stepwise method was performed. On this analysis, each of the independent factors was added subsequently based on the highest to the lowest variance of the independent variables on dependent variable. Since perceived risk was not related to intention to use, the regression analysis was performed excluding this factor. As can be seen in Table 4, the selected predictors were accounted about 64.2% of the variance of intention to use with F= significant at alpha < Out of the eleven predictors, perceived ease of use has the strongest impact on intention to use (beta=0.282) followed by perceived usefulness (beta=0.275). This finding implies that the e-filing is found to be easy to use and benefited to the citizen. This result confirms the relationship among the TAM constructs in the e-government context. Typical studies have been done to test these relationships in different form of G2C application (Phang et al., 2005; Carter & Belanger, 2005; Carter, 2008). The third strongest predictor is trust of the government (beta=0.195). The past experience from dealing with government agencies has built up a confidence among the citizens to use the e-filing. It is implied that to succeeding the e-government initiatives, the existing government should be trustworthy and keep the interest of the public. In addition, publicity would positively influence the citizen s perception to adopt the e-government services and share the experience with others (Belanger & Carter, 2008). 13

14 The fourth strongest predictor is image (beta=0.152). This indicates that those people who adopt the e- Filing are building up their own status in the community as savvy and progressive people. Their awareness of technology keeps them using the e-filing in the future as the system improved continuously and offer more convenience. The fifth strongest predictor is compatibility (beta=105). The findings have indicated that the e- Filing is compatible with the way they deal with the government in traditional way. The system is found to have the ability in facilitating the citizens to do tax transaction online. Furthermore, user s internet experience such as searching information or online shopping could be the factor influencing perception of adopting e-filing. Thus, by gaining the same experience with the other e-services, adopters tend to continue to use e-filing in the future. The findings are consistent with previous study which examine the impact of image and compatibility on intention to use in e-government context (Carter & Belanger, 2003). Lastly, service quality is found to be the weakest predictor (beta=0.097) influencing intention to use when it is combined with other predictors. Although the e-filing has improved and gain significant response from the citizen, the government should be more apprehended about the services they offer online. Server down, security and privacy remain the factors concerning the citizens when they use e-filing. The service offered through online is totally different compared to traditional form. Hence, the government should take a challenge to extend their services by providing convenience system in transforming the way the citizen conduct the tax transaction online without abandoning quality of service. In other words, a better quality of online service is likely affecting user s intention to use e-filing in the future. 7 CONCLUSION This study should provide a benefit to both researchers and practitioners, in this context government per se. Hence, the findings might give contributions theoretically and practically. Lack of studies in the area of G2C adoption is one of the reasons why such study need to be conducted to obtain a clearer picture of what is going on and the current status of e-government implementation in Malaysia. Some of the theoretical models of new technology adoption such as TAM, DOI, PCI and other contributing factors such as Social Influence, Service Quality, Trust, Perceived Risk, Internal and External Political self-efficacy, have been employed in the study to explain G2C adoption. Thus, the findings of current study is expected not only to strengthen or refute the previous works that adopt the theories mentioned above, but also, in general, to enrich a body of knowledge and theory of e-government adoption by citizen. Furthermore, the research model could be used as a reference for any further studies to cater different form of e-government technology. Practically, G2C system has been implemented in the country for years. But the impact has not yet been experienced fully by the citizens so far. Although numbers of studies have been conducted by the government, institution, independent researchers as well as practitioners, most of them were focusing on the infrastructure, policies and organizational perspectives. Hence, the findings of this study are expected to provide an empirical evidence of e-government adoption from the citizen s perspective. The findings also can be used considerably by the government to evaluate current system that could come with the improvement in the future. Initially, this study provides a basic platform for evaluating the e-government technology in different form and application. Future study should look at other important factors that may influence intention to use e-government services, especially online-tax. This study has integrated the TAM, DOI, PCI theory with social influence, service quality, trust, perceived risk and political self-efficacy. Other predictors such as cultural issue, internet security, and privacy are just some of the factors that might influence G2C adoption. Moreover, the findings of past studies in e-commerce context which have discussed the antecedents of TAM should be taken into an account as well for future study in e-government context. This study has several limitations. Firstly, the survey was only targeting the academic staffs as sample from selected universities in Malaysia. Larger sample size from various backgrounds should be appropriate to 14

15 provide a clearer picture of e-government adoption. Secondly, a stratified sampling method used in the study was focusing on e-filing adopters; hence future study should look at two sides, the adopters and non-adopters. Due to these limitations, extra caution needs to be taken in generalizing the results. References Al-Adawi, Z., Yousafzai, S., & Pallister, J. (2005). Conceptual Model of Citizen Adoption of e-government. The Second International Conference on Innovation in Information Technology Bandura, A. (1986). Social foundations of thought and action. Englewood, New Jersey: Prentice Hall. Belanger, F., & Carter, L. (2008). Trust and Risk in e-government Adoption. Journal of Strategic Information Systems. 17(2), Bernama (2008). E-filing, the preferred choice now of Malaysian taxpayers. Retrieved August 2009, from Carter, L. (2008). E-government Diffusion: A Comparison of Adoption Constructs. Transforming Government: People, Process and Policy, 2(3), Carter, L & Belanger, F. (2003). The Influence of Perceived Characteristics of Innovating on e-government Adoption. Electronic Journal of e-government, 2(1), Carter, L. & Belanger, F. (2005). The Utilization of e-government Services: Citizen Trust, Innovation and Acceptance Factors. Information Systems Journal. 15, Chen, L., Gillenson, M.L. & Sherrell, D.L. (2001). Enticing online consumers: an extended technology acceptance perspective. Information & Management, 39, Compeau, D.R. and Higgins, C.A. (1995). Application of Social Cognitive Theory to Training for Computer Skills. Information Systems Research, 6(2), Davis, F.D. (1989). Perceived usefulness, perceived ease of use, and user acceptance of information technology. MIS Quarterly, 13, Davis, F.D., Bagozzi, R.P. and Warshaw, P.R. (1989). User Acceptance of Computer Technology: A Comparison of Two Theoretical Models. Management Science, 35, Davis, F.D. & Venkatesh, V. (1996). A critical assessment of potential measurement biases in the technology acceptance model: three experiments. International Journal of Human-Computer Studies, 45, Fishbein, M. & Ajzen, I. (1975). Belief, Attitude, Intention and Behaviour: An Introduction to Theory and Research. Reading, MA: Addison-Wesley. Gardner, C. & Amoroso, D.L. (2004). Development of an instrument to measure the acceptance of internet technology by consumers. Paper presented at Proceedings of the 37 th Hawaii International Conference on System Sciences. Gefen, D. & Straub, D.W. (2002). Managing User Trust in e-services. e-service Journal. 2(1). Gefen, D., Karahanna, E., & Straub, D.W. (2003). Trust and TAM in Online Shopping: an Integrated Model. MIS Quarterly. 27(1) Gronlund, A & Horan, T.A. (2004). Introducing e-gov: History, Definitions, and Issues. Communications of the Association for Information Systems Hair, J., Anderson, R., Tatham, R., & Black, W. (1998). Multivariate Data Analysis. New Jersey: Prentice Hall. Hosmer, L.T. (1995). The Connection Link between Organizational Theory and Philosophical Ethics. Academy of Management Review. 20(3) Jarvenpaa, S.L. & Tractinsky, N. (1999). Consumer Trust in an Internet Store: A Cross-Cultural Validation. Journal of Computer-Mediated Communication. 5(2). Kelman, H.C. (1958). Compliance, Identification, and Internalization: Three Processes of Attitude Cahange?. Journal of Conflict Resolution, 2, Klopping, I.M. & McKinney, E. (2004). Extending the technology acceptance model and task-technology fit model to consumer e-commerce. Information Technology, Learning, and Performance Journal, 22(1). 15

16 Li, Y.N., Tan, K.C. & Xie, M. (2002). Measuring Web-based Service Quality. Total Quality Management, 13(5), Malhotra, Y. & Galletta, D.F. (1999). Extending the Technology Acceptance Model to Account for Social Influence: Theoretical Bases and Empirical Validation. Paper presented at the 32nd Annual Hawaii International Conference on System Sciences. Hawaii. Mathieson, K. (1991). Predicting User Intentions: Comparing the Technology Adoption Model with the Theory of Planned Behavior. Information Systems Research, 2, McKnight, D.H., Choudhury, V., & Kacmar, C. (2002). Developing and Validating Trust Measures for e- Commerce: an Integrative Approach. Information Systems Research. 13(3), Moore, G. & Benbasat, I. (1991) Development of an Instrument to Measure the Perceptions of Adopting an Information Technology Innovation. Information Systems Research, 2, Niemi, R.G., Craig, S.C., & Mattei, F. (1991). Measuring Internal Political Efficacy in the 1988 National Election Study. The American Political Science Review. 85(4). Parasuraman, A., Zeithaml, V.A. & Berry, L.L. (1985). A Conceptual Model of Service Quality and Its Implications for Future Research. Journal of Marketing. 49(Fall), Parasuraman, A. (2005). E-S-QUAL, A Multiple-Item Scale for Assessing Electronic Service Quality. Journal of Service Research, 7(3), Parent, M., Vandebeek, C.R., & Gemino, A.C. (2004). Building Citizen Trust through e-government. Paper Presented at Proceedings of the 37th Hawaiian International Conference on Systems Sciences. Pavlou, P.A. (2003). Consumer Acceptance of Electronic Commerce: Integrating Trust and Risk with the Technology Acceptance Model. International Journal of Electronic Commerce. 7(3) Phang, C.W., Sutanto, J., Li, Y., & Kankanhalli, A. (2005). Senior Citizens Adoption of E-government: In Quest of the Antecedents of Perceived Usefulness. Paper Presented at Proceedings of the 38th Annual Hawaii International Conference on System Sciences. Rogers, E. M. (1995). Diffusion of Innovations, New York : The Free Press. Rotter, L.B. (1971). Generalized Expectations for Interpersonal Trust. American Psychologist. 26(5), Taylor, S. and Todd, P.A. (1995). Understanding Information Technology Usage: A Test of Competing Models. Information Systems Research, 6(2), Tan, C. W. and Pan, S. L. (2003). Managing e-transformation in the public sector: an e-government study of the Inland Revenue Authority of Singapore (IRAS). European Journal of Information Systems, 12(4), Thomas, C.W. (1998). Maintaining and Restoring Public Trust in Government Agencies and Their Employees. Administration and Society, 30(2), Venkatesh, V. and Davis, F.D. (2000). A Theoretical Extension of the Technology Acceptance Model: Four Longitudinal Filed Studies. Management Science, 46(2), Venkatesh, V. and Morris, M.G. (2000). Why Don t Men Ever Stop To Ask For Directions? Gender, Social Influence, And Their Role In Technology Acceptance And Usage Behavior. MIS Quarterly, 24(1), Warkentin, M. Gefen, D., Pavlou, P., & Rose, G. (2002). Encouraging Citizen Adoption of e-government by Building Trust. Electronic Markets. 12(2) Welch, Eric W. & Charles C. Hinnant. (2003). Internet Use, Transparency and Interactivity Effects on Trust in Government. Paper presented at the 36th Annual Hawaii International Conference on System Sciences. Hawaii. Wu, I. L. & Chen, J. L. (2005). An Extension of Trust and TAM Model with TPB in the Initial Adoption of Online Tax: An Empirical Study. International Journal of Human-Computer Studies, 62, Zeithaml, V.A., Parasuraman, A., & Malhotra, A. (2002). Service Quality Delivery Through Web Sites: A Critical Review of Extant Knowledge. Journal of the Academy of Marketing Science. 30(4), Zimmerman, M.A. (1989). The Relationship between Political Efficacy and Citizen Participation: Construct Validation Studies. Journal of Personality Assessment, 53(3), Zucker, LG. (1986). Production of trust: Institutional sources of economic structure. Research in Organizational Behaviour, 8,