A Study of Taxpayers' Intention in Using E-filing System: a Case in Labuan F.t S

This study used the Technology Acceptance Model (TAM) to examine taxpayers' intention in using e-Filing system. Data are collected from three higher learning institutions' staffs particularly in Labuan F.T. The purpose of this study is to determine the relationship between taxpayers' intention to use e-Filing towards attitude, perceived usefulness, perceived ease of use, information system quality, information quality and perceived credibility of the system. Further, this study also examines critical determinant (attitude, perceived usefulness, perceived ease of use, information system quality, information quality and perceived credibility of the system) in TAM that influence most taxpayers' intention. This study has replicated few methods from previous studies. This study is analyzed with reliability analysis, correlation analysis and Standardized Regression Weight (using Structural Equation Modelling). The result confirms a strong relationship between TAM determinants and taxpayers' intention. Consequently, this paper will assist Inland Revenue Board (IRB) to improve their Internet e-Filing system. This in turn, will be useful for them to educate Malaysian taxpayers in order to improve their attitude using e-Filing as their mode to file tax return undoubtedly.


Introduction
TAM was developed to explain computer usage.The goal of TAM is to provide an explanation of the determinants of computer acceptance that is capable of explaining user behavior across a broad range of end user computing technologies and user populations, while at the same time being both economical and theoretically justified ( Davis, 1989).

Malaysian E-Filing
Previously, the tax return is manually processed.At present, Inland Revenue Board (IRB) shift to a new paradigm towards e-Filing due to Self Assessment System (SAS) and will focus more on audit field.IRB also decided to aim for paperless.E-Filing process is more convenience, fast, accurate and secured in terms of payments.There are four steps in e-Filing.First, taxpayers need to enroll and verify a digital signature and MyKad into the reader.Then, they are required to enter the gross earnings, relief and deductions before the system compute automatically.After that, IRB will receive the tax form electronically and email verification of tax form return will be sent to taxpayers.The most important aspect in e-Filing system is security.As stressed out by IRB CEO, the e-Filing system is secure and difficult to get into anyone's personal tax file because they need to enter PIN (personal identification number) and a password (The Star Online, April 27 th , 2007).According to IRB, about 657,000 taxpayers filed their Self-Assessment System (SAS) for year 2006 through the e-Filing system compared to year 2005 which were only 189,048 taxpayers (Berita Harian, June 26 th , 2007).Thus, the number of taxpayers used this e-Filing system as a method of filing tax return increased by 247%.

Problem Statement
It seems clear that there is lack of study that has been conducted in the area of TAM among e-Filing taxpayers', particularly in Labuan F.T. Thus, the study aims to contribute to the knowledge of information technology.Due to the problem on using e-Filing, taxpayers' were uncomfortable with e-Filing as they were unfamiliar with electronic transactions and some said they were not computer savvy (The Star, May 4 th , 2006).In addition, most of taxpayers' were very concerned if IRB directly changes the whole manual tax return process to e-Filing because of their inability to use Internet and less computer skill.Bedsides, slow response to e-Filing was mainly because of people's habit of doing their assessment at last minute.Some of them are difficult to accept a new technology since they are very concern about the security.

Significant of the Study
Through this study, the main reason of taxpayers' intention in using e-Filing will be explored.In addition, this study will assist Lembaga Hasil Dalam Negeri (IRB) to improve e-Filing performance according to TAM determinants and in line with the Government's Information Technology Policy.Based on the result of this study, it is expected there are more trainings and seminars might be conducted in order to improve e-Filing usage and compliance.However, this system will be fully implemented after it is widely accepted (Rahimah Abdullah, 2006).Besides, the IRB chief executive officer and director-general have stated that the process of upgrading e-Filing system services for taxpayers will improve the level of taxpayers' compliance.Consequently, the intention of taxpayers will be improved positively and in turn it will increase their compliance toward the IRB.

Objective of the Study
The objective of this study is twofold: To determine the relationship between taxpayers' intention to use e-Filing towards attitude, perceived usefulness, perceived ease of use, information system quality, information quality and perceived credibility of the system.
To examine critical determinant (attitude, perceived usefulness, perceived ease of use, information system quality, information quality and perceived credibility of the system) in TAM that contributes most to influence taxpayers' intention.
The remainder of this paper is organized as follows.A review of related literature on technology acceptance model and research questions is discussed.Next, the methodology employed in this study, research instruments used and data analysis method involved are described.Finally, the empirical results and discussion of the study are drawn

Technology Acceptance Model
The TAM adopts the theory of reasoned act (TRA) model to explore the IT acceptance.TAM and TRA, both of which have strong behavioral elements, assume that when someone forms an intention to act, they will be free to act without limitation (Davis. et al., 1989).In addition, Davis et al., (1989) also stated that TAM indicates both perceived usefulness (PU) and perceived ease of use (PEOU) as key independent variables that determine or influence potential users' attitudes toward IT intention of use.
This study also used DeLone and McLean model of information system success (2003) consists of information system quality (ISQ) and information quality (IQ).Another new dimension is perceived credibility of a computer system developed by Wang (2002) and Chang et al. (2005).
Previously, Wang (2002); Chang et al. (2005);Hung. et al., (2006);and Fu et al., (2006) applied TAM in their study on tax filing methods especially in Taiwan.However, most of researchers construct hypothesized affect the use of Internet tax filing indirectly through their affect on perceived usefulness (PU), perceived ease of use (PEOU), information system quality (ISQ), information quality (IQ), and perceived credibility (PC) toward attitudes of using (ATT) and behavior intention (BI).
However, there has been little study on TAM and DeLone and McLean model particularly in e-Filing system.Nevertheless, Lai et al. (2004) examined the level of technology readiness of Malaysian tax practitioners and their usage intention towards an e-Filing system.They found a significant positive relationship between the level of technology readiness and the usage intention towards the e-Filing system.Besides, Hanudin et al. (2006) has applied TAM particularly on the intention to use the SMS as a mode for banking transactions.According to this study, perceived expressiveness, perceived usefulness and perceived ease of use are important determinants of intention to use SMS banking among male respondents.

Attitude
Attitude is defined in terms of individual preferences and interests regarding the use of Internet tax-filing system.The measurement is adapted from Davis et al. (1989).Attitude is one of important determinant in increase the level of behavior intention among taxpayers.This can be supported by Chang et al. (2005) that stated as attitude has a significant impact on behavior intention (BI) of using the system.

Perceived Usefulness
Perceived usefulness is defined as the degree of taxpayers' believes from using Internet tax-filing system that would enhance their job performance and the measurement adapted from Davis (1989).In addition, Davis (1989) has stated that perceived usefulness was found to have a strong influence on people's intentions.However, Chang et al.(2005) study has found that perceived usefulness has no direct impact on behavior intention but has significant on attitude, which consequently has an impact on behavior intention of using the system.

Perceived Ease of Use
Perceived ease of use was defined as the degree to which a user aspects the use of Internet tax-filing system to be free of effort and was measured by Davis (1989).In Davis (1989), perceived ease of use which test to had a smaller but significant affect that subsided over time.According to Chang et al., (2005), perceived ease of use also found to have a significant impact on attitude, thus affects behavior intentions.

Information System Quality
According to DeLone and McLean (2003), information system quality is associated with the issue of whether the technical components of delivered is provide the quality of information and service required by stakeholders.Besides, information system quality was defined by the degree to which the technical components of Internet tax-filing provide the quality information and service required by users (Chang et al., 2005).

Information Quality
Based on Chang et al. (2005), information quality has been defined by the degree to which users are provided with quality information regarding their needs.Information quality also represents the users' perception of the output quality generated by an information system and includes such issues as the relevance, timeliness and accuracy (DeLone and McLean, 2003).

Perceived Credibility
Perceived credibility is defined as the extent of users' confidence in the Internet tax-filing system's ability to protect the user's personal information and security.This measurement was adapted from Wang (2002).According to Chang et al. (2005), a credible website needs to safeguard personal information from unauthorized access or disclosure, accidental loss and alteration or destruction.In Lai et al. (2004) study, some of the respondents specifically expressed that they would only use the e-Filing system if the IRB could assure them that the e-Filing system were safe and secure, and if the usability and reliability of the e-Filing system were fully tested and well documented.

Research Questions
RQ1: How strong the six TAM determinants influence taxpayers' intention?RQ2: What are critical determinants in TAM those contribute most to influence taxpayers' intention?

Study Sampling Procedure
The population of this study covered e-Filing users from three higher learning institutions in Labuan F.T. Before data collections were carried out, phone calls were made to each institution in order to ensure total number of academic and administration staffs.Respondents consist of experience and non-experience individual taxpayers from Universiti Malaysia Sabah (UMS), Institut Latihan Perindustrian (ILP) and Pusat Matrikulasi Labuan (PML).Questionnaires were distributed during January until April 2007.This is due to the fact that individual taxpayers need to submit their e-Filing return before April 30 th 2007.We have distributed 80 questionnaires for every institution and the total samples are 240 respondents.Finally, only 100 respondents completed and returned the questionnaires, which represents about 42.0% response rate.

Instrumentation
The questionnaire has two sections namely TAM determinants and demographic section.The instrument of this study is based on Chang et. al (2005).It presents a new set of instrument with some modification according to seven variables using seven-point Likert scale.The variables consist of behavior intention, attitude, perceived usefulness, and perceived ease of use (Davis, 1989); information system quality and information quality (DeLone and McLean, 1992) and perceived credibility (Wang, 2002).Demographic section consists of gender, education level, job, time of computer using, experience of handling e-Filing and experience of learning e-Filing.
In addition, about 35% of respondents claimed that they used computer more than 28 hours per week.Most of respondents do not have an experience in handling and learning e-Filing system.About 33% of them have experience in handling e-Filing system and 36% have learning experience through IRB courses and seminar.
Table 2 presents the Reliability Analysis, Cronbach's Alpha reliability coefficients of the six determinants and behaviour intention (dependent variable).All the determinants were all above 0.7.It seems that this study provides quite reliable instruments because the score is higher than Chang (2005).For example, the behaviour intention is 0.96 as compared to 0.94; attitude = 0.96 (0.94); and perceived credibility = 0.80 (0.70).Perceived credibility is a new determinant and presents more reliable compared to previous study.Reliability less than 0.6 is considered poor, those in the 0.7 ranges, acceptable, and those 0.8 good (Sekaran, 2000).It is of evidence that the Cronbach's alpha value for the seven factors in this study ranged from 0.8 to 0.97.Therefore, the internal consistency reliability of the measures used in this study can be considered to be good.

Correlation Analysis
A correlation analysis in Table 3 indicates that all the six TAM determinants (i.e.attitude, perceived usefulness, perceived ease of use, information system quality, information quality and perceived credibility of the system) are positively correlated.Each construct shares greater variance with its own block of measures than with the other constructs representing a different block of measures.

Structural Equation Modeling
In order to answer the second objective of this study which is to examine the critical determinant (attitude, perceived usefulness, perceived ease of use, information system quality, information quality and perceived credibility of the system) in TAM that contributes most to influence taxpayers' intention tu use e-Filling system, advanced statistical technique know as Structural Equation Modeling (SEM) was utilized.
SEM is a versatile statistical technique that is particularly useful for analyzing nonexperimental data (Byrne, 2001).It has become an increasingly popular data-analytic technique in psychology, counseling, and rehabilitation.Quintana and Maxwell (1999) highlighted several applications of SEM to research, including the use of SEM for testing for mediational relationships, interaction effects, and mean differences; for confirmatory factor analysis and multiple sample analysis; for longitudinal designs; and for handling missing data.Recent innovations have allowed SEM to become a broad data-analytic framework with flexible and unique capabilities.Furthermore, SEM involves an analysis of carefully defined a priori hypotheses about the relationships among both measured and latent variables.It is imperative for researchers to become familiar with this data-analytic technique so that they can use this technique in their research endeavors.It is equally important for practitioners to become familiar with SEM to make judicious assessments of published studies.

Model-estimation
Analysis of Moment Structure (AMOS) Version 5 was used to estimate the model using SEM with observed variables.Recognition of the reliability of AMOS computations has been established by its increasing use in published studies in reputable journals over the last few years (e.g.Zuroff et al., 1999).Prior to model estimation, each of the multi-item constructs were transformed into totalled scores using equally weighted scales developed from the results of the CFA.This path analytic procedure was used due to the complexity and difficulty of using a full structural equation model.For a similar use of this technique, see Li and Calantone (1998, p. 88) and the references cited by these authors to justify this approach.

Model Testing Results
The structural model was assessed by using established measures and evaluative criteria for model fit.Several goodness-of-fit indices are commonly used to evaluate how well the structural model fits the data.The chi square goodness-of-fit test is one of the most commonly used indices.In SEM, a nonsignificant chi square value is an indication that the hypothesized model has a good fit with the data.The problem with using chi square, however, is that it is hypersensitive to sample size (Ullman, 2001).Because SEM is grounded in large-sample theory, finding well-fitted hypothesized models, where the chi square value approximates the degrees of freedom, has proven unrealistic, leading SEM methodologists to develop additional practical or ad hoc indices of fit.
One approach is to divide the chi square (χ 2 ) value by the degrees of freedom.According to Carmines and Mclver (1981), χ 2 /df ratios in the range of 2:1 or 3:1 indicated an acceptable fit between the hypothetical model and the sample data.The most popular alternative measures of fit for SEM analysis, however, are the goodness-of-fit index (GFI), the normed fit index (NFI), the comparative fit index (CFI), and the root mean square error of approximation (RMSEA).The GFI, NFI, and CFI all have values ranging from 0 to 1; a good fit is indicated by values greater than .90for GFI and NFI and .95 and greater for CFI.For RMSEA, a value of 0 is interpreted as an exact fit; values less than .05are a close fit, values between .05 and .08 are a fair fit, values between .08 and .10 are a mediocre fit, and values more than.10 are a poor fit.Regarding the precision of the RMSEA estimates, AMOS reports a 90% confidence interval around the RMSEA value.MacCallum, Browne, and Sugawara (1996) indicated that a small RMSEA and a very narrow confidence interval suggest good precision of the RMSEA value in reflecting model fit in the population.Finally, Martens (2005) indicated that chi square/df, GFI, and NFI tend to be substantially affected by sample size and number of indicators per factor and do not generalize well across samples.Marten (2005) recommended using CFI and RMSEA as the primary goodness-of-fit indexes.
The results suggest that the data fit the current conceptual model well, with a χ 2 of 9.158 (df =6, p =0.165), χ 2 /df =1.526, GFI =0.975, CFI =0.996, NFI =0.987, and RMSEA =0.073.Moreover, the squared multiple correlation for the predictors of INT is 0.701, which shows that the variables included in the model explain 70.1 per cent of the variance in the outcome variable.In other words, the error variance of INT is approximately 29.9 percent of the variance of INT itself.

Hypotheses Testing
The current study proposed to test six hypotheses in identifying the critical determinant in TAM that contribute most to influence taxpayers' intention.Details of the hypotheses are stated below: H1: Attitude of the taxpayers has significant influence on taxpayers' intention to use e-Filing H2: Perceived usefulness of the system has significant influence on taxpayers' intention to use e-Filing H3: Perceived ease of use of the system has significant influence on taxpayers' intention to use e-Filing H4: Information system quality has significant influence on taxpayers' intention to use e-Filing H5: Information quality of the system has significant influence on taxpayers' intention to use e-Filing H6: Perceived credibility of the system has significant influence on taxpayers' intention to use e-Filing The results of the hypotheses testing are accessible in Table 4.It can be clearly seen in the table that from a total of six hypotheses stated above, significant relationship was found in Hypothesis 1 between the influence of attitude of the taxpayers on their intention to use e-Filing (ß=0.605,C.R=5.955, p=0.000).By comparing the standardized path coefficient of this construct among the other proposed hypotheses, attitude construct lead the list.Therefore, it is certainly become the most critical determinant in TAM that contributes most to influence Malaysian taxpayers' intention to use e-Filing.This infers that taxpayers have positive and strong attitude and interest to use e-Filing.The upside to e-Filing is that it's greatly reducing the volume of paperwork.New technology allows a paperless filing system and provides taxpayers with an electronic version for their files as well.A paperless process saves paper, toner, and file storage costs.Another benefit of e-Filing was that it enabled e-filers to be more productive, presumably because it saves on paperwork costs, makes it easier to correct errors, and is quicker than filing on paper.Further, e-Filing reduced number of errors, attributable to the one-time entry of figures and the checks performed by preparation software.
Next, Hypothesis 2 avers that perceived usefulness of the system has significant influence on taxpayers' intention to use e-Filing.Table 4 and Figure 1 illustrate that the effect of perceived usefulness of the system on taxpayers' intention to use e-Filing was not recognized (ß=0.026,C.R=0.223, p=0.823).Malaysian taxpayers' have less intention to use e-Filing because of the downside they perceived to obtain such as using an electronic technology that they may not thoroughly understand can be daunting.At the same time, the idea of a paper-based tax return over which they have more control is more comfortable.They still prefer to go to the basics where traditional way to fill up form using pen and paper.
Further, null hypothesis of Hypothesis 3 asserts that the extent of perceived ease of use of the system has significant influence on taxpayers' intention to use e-Filing.The comparison of the effect as reflected in Table 4 evinced that perceived ease of use of the system (ß=0.059,C.R=0.502, p=0.616) had insignificant relationships on taxpayers' intention to use e-Filing.Thus, Hypothesis 3 of perceived ease of use of the system has significant influence on taxpayers' intention to use e-Filing was not established.The present study provides evidence of less effect of perceived ease of use of the system in relation to Malaysian taxpayers' intention to use e-Filing.There are Malaysians taxpayers who are still reluctant to use e-Filing to reveal tax transactions though they can e-file any time of the day or night, and both complex and simple returns can be filed electronically.Most probably, some taxpayers feel that this e-Filing is very complex and they need to take a longer time to accustom with the system.Meanwhile, Hypothesis 4 states that information system quality has significant influence on taxpayers' intention to use e-Filing.Inspection of the structural loading of the standardised path coefficients results as exemplified in Table 4 showed insignificant relationship between information system quality and taxpayers' intention to use e- Filing (ß=0.189,C.R=1.139,p=0.255).Thus, Hypothesis 4 of information system quality has significant influence on taxpayers' intention to use e-Filing was not acknowledged.However, it should be noted that information system quality is an important construct to influence Malaysian taxpayers to use e-Filing as it has significant influence with perceived usefulness, perceived ease of use, attitude of taxpayers, information quality and perceived credibility of the system with p-values of 0.000 (see Table 4).It is clearly understood that e-Filing can be done at any time from a personal computer.The taxpayer able to get his/her refund quicker or schedule payment to meet the deadline.Indeed, Malaysian taxpayers found that the data transmitted nearly instantaneously to the clearinghouse and then to the system, nothing needs to be scanned computers because the data is transmitted directly into them.However, there still a bigger drawback to doing own taxes is that taxpayer might make errors and/or overlook legitimate deductions.Taxpayers also face risk making mathematical errors.It is advisable to check arithmetic carefully.Even if use a computer, inputting errors occur, thus check each field carefully to make sure figures correspond to withholding and expense records.Occasionally, glitches also show up even in reputable tax programs.
The hypothesis about information quality of the system has significant influence on taxpayers' intention to use e-Filing (Hypothesis 5) was not confirmed (ß=-0.026,C. R=-0.237, p=0.812).Certainly, barriers to adoption of e-Filing perceived by Malaysian taxpayers includes a continued preference by taxpayers and certain segments of tax practitioners for paper filing, lack of awareness of e-Filing and how to do it and concern about privacy, security and the role of third parties in the process.
Hypothesis 6, the final hypothesis tested in this study, hypothesized that perceived credibility of the system has significant influence on taxpayers' intention to use e-Filing.Surprisingly, the result inferred that the p-value is greater than 0.05 (ß=0.033,C.R=0.426, p=0.670).Thus, the proposed relationship is not significant.Malaysian taxpayers perceived that it takes a lot of time to test software compatibility on network computer systems, develop training programs, and convince users that they should use a new system when their old system works just fine and they have yet to see any benefit in changing.Overall, they found that using the electronic filing is yet to be compatible with their lifestyle.

Concluding Remarks
This study is assured to have strong reliable determinant to assess taxpayers' intention in using e-Filing as a medium to file their tax return.This is based on the result of Cronbach Alpha that performed by reliability analysis for six independent variables and one dependent variable.The empirical results of our study can provide support for Davis (1989), DeLone andMcLean (1992), Wang (2002), andChang (2005) models.
Furthermore, this study proves strong and positive relationship between TAM determinants with taxpayers' behaviour intention.The intention of taxpayers seems to be influenced by attitude (0.823), perceived usefulness (0.724), perceived ease of use (0.691), information system quality (0.751), information quality (0.639) and perceived credibility (0.545).From the result, attitude seems to play strong role to influence taxpayers' intention to file tax return in future.This study suggests that taxpayers' attitude can be changed based on their first experience of handling e-Filing system.Furthermore, taxpayers' experience is influenced by their perceived usefulness, perceived ease of use, information system quality, information quality and perceived credibility to file tax return.

Implication and Future Research
Some of these barriers can be addressed through education and marketing of the advantages of e-Filing, such as faster refunds, electronic receipts that offer proof of filing, convenience, accuracy and reduced likelihood of receiving a notice from the system provider.The result of this study is able to assist Inland Revenue Board (IRB) to improve their e-Filing system and implement e-Filing seminar or on-hand course for taxpayers.This in turn, will also increase taxpayers' compliance toward their responsibility as a Malaysian resident.However, Inland Revenue Board (IRB) need to implement useful activities and programmes to educate taxpayers to use e-Filing.In that case, taxpayers will believe this medium is more useful compared to fill in the form and submit to IRB office manually before the due date.They also might realize e-Filing will process their work faster and more easier as compared their experience before.Once taxpayers' attitude change, the five determinants of TAM will also influence taxpayers' intention.In addition, the intention of taxpayers to file tax return in future might also influence the level of tax compliance.
This study is an exploratory study of TAM particularly in e-Filing system.However, it is quite difficult to be generalised due to small sample sizes, which are only 100 public servants at Labuan.In future, the sample of the study should consider both public and private sectors' staffs in Malaysia.This is because the result can be generalized to Malaysian taxpayers as a whole.
(18th April 2007).New tax assessment system to be in place next year.New Strait Times.  .

(
27 th April 2006).Good response to E-Filing.The Star Online.(4 th May 2006).Taxpayers need to switch to e-Filing next year, say IRB.The Star.(3 rd May 2006).Manual Income Tax Return Are Acceptable Next Year.Bernama.

Table 1 .
Demographic Data