factors influencing knowledgeable consumers' level of - USM

FACTORS INFLUENCING KNOWLEDGEABLE CONSUMERS’ LEVEL OF

ACCEPTANCE OF GOODS AND SERVICES TAX

SHALENE KALYANASUNDARAM

Research report in partial fulfilment of the requirements for the

Degree of Master of Business Administration

Universiti Sains Malaysia

2015

This thesis is dedicated to my Papa, S.K. Sundaram who is my sturdiest pillar of

support and who inspired my interest in GST.

I can never thank you enough.

ii

ACKNOWLEDGEMENT

First and foremost, I am grateful to God for being able to conduct a research in an

area of much interest to me and for surrounding me with the best of people who have

supported me endlessly throughout this challenging journey. I am immensely indebted to my

mentor, Professor Dato’ Dr Hasnah Haron for her impeccable guidance, support, motivation,

approachability and constructive feedback without which this research would not have been

possible. Her diligence and dedication was highly inspiring and words cannot express the

honor and gratitude I feel to have worked with her.

I would also like to thank respected professors, namely Associate Professor Dr

Jayaraman and batch-mates who were so generous with their time and knowledge in assisting

me throughout the duration of this research paper. Their kind assistance and valuable

feedback enabled the successful completion of this paper. I would also like to take this

opportunity to thank my good friends; Bong-Arkya, Suria Chetta, Deepa, Thivagar and

Ameet from the bottom of my heart as they were my pillars of support and encouragement as

we travelled through this journey, spending days and nights together to finish this paper.

I would also like to thank my respondents who have participated in the study. I would

not have been able to contribute this study without their assistance. They have played a key

role in this research paper and I would like to pass my sincere gratitude to them.

Last but not at all least, I would like to thank my ever supporting parents, siblings,

employers, colleagues, friends, FREAKSisters and devoted fiancé for their constant show of

support, understanding, love and encouragement. Everyone played an important part in

easing the challenges I had to face throughout this period. Without this amazing support

system, this research paper would not have been a reality.

iii

TABLE OF CONTENTS

Page

ACKNOWLEDGEMENT…………………………………………………………….ii

TABLE OF CONTENTS……………………………………………………………..iii

LIST OF TABLES………………………………………………………………….....ix

LIST OF FIGURES…………………………………………………………….……..xi

ABSTRAK……………………………………………………………………….......xiii

ABSTRACT…………………………………………………………………………..xiv

LIST OF ABBREVIATION………………..…………………………………………xv

CHAPTER 1 INTRODUCTION………………………………………………………1

1.0 Introduction…………………………………………………………………........1

1.1 Background of the Study…………………………………………………………1

1.2 Problem Statement………………………………………………………………..3

1.3 Research Objectives ……………………………………………………………...7

1.4 Research Question……………………………………………………………......7

1.5 Definition of Key Terms …………………………………………………………8

1.6 Significance of the Study………………………………………………..………..9

1.6.1 Practical Implication…..………………………………………………...……..9

1.6.2 Theoretical Implication……………………………………………………….11

1.7 Organization of the Remaining Chapters……………………………………….12

iv

CHAPTER 2 LITERATURE REVIEW………………………………………..…….13

2.0 Introduction……………………………………………………………………13

2.1 Malaysian Taxation……………………………….……………………………13

2.2 Consumers………………………..……………………………………………14

2.3 GST…………………………………………………………………………….16

2.3.1 Reasons for Implementing GST……………………………………………..16

2.3.2 Significance of GST…………………………………………………………18

2.3.3 Mechanism of GST……………………………………………………….....19

2.3.4 GST rate in the ASEAN countries…………………………………………..24

2.3.5 Implementation of GST……………………………………………………..25

2.4 Theory…………………………………………………………………………26

2.4.1 Theory of Planned Behavior (TPB)………………..……………………….26

2.5 Dependent Variable……………………………………………………………28

2.6 Antecedent and Independent variables ………………………………………..29

2.6.1 Attitude………………………………………………………………………29

2.6.2 Rule Observance Behavior……………………………………………..…....31

2.6.3 Perception of fairness of GST…………………………………………..…...32

2.6.4 Subjective Norm……………………………………………………………..34

2.6.5 Perceived behavioral control (PBC)…………………………………………35

2.6.6 Level of GST Knowledge………………………………………………...….36

2.6.7 Self-Efficacy…………………………………………………………………37

2.7 Theoretical Framework………………………………………………………...39

2.8 Factors Affecting Level of Acceptance of GST………………………………..41

2.8.1 Studies on Goods & Services Tax…………………………………………...44

2.8.2 Studies on Tax compliance……………………………………………….….44

v

2.8.3 Studies on Intention to Accept…………………………………………….....44

2.8.4 Other Studies……………………………………………………………..…..45

2.9 Hypotheses Development………………………………………………………46

2.9.1 Rule observance behavior and attitude ………………………………………46

2.9.2 Perception of Fairness of GST and attitude…………………………………..46

2.9.3 Level of GST Knowledge and Perceived Behavioral Control………………..47

2.9.4 Self-efficacy and Perceived Behavior Control ……………………………….48

2.9.5 Attitude and Level of Acceptance…………………………………………….50

2.9.6 Subjective Norms and Level of Acceptance………………….……………….51

2.9.7 Perceived Behavioral Control and Level of Acceptance……………………...52

2.10 Summary……………..………………………………………………………..53

CHAPTER 3 RESEARCH METHODOLOGY………………………………………55

3.0 Introduction……………………………………………………………………...55

3.1 Research Design………………………………….……………………………...55

3.1.1 Type of Study……………………………………………………………..…..55

3.1.2 Population and Unit of Analysis……………………………………………...56

3.1.3 Sample Size…………………………………………………………………..58

3.1.4 Sampling Method…………………………………………………………….59

3.1.5 Data Collection Techniques………………………………………………….59

3.1.6 Survey Instrument…………………………………………………………....60

3.1.7 Questionnaire Design………………………………………………………...61

3.2 Measurements of Variables…………………………………………..…….…..63

3.2.1 Measurements of Antecedent Variables and Independent Variables………..63

3.2.2 Level of Tax of Knowledge Scoring………………………………………...64

3.2.3 Measurement of Dependent Variable………………………………………..65

vi

3.3 Pre Test…………………………………………………………………...……67

3.4 Pilot Test….……………………………………………………………………68

3.5 Data Analysis………………………………………………………….……….71

3.5.1 Structured Equation Modeling (SEM)………………………………………71

3.5.2 Covariance-based SEM (CBSEM) against Variance-based SEM

with Partial Least Squares (PLS)……………………………………………72

3.5.3 Statistical Analysis…………………………………………………………..74

3.5.4 Demographic Analysis………………………………………………………74

3.5.5 Common Method Bias………………………………………………………74

3.5.6 Descriptive Analysis………………………………………………………...75

3.5.7 Goodness of Data Test………………………………………………………76

3.5.8 Factor Analysis……………………………………………………………....76

3.5.9 Validity Analysis…………………………………………………………….77

3.5.10 Reliability Analysis………………………………………………………….78

3.5.11 Goodness-of-Fit (GoF)………………………………………………………79

3.6 Summary……………………………………………………………………….79

CHAPTER 4 RESULTS…………………………………………………………….81

4.0 Introduction……………………………………………………………………81

4.1 Response Rate…………………………………………………………………82

4.2 Profile of Respondents…………………………………………………….…..83

4.3 Descriptive Statistics and Multicollinearity ………………………………..…86

4.4 Common Method Bias using Principal Component Analysis ……...…………90

4.5 Data analysis and results ………………………………………………….…..91

4.5.1 Goodness of measures……………………………….………………….......91

4.5.2 Pathway of Each Model Construct………………………………………….92

vii

4.5.3 Construct validity…………………………………………………………...93

4.5.4 Convergent validity…………………………………………………………93

4.5.5 β value and R Square………………………………………………………..94

4.5.6 Discriminant validity………………………………………………………..95

4.5.7 SEM Model…………………………………………………………………95

4.5.8 Results of each model………………………………………………………96

4.5.9 Reliability Analysis………………………………………………………..110

4.5.10 Summary on Validity and Reliability of Models………………………….111

4.5.11 Summary of Hypothesis Testing…………………………………………..113

4.5.12 Goodness-of-Fit……………………………………………………………114

4.6 Summary……………………………………………….…………………….115

CHAPTER 5 DISCUSSIONS AND CONCLUSIONS……………………………116

5.0 Introduction…………………………………………………………………..116

5.1 Recapitulation of the Study Findings………………………………………...116

5.2 Discussion…………………………………………………………………….119

5.2.1 What is the GST acceptance level amongst Malaysians?.............................120

5.2.2 What is the relationship of rule observance behavior and perceptions

of tax fairness with attitude of consumer?...........................................................123

5.2.3 What is the relationship of Self – Efficacy and Level of Tax Knowledge

with Perceived Behavior Control?.........................................................................125

5.2.4 What is the relationship of Attitude, Subjective Norms and

Perceived Behavioral Control with Level of Acceptance? .............................128

5.3 Summary and findings…….……………………………………………………130

5.4 Implications...……………………………………………………………….….131

5.4.1 Theoretical Implication……………………………………………………...131

viii

5.4.2 Practical Implication………………………………………………………...133

5.5 Limitation………………………………………………………….…………...138

5.6 Suggestion for Future Research ………………………………………………..138

5.7 Conclusion……………………………………………………………………...140

REFERENCES……………………………………………………………………….142

APPENDIX A – ECONOMIC REPORT 2010-2014- FEDERAL GOVERNMENT

REVENUE ……………………………………………..................167

APPENDIX B – ECONOMIC REPORT 2010-2014- FEDERAL GOVERNMENT

FINANCE ……………………………………………..................168

APPENDIX C – RESULTS OF MERDEKA CENTRE SURVEY ON GST....……169

APPENDIX D – QUESTIONNAIRE……………………………………………….176

APPENDIX E – PLS ALGORITHM REPORT (PILOT STUDY)………………….187

APPENDIX F – PLS ALGORITHM REPORT……………………………………..188

APPENDIX G – BOOTSTRAPPING PLS REPORT……………………………….199

APPENDIX H – IBM SPSS STATISTICS REPORT……………………………….203

ix

LIST OF TABLES

Page

Table 2.1 Key Studies on Malaysian Goods and Services Tax 42

Table 3.1 Profile of Contacts 60

Table 3.2 Summary of Questionnaire Sections 63

Table 3.3 Sources of measurement of variables 66

Table 3.4 Details of Models for analysis 69

Table 3.5 Cronbach’s Alpha for pilot study 70

Table 4.1 Summary on the Response Rate 82

Table 4.2 Socio-Demographic table – Respondent’s profile 83

Table 4.3 Descriptive Statistics and Correlation Matrix (n=228) 86

Table 4.4 Total Variance Explained 91

Table 4.5 Measurement model of VB-SEM- Model 1 97

Table 4.6 Discriminant Validity- Model 1 98

Table 4.7 Summary of PLS Results – Direct Effects- Model 1 99

Table 4.8 Measurement model of VB-SEM-Model 2 101

x

Table 4.9 Discriminant Validity- Model 2 101

Table 4.10 Summary of PLS Results – Direct Effects-Model 2 103

Table 4.11 Measurement model of VB-SEM-Model 3 105

Table 4.12 Discriminant Validity-Model 3 107

Table 4.13 Summary of PLS Results – Direct Effects- Model 3 109

Table 4.14 Cronbach’s Alpha and Composite Reliability for Model 1 110



Table 4.17

Summary of Hypotheses Testing

113

Table 5.1

Summary on t-test for Gender and Mode of Study

121

Table 5.2

Descriptive Table on Institutes and Level of Acceptance of GST

121

Table 5.3

ANOVA table of results for institution groups

122

Table 5.4

Descriptive Table on Income Level and Level of Acceptance of GST

122

Table5.5

ANOVA table of results for income level groups

122

xi

LIST OF FIGURES

Page

Figure 2.1 Cascading effect -Sales & Service tax 17

Figure 2.2 Cascading effect - Service tax 17

Figure 2.3 Input and output tax 20

Figure 2.4 Standard rated supply 21

Figure 2.5 Zero-rated supply 21

Figure 2.6 Exempt supply 22

Figure 2.7 GST Mechanism- standard rated 22

Figure 2.8 GST rates in the ASEAN countries 25

Figure 2.9 Theoretical Framework 40

Figure 4.1 PLS Pathway for Model 1 92



Figure 4.4 PLS output for Beta-value and R square value(Model 1) 98

xii

Figure 4.5 PLS output for testing population regression coefficients 100

Figure 4.6 PLS output for Beta-value and R square value – Model 2 102

Figure 4.7 PLS Output for testing population regression coefficients 103

Figure 4.8 PLS output for Beta-value and R square value-Model 3 108

Figure 4.9 PLS Output for testing population regression coefficients 109

xiii

ABSTRAK

Pelaksanaan Cukai Barangan dan Perkhidmatan (GST) pada 1 April 2015 adalah satu

kejayaan bagi Malaysia. Pelaksanaan GST adalah amat penting untuk menyokong

pembangunan negara. Kajian ini telah dibangunkan untuk mengenal pasti faktor-faktor yang

mempengaruhi tahap penerimaan GST oleh pengguna menggunakan’Theory of Planned

Behavior’. Faktor-faktor yang dikaji termasuk Peraturan kelakuan pematuhan, Persepsi

keadilan GST, Aras pengetahuan GST, ‘Self-efficacy’, Sikap, ‘Subjective Norm’ dan

‘Perceived Behavorial Control’. Borang soal selidik telah diedarkan kepada pelajar-pelajar

dalam talian MBA daripada Universiti Sains Malaysia, Universiti KebangsaanMalaysia,

Universiti Malaya dan Universiti Putra Malaysia pada Februari 2015. 228 MBA pelajar

mengambil bahagian dalam kajian ini dan Smart PLS telah digunakan untuk menguji kesahan

dan kebolehpercayaan data dan hipotesis.. Hasil kajian membuat kesimpulan bahawa 6

hipotesis daripada 7 disokong. Tahap penerimaan GST rendah dengan Sikap dan Persepsi

Kawalan Tingkahlaku menjadi pengaruh yang penting. Pematuhan Peraturan Kelakuan dan

Persepsi GST keadilan adalah ‘ántecedent’ yang penting kepada Sikap. Self-efficacy dan

Tahap pengetahuan GST adalah ántecedent’ yang penting kepada ‘Perceived Behavorial

Control’. Kajian menunjukkan bahawa walaupun tahap pengetahuan GST dipertingkatkan di

kalangan pengguna, tahap penerimaan tidak akan bertambah baik tanpa langkah-langkah

berkesan yang diambil untuk meningkatkan persepsi mereka terhadap keadilan GST. Di

samping itu, pemahaman GST tidak akan mencukupi untuk mempengaruhi penerimaan GST.

Pengguna perlu dididik bagaimana untuk menguruskan penggunaan dan kewangan mereka

serta diberikan sokongan yang mencukupi dalam tempoh peralihan ini.

xiv

ABSTRACT

The implementation of Goods and Services tax (GST) on the 1st of April 2015 is a milestone

for Malaysia. The implementation of GST is of vital importance in order to support the

development of the nation. This study was developed in order to identify the factors that

influence knowledgeable consumers’ level of acceptance of GST using the Theory of Planned

Behaviour. Factors examined include Rule observance behavior, Perception of fairness of

GST, Level of GST knowledge, Self-efficacy, Attitude, Subjective Norm and Perceived

Behavioral Control. Questionnaires were distributed online to MBA students of Universiti

Sains Malaysia, Universiti Kebangsaan Malaysia, Universiti Malaya and Universiti Putra

Malaysia in February 2015. 228 MBA students participated in the study and Smart PLS was

used to test the validity and reliability of the data as well as the hypotheses developed.

Findings of the study concluded that 6 hypotheses out of 7 were supported. The acceptance

level of GST was slightly low with Attitude and Perceived Behavioural Control being

significant influencers of the acceptance level. Rule Observance Behaviour and Perception of

GST fairness were significant antecedents to Attitude with perception of fairness being the

variable with higher weightage. Self-efficacy and Level of GST knowledge were acceptable

antecedents to Perceived Behavioral Control with Self-Efficacy carrying the higher

weightage of significance. This implies that although the level of GST knowledge is

enhanced amongst consumers, level of acceptance will not improve without effective

measures taken to enhance their perception of GST fairness. Besides that, the mere

understanding of GST will not be sufficient to influence acceptance of GST. Consumers need

to be educated on how to manage their consumption and finances as well as be given ample

support during this transition period.

xv

LIST OF ABBREVIATIONS

ATT Attitude

AVE Average Variance Extracted

BR1M Bantuan Rakyat 1 Malaysia

CBSEM Covariance-based Structural Equation Modelling

CFA Confirmatory Factor Analysis

CMB Common Method Bias

FAQ Frequently Asked Questions

GST Goods & Services Tax

ITA Income Tax Act

LA Level of Acceptance

LTK Level of Tax Knowledge

PBC Perceived Behavorial Control

PLS Partial Least Squares

PTF Perception of Tax Fairness

RMCD Royal Malaysian Customs Department

ROB Rule Observance Behavior

SE Self Efficacy

SEM Structural Equation Modelling

SN Subjective Norms

SST Sales & Services Tax

TPB Theory of Planned Behavior

xvi

UM Universiti Malaya

UKM Universiti Kebangsaan Malaysia

UPM Universiti Putra Malaysia

USM Universiti Sains Malaysia

VAT Value-Added Tax

VBSEM Variance-based Structural Equation Modelling

1

CHAPTER 1

INTRODUCTION

1.0 Introduction

This chapter presents the research framework of the study. It begins with discussing

the background of the study and the problem statement followed by research

objectives and research question. Definition of key terms of major variables will also

be included to enhance understanding of the concepts. The significance of the study

is the contribution of the scope of study to both practical and theoretical

implications. The scope of study focuses on consumers as the Goods and Services

tax being studied is a consumption tax. Manufacturers, retailers and distributors are

able to set off the tax they pay against the tax charged to their customers but

consumers are unable to claim the GST they pay on their consumption. Towards the

end, the chapter will give a brief overview of the remaining chapters in the thesis.

1.1 Background

Malaysian taxation system is divided into two; direct and indirect taxes. Indirect

taxes are controlled by the Royal Malaysian Customs Department (RMCD) and

consist of four parts; customs duties, excise duty, sales tax and service tax. Direct

taxes are administered by the Inland Revenue Board of Malaysia (IRBM). This

comprises individual income taxes, corporate taxes, petroleum income taxes, real

property gains taxes and stamp duty.

2

According to Jeyapalan Kasipillai(2005), taxation is a vital economic tool for

governments to regulate the economy, to revitalize economic growth through the

granting of fiscal incentives and to provide funds for development projects.

Attached in Appendix A is a table which provides information on the federal

government revenue obtained from the Economic Report 2013/2014 of Ministry of

Finance. According to the report, in 2013, 57.6% of the revenue was in the form of

direct taxes while 16.6% was in the form of indirect taxes.

With reference to the second table in the Appendix B, the Economic Report

2013/2014 reports an overall deficit yet again. From the table, it can be seen that

Malaysia faces deficits even from 2010. This has accentuated the desperate need for

Malaysia to develop a more efficient, effective taxation system which will be able to

provide Malaysia with a stable source of revenue.

Since the Sales and Service tax had inherent weaknesses in its system in terms of

generating stable revenue, Malaysia was forced to look into the implementation of

GST. GST is a broader consumption-based tax which is proven to be able to generate

stable revenue for the nation's financial needs. GST is levied on all goods and

services except for essential ones which are exempted from tax, or those which are at

zero-rated tax for now (Palil & Ibrahim, 2011).

Sales tax is an indirect tax levied on certain imported and locally manufactured

goods introduced in 1972. It is also known as manufacturer's tax. It is a single stage

3

tax and 10% sales tax is charged to manufacturers of locally manufactured goods

when goods are sold or for imported goods, at the time when goods are cleared at

Customs. Service tax of 6% was introduced in 1975 and is levied on taxable services

which include prescribed goods, i.e tobacco as well as professional and consultancy

services (Palil & Ibrahim, 2011). The Goods and Services Tax will replace both

these taxes at a rate of 6%. It is also categorized into standard-rated, zero-rated and

tax-exempt items. It is mandatory and involves all Malaysians nationwide as it is a

consumption based tax.

1.2 Problem Statement

Budget 2014 was a milestone budget as the Honourable Prime Minister Dato’ Seri

Najib Tun Razak announced the implementation of Goods and services tax (GST) on

the 1st of April 2015. According to The Star newspaper the Prime Minister in his

budget speech stated that the Government had to take bold measures to overcome the

weaknesses of the current sales and service tax in order to strengthen the fiscal

position of the nation. Thus, after years of detailed and comprehensive studies as

well as views of all segments of society including chambers of commerce, investors,

economists, academicians, consumer associations and NGOs, the government has

decided to implement a fair tax system that benefits all Malaysians, Goods and

Services Tax (GST). The GST was publicized at a rate of 6%, higher than what was

promised in earlier proposals (Singh 2013). However, this rate is the lowest amongst

the ASEAN countries compared with 10% in Indonesia, Vietnam, Cambodia, the

Philippines, and Laos and 7% in Singapore and Thailand.

4

Amongst the main highlights associated with the implementation of GST was that

GST will not be implemented on approximately 40 essential goods and services, the

Government will be assisting the Rakyat with the GST transition period by providing

them with one off cash allowance and a reduction of 1% in individual income tax

rates. The Prime Minister also iterated that the GST is vital in assisting the

Government to generate revenues in an effective way in order to invest in the

development of the country (Lai 2014; New Strait Times, 2014).

Despite the implementation of GST being a topic already put forth many years ago,

the acceptance level amongst Malaysians is poor as represented by the massive rally

organized by a coalition of 89 various parties on May 1 2014. The rally with the

theme “GST: Protest Till It’s Dropped’ was aimed at displaying the protest of

citizens against the implementation of GST and enforced the need of the

Government to first resolve the issue on the high personal debt level of the nation

and minimum wage as well as that the people are currently still adjusting to the hike

in petrol prices and other essential goods. (Star Online, 2014)

Moreover, a survey conducted by the Merdeka Centre in May 2014 of 1009

registered voters comprising 60% Malay, 31% Chinese and 9% Indian concluded

that 62% of Malaysians are not in favour of GST. The respondents were selected on

the basis of random stratified sampling along ethnicity, gender and state of residence

in order to obtain results which are generalizable. At the same time, 64% of the

respondents indicated that they are not aware of the working of the national

economy. To add on, 72% of those with household income less than RM1, 500 and

56% of those with household income between RM1, 501 and RM3,000 do not

5

understand GST. More than half of those who earn between RM3, 000 and RM5,

000 and 67% of those who have a household income of over RM5, 000 understand

the issue.(MerdekaCenter,2015).

Also, 45% of the respondents have indicated that GST is not a fair tax system when

in fact more than 160 countries have successfully implemented GST. According to

the poll, 62% Malays do not understand GST, while less than half of the Chinese

(41%) and Indian (35%) respondents do not understand the issue

(MerdekaCenter,2015).The results of the survey are attached in Appendix C. To

further corroborate this phenomenon, some of the riot participants who were

interviewed by the Star newspaper displayed the same lack of knowledge in the

system and high level of uncertainty in its impact to their daily lives.

Previous studies (Nordiana, 2012; Bidin & Shamsudin, 2013; Ramalingam et al.,

2014) were focused on the compliance of organizations with GST and their adoption

of the GST application systems. Previous studies have also explored the awareness

and acceptance levels of consumers (Saira et al., 2010; Palil & Ibrahim, 2011;

Shamsuddin et al., 2014; Moomal & Zakarian, 2014). However, in the nation’s

current scenario, it is vital for the government to understand the causes of the low

levels of acceptance. In line with that, this study will delve deeper into factors that

are influencing consumers’ acceptance of GST using the well-established Theory of

Planned Behavior. In order to further understand the factors influencing the

consumers’ acceptance of GST, antecedent variables to the attitude and perceived

behavorial control are examined. The antecedent variables determined were Rule

Observance Behaviour, Perception of GST Fairness, Level of GST knowledge and

6

Self-Efficacy. If these antecedent variables together with the other variables

examined in the study (attitude, subjective norm and perceived behavioral control)

proved significant, recommendations can be suggested to ensure a more efficient and

smooth implementation of GST. This is because the relevant authorities can now

identify specific areas for improvement in order to address any acceptance issues

amongst the consumers.

The focus of this study is knowledgeable consumers who are currently pursuing their

Masters in Business Administration. This is because the Merdeka Center survey had

portrayed that there was a higher level of unacceptance amongst private sector

employees, self-employed individuals and those involved in business. Moreover, the

age groups comprising of those in their twenties and thirties showed a higher level of

unacceptance (Merdeka Center, 2015). MBA students will be able to represent the

demographics of the Merdeka Center survey who are not accepting GST. MBA

students are normally within the above-mentioned age group and majority are

employed (Tay, 2001) Moreover, lack of GST acceptance has been associated with

lack of knowledge (Borneo Post Online, 2015; Jalil, 2015; GST Malaysia Info,2014)

and by choosing MBA students we will be able to put that assumption to test as

MBA students are expected to have the required basic knowledge on GST. Thus, for

this study, MBA students will represent the consumers in Malaysia.

7

1.3 Research Objectives

1) To examine the level of acceptance of Goods and Service Tax (GST)

amongst consumers.

2) To examine the relationship between the existence of rule observance

behavior and attitude of consumer.

3) To examine the relationship between perception of GST fairness and attitude

of consumer.

4) To examine the relationship between level of GST knowledge and perceived

behavioral control of consumer.

5) To examine the relationship between self-efficacy and the perceived

behavioral control of consumer.

6) To examine the relationship between attitude of consumers and their level of

acceptance of GST

7) To examine the relationship between subjective norms of consumers and

their level of acceptance of GST

8) To examine the relationship between perceived behavioral controls of

consumers and their level of acceptance of GST

1.4 Research Questions

1) What is the level of acceptance of GST amongst consumers?

2) What is the relationship between rule observance behavior and attitude of

consumer?

8

3) What is the relationship between perception of GST fairness and attitude of

consumer?

4) What is the relationship between the level of GST knowledge of consumers

and perceived behavioral control?

5) What is the relationship between self-efficacy of a consumer and perceived

behavioral control?

6) What is the relationship between attitude of consumers and their level of

acceptance of GST?

7) What is the relationship between subjective norms of consumers and their

level of acceptance of GST?

8) What is the relationship between perceived behavioral control of consumers

and their level of acceptance of GST?

1.5 Definition of Key Terms

1. Level of Acceptance refers to the level of intention to accept (Shamsuddin et al.,

2014).

2. Attitude is an individual’s positive and negative feelings about performing a

target behavior. The feelings are brought about after evaluations conducted based on

beliefs (Fishbein & Ajzen, 1975).

3. Subjective Norm is the individual’s opinion that those who are important to

him/her think that he/she should not perform the behavior in question

(Fishbein&Ajzen, 1975).

9

4. Perceived behavioral control is the perceived internal and external constraints on

behavior or intention to behave (Taylor & Todd, 1995).

5. Rule observance behavior relates to obedience to authority (Trevino, 1986).

6. Perception of fairness is the judgement of individuals that something or someone

is free of biasness and injustice (Saad, 2010).

7. Self-efficacy is an individual’s perception of his/her capability to organize and

execute courses of action required to accomplish selected types of performances

(Bandura, 1986).

8. Level of GST knowledge is an individual’s technical and general knowledge on

the subject matter of GST. (Saad, 2010)

1.6 Significance of Study

Both theoretical and practical implications of the study will be discussed.

1.6.1 Practical Implication

With the Goods and Services Tax finally being implemented, a study of this nature

would be imperative for the government in identifying areas at which they should

improve in and allocate resources to. This is very crucial especially since according

to Deputy Finance Minister Datuk Ahmad Maslan, there is RM250 million budget

10

allocation to educate the public about GST and to facilitate its implementation (Ian,

2014 ; Malaysian Insider, 2014).

This will therefore enable the Royal Malaysian Customs Department to undertake

measures for a smooth and effective transition to the GST system for consumers and

the Government as a whole. This includes enhancing their current support system

and enforcement measures (Kasipillai, 2013; Tan, 2009; Tan, 2014; Choo, 2014).

Previous studies (Saira et al., 2010; Palil & Ibrahim, 2011) have been conducted on

an exploratory mode as the GST implementation was not finalized. Other studies

Shamsuddin et al. (2014) and Moomal & Zakarian (2014) studied on the correlation

between awareness and acceptance. However, all these studies were not backed up

by a theory. Nordiana (2012), Bidin & Shamsudin (2013) and Ramalingam et al.

(2014), conducted a study on the compliance with GST and adoption of GST

application system of corporate tax collectors. Thus, this study would be the first

study to examine the factors that influence the acceptance level of consumers at large

amidst the announcement of the confirmed implementation of GST on 1 April 2015.

The key idea to understand here is that with corporate tax payers, their compliance is

harped on because under the GST implementation they are required to file in taxes to

the Royal Customs on a monthly or quarterly basis. Although they bear expenses on

preparing infrastructure and staff to ensure compliance to GST, which is usually a

one-off expenditure, it is the consumers who bear the tax burden of GST and

corresponding price hikes. The findings of this study will ultimately provide insight

11

to the Consumers’ association on the information they can disseminate to the

consumers in order to assist them to accept and adapt to the implementation of GST.

1.6.2 Theoretical Implication

Although the implementation of GST has been decided upon-1 April 2015, however,

the acceptance level of consumers is still low. As such, the only issue at hand now is

to handle the transition process as smoothly as possible.

For this to be a success, the theory of planned behavior comes in handy to explain

the various possible factors to be addressed in order to enhance the acceptance of

GST amongst consumers. Previously this theory has been used to study corporate tax

payers’ compliance to GST (Nordiana, 2012; Bidin & Shamsudin, 2013) but through

this research, the theory will be extended to understand the factors that contribute

towards the acceptance of GST by consumers. Consumers do not need to comply

with the GST but rather just accept it as the tax will already be embedded in prices of

goods and service. In this case, the study will examine the consumers’ level of

acceptance of GST by determining their intention to accept GST because at the time

of data collection GST has not been implemented yet.

Hence, this study will test the theory’s rigidity in instances where the intention to not

behave in a certain way does not enable the respondent to escape from the behavior

itself.

12

Antecedent variables are also used in this study to enhance the explanatory power of

the Theory of Planned Behavior. With these antecedent variables, the research will

be able to demonstrate very specific key areas for improvement that the government

should allocate resources to. The antecedents were Rule Observance Behavior,

Perception of GST Fairness, Self-Efficacy and Level of GST Knowledge.

1.7 Organization of the remaining chapters

This research is made up of a total of five chapters in order to present the study in a

systematic manner. The next chapter, Chapter two presents an overview of previous

literature on the variables put forth. This then leads to the development of the

theoretical framework and hypotheses for this study.

Chapter three consists of the research methodology, research design, data collection

methods and data measurement methods that will be employed for this study.

Chapter four presents the findings of this study where the goodness of the data used

is confirmed and the hypotheses developed are tested.

The final chapter, Chapter 5 will discuss the implications of the findings, practical

and theoretical applications of this study, limitations of the study, future research

suggestions and the conclusion.

13

CHAPTER 2

LITERATURE REVIEW

2.0 Introduction

This chapter aims to examine all existing literature on the content, theory and latent

variables of this study as well as the relationships being examined in relation to

Goods and Services Tax which will be implemented come 1st April 2015.This will

then lead to the formation of the theoretical framework and the development of the

hypotheses.

2.1 Malaysian Taxation

The British colonialism introduced taxes into the Federation of Malaya in 1947

which has resulted in the taxation system in Malaysia today (Singh, 1999). At first,

the Income Tax Ordinance 1947 was gazetted as the fundamental act yet this was

accordingly changed and replaced by Income Tax Act 1967 (ITA) with effect from

January 1, 1968. At that time, ITA combined the three acts of income taxation in

particular Sabah Income Ordinance 195633, Sarawak Inland Revenue Ordinance

196034 and Income Tax Ordinance 194735. As of now, ITA 1967 is the fundamental

act to administer direct taxes in Malaysia including individual and corporate income

tax (Palil, 2010).

14

The Lembaga Hasil Dalam Negeri or the Inland Revenue Board (IRB) is the tax

authority which oversees the operationalization of direct taxes in Malaysia. Other

than ITA, the IRB are also accountable for supervision, assessment, collection and

enforcement of petroleum taxes, real property gain taxes and stamp duties. The

indirect taxation system is controlled by the Royal Malaysian Customs (RMC). This

includes customs duties, excise duty, sales tax and service tax (Palil, 2010).

2.2 Consumers

In previous research, (Nordiana, 2012; Bidin & Shamsudin, 2013; Ramalingam et

al., 2014) their scope of study was on industrial taxpayers such as manufacturers,

retailers and distributors. The focus was on businesses that were registered under

GST compliance. In this study, the scope of study is on consumers, individuals who

consume goods with taxes already embedded in the pricing. Thus, the matter of non-

compliance or evasion does not exist in this context. In this study, MBA students

will be representing the consumers as majority of them are within the age group and

employment status highlighted in the Merdeka Center Survey of those who had a

higher percentage of not accepting GST.

An in-depth discussion of GST in the following component will highlight the very

fact that although GST will be charged at different stages, in the end the burden will

be on the consumer (Voon, 2013). The implementation of GST will have quite a

significant impact on consumers especially since according to the Report on

Household Expenditure by the Department of Statistics Malaysia, average monthly

expenditure per household has increased from RM 1,161 in 2004 up to RM 2,190 in

15

2010. Even without GST implemented, consumers are facing an increase in

expenditure by 89% over 6 years. Thus, it is vital to study the perceptions and

acceptance of GST by these Malaysian consumers and the factors that are

influencing their perceptions.

Moreover, according to Bank Negara, Personal/Household Debt is standing at an

alarming rate of 83% of Gross Domestic Product of the nation compared to a 70% in

2009 and the debt to household income ratio is standing at an alarming 140%

(Joseph, 2013). Additional statistics creates more concern over this implementation

of GST. The increase in GDP of 5.1% in 2012 is less than half of the growth of

household debt of 12% in 2012, indicating severe unsustainability (Joseph, 2013).

Consumers are already high in debt pre-GST and it is worrying to imagine the

impact on consumers’ post-GST, where prices of goods and services are expected to

increase.

Hence, with the undertaking of this study, the reasons for knowledgeable consumers

accepting or not accepting GST can be determined and consequently influence the

channeling of resources towards customizing assistance and support for the

consumers in the most effective way.

16

2.3 GST

Goods and Services Tax or GST is a tax on consumption based on a value-added

concept. In a nutshell, the more you consume the more tax you pay. Unlike the

present sales tax or service tax which is a single stage tax, GST is a multi-stage tax.

The tax is not a cost to the intermediaries since they are able to claim back GST

incurred in their business operations. GST is imposed on goods and services at every

stage in the supply chain including situations in which the good or service is

imported. It is not a new tax but instead is a tax to replace the current Sales Tax and

Service Tax (SST). GST is also known as VAT, Value-Added Tax in certain

countries (RMCD, 2013)

2.3.1 Reasons for implementing GST

The key reason for many countries adopting GST/VAT is the inefficiency of the

current tax system in financially supporting the development of the nation’s

economy. The same scenario occurred here in Malaysia and the Ministry of Finance

has studied and confirmed that the implementation of GST can overcome the various

weaknesses under the current Sales & Service Tax system. These weaknesses

include tax compounding and cascading, transfer pricing and reliefs on certain goods

(RMCD, 2013).

17

Figure 2.1: Cascading effect -Sales & Service tax (RMCD, 2013)

Figure 2.2: Cascading effect - Service tax (RMCD, 2013)

In 1990, direct taxes contributed to 35.2% of the nation’s revenue whilst indirect

taxes contribute to 36.7%. The contributions of both categories of taxes were equal

but trade liberalization policies were amended causing major inequality in the

contribution of direct and indirect taxes. This was clearly depicted in 2012 when the

contribution from direct tax was 56.4% and indirect tax 17.2%. At this point of time,

Malaysia relies profoundly on direct taxes and petroleum revenues, a situation which

economic experts do not recommend. According to these experts, over-dependence

on only certain taxes will have an adverse effect on the nation’s financial position.

Hence, to eliminate these adverse effects this is an appropriate time for the

Government to engage in an overall tax reform and implement GST. (RMCD, 2013)

18

2.3.2 Significance of GST

GST was introduced by the government in the 2005 budget but postponed to a later

date. Following that, the GST Bill was tabled for the First Reading in the Dewan

Rakyat in December 2009 but in 2010 the implementation was postponed again.

However, the implementation of GST is inevitable to support the nation’s long term

economic growth and the pressure was on the government to ensure a smooth

transition to GST. In order to do this, the Government has to take into account the

welfare and concerns of the society as a whole so that it is well-received (RMCD,

2013).

With this in mind, the Government had undertaken social impact studies of the

implementation of GST. To ease the tax burden on the consumers, certain goods and

services which are indispensable to the low and middle income group have been

proposed not to be subjected to GST. To add on, basic items such as poultry, rice,

flour, meat, vegetable, sugar, flour, cooking oil, residential and agricultural

properties, health services and education will not be subjected to GST (RMCD,

2013).

GST is a more transparent, competent, effective and less bureaucratic taxation

system. The double taxation phenomenon under the Sales & Services Tax regime is

eliminated with GST in place. From a business point of view, they are able to reduce

costs as they now can claim back the GST incurred which was not possible under the

SST regime. A company which back then had to absorb the sales and service taxes

19

2.3.3 Mechanism of GST

The mechanism of GST can be examined from the consumers’ and businesses’ point

of view.

(a) Business

GST is charged on the supply of goods and services made in Malaysia and on the

imports of goods and services into Malaysia. GST is charged on the selling price of

the goods or services and only the GST amount will be forwarded to the

Government. The GST value is representative of the value added to the goods or

services at each level of the supply chain. The value added is the value added to a

raw material or purchases before selling the new or improved product or service. To

operationalize this, GST adopts a credit offset mechanism whereby GST charged

on the output of the business is offset against the GST paid on the goods or services

acquired as inputs by the business. Along these lines, a company will be charged

GST by its suppliers and simultaneously, the company will charge GST to its

paid can now claim the GST paid from the Customs. The net effect is definitely a

reduction in costs for these companies (RMCD, 2013).

Exportation of goods and services are not subjected to GST which will in return

ensure Malaysian exports are competitive in the global market. Consequently, this

will improve the Gross Domestic Product of the nation (News Strait Times, 2015)

With GST in place, shadow economy activities such as illegal trading, black money

and tax evasion can be curbed. (Schneider & Frey, 2000 ; MalayMail Online, 2015)

20

customers. GST charged on output is called output tax and GST incurred on

purchase is called input tax. This offsetting mechanism is to guarantee GST paid by

businesses are recoverable leading to a reduction is cost. Most importantly, this

ensures there’s no double taxation and the net tax effect on the end consumer is only

6%. (RMCD, 2013)

Diagram 2.3: Input and output tax (RMCD, 2013)

(b) Consumers

GST-registered businesses will collect GST from consumers on the goods and

services purchased. There are three types of supplies listed under GST which

are(Examples are derived from Royal Malaysian Customs Guide on Supply

under GST):

(i) Standard rated supplies are taxable supplies of goods and services which are

subject to a positive rate of 6%. Examples of these supplies are sales of

commercial properties, vehicles, accessories and packaged food items.

(ii) Zero rated supplies are taxable supplies which are subject to a zero rate.

21

Examples of zero-rated supply are fresh vegetables, live animals, books and

exports of goods and services. Businesses dealing with zero-rated supplies are

still able to claim the input taxes incurred.

(iii) Exempt supplies are non-taxable supplies which are not subject to GST.

However, the GST paid on input by the businesses cannot be claimed as tax

credit. Examples of exempt supply of services are domestic transportation of

passengers for mass public transports, private education and private health

services. Examples of exempt supplies of goods are residential properties, land

for agricultural and land for general use.

Below is an illustration showing how GST works:

Diagram 2.4: Standard rated supply (RMCD, 2013)

Diagram 2.5: Zero-rated supply (RMCD, 2013)

22

Diagram 2.6 : Exempt supply (RMCD, 2013)

Although GST is imposed at every stage of the supply chain, businesses can claim

the GST incurred on inputs. Such mechanism implies that the end consumer only

pays GST at the rate of 6% and not 24%.

Diagram 2.7: GST Mechanism- standard rated (RMCD, 2013)

Hence, for consumers patronizing zero-rated and exempt-rated products, there would

not be any tax burden unlike purchasing a standard rated product. However, delving

23

deeper, the consumer should understand that producers of exempt rated goods and

services cannot claim back their incurred input GST and will have no choice but to

channel the cost to the consumers by increasing prices. For example, a private

education institute will have to pay their suppliers GST for purchases of

infrastructure and maintenance services but will not be able to claim this GST

expense from the customs as their service is considered an exempt rated service.

However, the institute is now incurring higher costs than pre-GST implementation

and will have to increase their selling prices as well to cover the costs incurred.

An equally important point to understand is that GST, if well-implemented

should not result in increase of prices of all standard-rated products and

services. With the discussions below, an understanding can be obtained on the

three main implications of standard-rated GST on consumers.

Scenario 1:

10% SALES TAX abolished, 6% GST implemented

The sales tax of 10% is a business to business transaction and this tax burden is

usually embedded in the final pricing to end consumers. With the

implementation of GST, consumers should end up paying less because a

product manufactured or imported now is subject to 6% GST rather than 10%

Sales Tax.

24

Scenario 2:

6% SERVICE TAX abolished, 6% GST implemented

While most consumers don’t experience the direct effect of the 10% Sales Tax,

most of them would have paid for the 6% Service Tax. In the scenario of

consumers engaging in service for which they were previously charged 6%, the

implementation of GST will not impose any price increase on items.

Consumers will be paying the same.

Scenario 3:

No Service or Sales Tax previously, GST 6% implemented

As GST is a broad-based taxation system, more sectors of the economy are

covered as compared to under the SST regime such as the paper and printing

industry, food preparation industry, medical and educational equipment

industry as well as repackaging industry. (Customs Guide Book,2003) Unless

the goods or services are zero-rated, prices of goods not previously covered

under SST will now be affected. Consumers will have to pay a higher price.

2.3.4 GST rates in the ASEAN countries

The GST rate of 6% suggested is the lowest in comparison to our neighboring

countries. Although Singapore initially introduced GST at 3% but the current GST

rate is 7%.

25

The following are the GST rates among the ASEAN countries:

Figure 2.8: GST rates in the ASEAN countries (RMCD, 2013)

2.3.5 Implementation of GST

Budget 2014 has announced the confirmed implementation of GST on 1st April

2015. Many steps are undertaken to ensure the smooth implementation of GST.

Currently the Ministry of Finance is conducting a series of awareness program on

GST to the public and businesses. The aim of this awareness program is to help the

public and business to have a better understanding on the proposed GST model in

Malaysia, the mechanism and steps taken by the Government to overcome issues

raised concerning GST. (RMCD, 2013)

The people need to know that GST is charged and collected on all taxable goods and

services produced in the country including imports. Only businesses registered under

GST can charge and collect GST. GST collected on output must be remitted to the

26

Government. However, businesses are allowed to claim the input tax credit.

From the nation’s point of view, Minister in the Prime Minister's Department, Datuk

Seri Idris Jala stated that GST implementation is projected to result in additional

revenue of RM20 billion to RM27 billion, at maturity. At maturity refers to the point

in time when every Malaysian starts to contribute towards the GST. The additional

revenue thus can be used to enhance the well-being of Malaysians as a whole

(Xavier, 2013).

2.4 Theory

A theory in essence is a tested general proposal, universally regarded as correct,

that can be used as a principle of explanation and prediction for a scenario (Oxford

Dictionary, 2015). This study employs the Theory of Planned Behavior to determine

the consumers’ acceptance behavior of GST. The theory is used to support the

framework of this study. Antecedent variables are added in to enable better

explanation of the components of TPB. Detailed discussion will be done in following

sections.

2.4.1 Theory of Planned Behavior (TPB)

Through his article "From intentions to actions: A theory of planned behavior", Icek

Ajzen put forth the Theory of Planned Behavior in 1985. This theory was expanded

from the theory of reasoned action proposed by Martin Fishbein together with Icek

Ajzen in 1975. (Azjen 1991) The theory of reasoned action portrays behavioral

http://en.wikipedia.org/w/index.php?title=Martin_Fishbein&action=edit&redlink=1

27

intention as antecedent to actual behavior. Behavioral Intention is determined by two

independent variables; attitude and subjective norms.

In the Theory of planned behavior, to take into account conditions of variable

control, an additional construct was included- Perceived Behavioral Control. This

particular component reflects the respondent’s perception of how much control they

have over the behavior. (Taylor & Todd, 1995) Therefore, according to TPB,

intention to behave in a certain way is determined by the attitude towards the

behavior, subjective norms and perceived behavioral control. The TPB is highly

valuable due to the fact that it can be used in numerous applications and has a more

powerful predictive framework compared to TRA (Ajzen, 1991)

Predicting behavior is the ultimate objective of TPB thus it does not explain the

behavior (Conner & Sparks, 2005). The TPB model is not restricted to predicting

behaviors in information systems (Davis et al., 1989) and other human behaviors

(Paris &Broucke, 2008; Guo et al., 2007; and Chang, 1998), but is also useful in

explaining tax compliance behavior (Saad, 2010; Palil & Ibrahim 2011). Research on

tax compliance has tried to prove this theory in a number of countries. Trivedi &

Shehata (2005) proved the Theory of Planned Behavior in Canada, Bobek and

Hatfield (2003) proved it in the USA, Damayanti (2012) proved it in Indonesia and

this study aims at proving the applicability of this theory in the context of Malaysian

Goods and Services Tax.

28

2.5 Dependent Variable

In this study, level of acceptance of Goods & Services Tax has been identified as the

dependent variable of the study. Level of acceptance of GST is measured by the

individual’s intention to accept GST. Intention is defined as being mentally

determined to accept GST (Hung, Chang & Yu, 2005). Level of acceptance is simply

the level of agreement of the individual with the implementation of GST. The TPB

advocates that the intention to behave in a certain manner is highly influenced by the

attitude towards the behavior in question, whether or not those who are influential to

the respondents have the behavioral intention-subjective norm and perceived

behavioral control- the ease or difficulty of performing the behavior caused by

internal and external constraints (Ajzen, 1991; Sommer, 2011).

The type of behavior impacts the degree of intention-behavior uniformity. Findings

confirmed that intentions are significant predictors of an action. The intention to

behave in a certain way is a sufficient proxy measure of behavior. Previous studies

have concurred that intention to comply with tax obligations explains an individual’s

decision to comply with tax obligations (Bidin & Shamsudin 2013; Saad, 2010;

Trivedi et al.,2005; Palil & Ibrahim,2011). Cheng, Lam & Yeung (2006) also argued

that for a survey-based research design, measurement of behavioral intention is more

appropriate than actual behavior.

29

2.6 Antecedent and Independent variables

Independent variables of this study consist of attitude, subjective norm and perceived

behavioral control. Antecedent variables consist of rule observance behavior,

perception of GST fairness, level of GST knowledge and self-efficacy.

2.6.1 Attitude

Attitude is an individual’s assessment of performing a behavior. The evaluation can

be either positive or negative (Fishbein & Ajzen, 1975). Attitude towards behavior is

defined as a function of an individual’s beliefs towards a behavior and a subjective

evaluation of that behavior (Fishbein & Ajzen, 1975). The belief component captures

a person’s perceptions about a certain behaviour. Many studies confirmed the

correlation between attitudes and behavioral intention. (Wan et al., 2012) It is highly

related to behavioral intention because individuals form intentions to perform

behaviors towards which they have positive feelings. Ajzen (1991) specifies that

attitudes towards compliance reflect feelings of favour and disfavour towards

behavior, in this case level of acceptance of GST.

Any attitude is a hypothetical or latent variable rather than an immediately

observable variable. It is, in other words, an abstraction. The concept of attitude does

not refer to any one specific act or response of an individual, but it is an abstraction

from a large number of related acts or responses. When we state that a certain

individual, A, has a less favorable attitude towards GST than another individual, B,

30

we mean that A's words and actions are consistently less favorable to GST than B's

words and deeds (Jones, 1970).

There are a number of functional approaches to attitude and the approach applicable

to this study would be the adjustive function. This function of attitude has relevance

in the realm of behavioral theory. It follows from the nature of the adjustive function

of attitudes that the clarity, consistency and nearness of rewards and punishments are

important factors in the acquiring of new attitudes. In this case, the possible rewards

and benefits from GST could be an important factor in consumers acquiring a

positive attitude towards GST (Jones, 1970)

Social psychologists agree that an attitude involves at least three things

(Culbertson,1986):

1. An attitude object which may be physical or an abstraction. In this context, it

is the GST.

2. A set of beliefs that the object is either good or bad. Hence in this case it

would be the belief of whether GST is good or bad.

3. A tendency to behave towards the object so as to keep or eliminate it. This

refers to the acceptance of GST.

Using belief-based attitudes measurements in a taxation context will result in

attitudes explaining compliance behavior (Bobek, 1997). Thus, it is anticipated in

this study that a positive attitude towards the GST tax system would encourage

31

consumers to be accepting of it. In this study, we have considered two antecedents to

attitude, namely Rule Observance Behaviour and Fairness Perception.

Attitude is operationalized in previous studies (Loo et al.,2007 & Bobek 1997) by

determining the individual’s positive or negative feeling towards the behavioral

intention. The same concept is applied here to determine the beliefs and perspectives

that these consumers have on GST.

2.6.2 Rule Observance Behavior

Rule Observance Behaviour is an act or instance of following a rule. Trevino (1986)

discussed a model of ethical decision making wherein situational and individual

moderators interact with an individual’s cognitions to determine ethical or unethical

behavior.

One situational moderator included in the framework is “obedience to authority”.

Obedience is a form of social influence where an individual acts in response to a

direct order from another individual, who is usually an authority figure. (McLeod,

2007) Rule Observance Behavior is studied as an antecedent to attitude in this study.

Previous study (Sarina et al., 2007) measured the obedience of auditors to authority

and rules whilst another previous study (Jeffrey &Weatherholt, 1996) measured the

rule observance behavior amongst corporate accountants. This study will undertake

32

the same operationalization of this variable as it is a straightforward and simple

variable.

2.6.3 Perception of fairness of GST

The term fairness relates to the justification and validity a consumer assigns to

the burden of the GST. The term perception of fairness is used here because each

individual has a different perspective of the fairness of the taxation system which

does not conclude on the ultimate fairness of the system (Jones, 2009).

Previous studies indicate that fairness perceptions can take various forms. The first

dimension that can be looked at is the vertical fairness, which stresses that different

tax rates should be enforced for taxpayers with different economic situations (Erich

et al., 2006). This would result in higher income earners paying tax at higher rates

than low-income earners. The next dimension is horizontal fairness, defined as ‘the

equal treatment of equally circumstanced individuals’ (Michael, 1978). In this

horizontal fairness dimension, taxpayers of similar economic positions should pay

the same amount of tax. Horizontal and vertical fairness are derived from the

Distributive Justice Theory (DJT) which argues that individuals in similar

circumstances must be treated in an equivalent manner without compromising their

individual needs.

Besides vertical and horizontal fairness, a study on the US tax system brought forth

the concept of procedural fairness and policy fairness (Bobek, 1997). Procedural

33

fairness relates to the procedure engaged to reach distribution outcomes while policy

fairness deals with the content of the tax law. Moving on, exchange fairness is a very

significant dimension (Gilligan & Richardson, 2005; Gerbing, 1988), which

represents the exchange of contribution and benefit between taxpayers and

government. In simple terms, the exchange fairness dimension highlights that

taxpayers will have fair perceptions of the tax system if the benefits received from

the government are justifiable compared to their tax contributions. Other dimensions

of fairness include a preference for either progressive or proportional taxation

(Turman, 1995), personal fairness, tax rate fairness, special provisions and general

fairness (Gilligan & Richardson, 2005; Richardson, 2005a; Christensen &

Weichrich, 1996; Christensen et al., 1994; Gerbing, 1988). The above review on

studies of tax fairness suggests approximately ten dimensions of fairness.

However, in this study, only two dimensions are identified to be important in

assessing the fairness of the GST system; general fairness, exchange fairness (Saad,

2010) General fairness simply measures individuals’ judgments whether the GST tax

system is generally fair or not while exchange fairness is concerned with a mutual

exchange between taxpayers and the government. The rest of the dimensions would

not be applicable as GST is a tax on consumption and not income. Perception of

fairness of GST is studied as an antecedent to attitude.

In general, equity theory is engaged when studying tax fairness. Equity

theory explains relational satisfaction in terms of fairness perceptions of distributions

of resources within interpersonal relationships (Jones, 2009).

34

2.6.4 Subjective Norm

Subjective norm reflects the level of influence significant referents have on

individuals. In this case, the influence would be in reference to accept or not accept

the implementation of GST. Ajzen (1991) describes subjective norm as ‘the

perceived social pressure to perform or not to perform the behaviour’ or ‘an

individual’s perception that important others would approve or disapprove of his or

her performing a behaviour’ Subjective norm addresses the individual’s belief that

others in their support network want them to support or oppose GST, and their

motivation to comply with the views of those persons. This is simply known as the

social influence component of decision making.

According to the theory of planned behavior, the support of significant others will

influence consumer’s acceptance of GST. The opinions of peers are critical in

shaping the views of individuals due to the necessity for interpersonal support during

a time of change (Gerpott, 1990) and the effects of information processing on

socialization (Salancik and Pfeffer, 1978) In this context, subjective norms in

intention to accept GST refer to how other consumers feel about the subject matter.

As subjective norm is related to significant referents, peers and family members are

taken up as the influential referents (Damayanti, 2012).

No antecedents were identified for the subjective norm as influences from peers and

family members would have the same impact and thus would not provide additional

insight. The same measurements used in Bhattacherjee (2000) were used for this

35

study as the two main areas of peers and family members have to be covered in order

to have a valid measurement.

2.6.5 Perceived behavioral control (PBC)

PBC is the latest and unique construct of the TPB and can be linked to behavior both

directly and indirectly through intentions. The direct impact is the actual control an

individual has over behavioral performance whilst indirect impact refers to the

motivational influence of control on behavior. In this case study, we are looking at

the indirect link of perceived behavioral control and behavior through the intention

to behave.

Perceived behavioural control reflects an individual’s perception on the ease or

difficulty in performing a particular behaviour. In a more detailed explanation, PBC

refers to an individual perception of accessibility of resources or opportunity

necessary for performing the behavior (Ajzen, 1985; Ajzen & Madden, 1986) Ajzen

(1991) stipulates that high perceived behavioural control results in a behaviour being

easy to perform, while one that is difficult to perform is low in perceived behavioural

control.

There are internal and external perceived behavioral controls. The internal factors

refer to individual’s disposition which includes the amount of information a person

has. The external factors determine the extent to which circumstances support or

inhibit the performance of behavior (Ajzen, 2001) PBC has to be customized to the

36

type of behavioral issues under study (Sparks et al., 1997). Perceived behavioral

control is the perception of the resources, knowledge and ability to perform the

behavior (Giantari, 2013). In this case, the two antecedents to PBC are level of tax

knowledge and self-efficacy. Both are internal perceived behavioral controls.

Previous studies (Sarina et al., 2007; Nordiana 2012) operationalized this variable by

examining the perceived internal and external constraints on behavior or intention to

behave. The same measurement is undertaken here in order to measure the

constraints or resources that individuals face or have in accepting GST.

2.6.6 Level of GST Knowledge

A voluntary tax compliance system is effectively functional only with tax knowledge

(Kasipillai, 2000). The tendency to not comply with tax whether intentionally or

unintentionally increases with the lack of knowledge. This was proposed by

McKerchar (1995) who studied small business taxpayers. She concluded that

taxpayers were not even aware of their tax knowledge deficit and this might lead to

unintentional non-compliance behavior.

An earlier study by Harris (1989) separated tax knowledge into fiscal awareness and

technical knowledge. Saad (2010) has divided tax knowledge into four categories but

only two are applicable here as we are not discussing compliance bur rather just

acceptance. The two are general knowledge and technical knowledge. Previous

37

studies have evidenced that general tax knowledge has a very close relationship with

taxpayers’ ability to understand and appreciate the laws and regulations of taxation.

(Singh, 2003). This can also be translated for the purpose of this case study, to the

fact that with the general and technical knowledge on GST, consumers will be able

to manage their consumption better in order to minimize the impact of the GST on

their expenses especially since many essentials are tax-exempt and zero-rated. Level

of tax knowledge is taken as an antecedent to perceived behavorial control in this

study.

This variable is measured both in a technical and practical way in this study and it is

customized to GST issues.

2.6.7 Self-Efficacy

Self-efficacy is associated with effective behavioral strategies and this has led to this

concept being studied as an influence to improved performance in various situations.

(Ben-Ami et al., 2014) Self-efficacy theory postulates that individuals judge their

ability to cope effectively with challenges when faced with environmental demands.

Based on this judgment, individuals initiate behavioral strategies to manage

challenges efficiently and attain desired outcomes (Bandura, 1997). Self-efficacy is

the confidence an individual has of his/her competence to perform specified tasks

(Bandura, 1997). It is a well-researched concept originating from the Social

Cognitive Theory (Bandura, 1986). Self-efficacy is comparable to self-confidence

however self-efficacy is more situational in nature. Self –confidence is more task-

38

specific whilst self-efficacy is a dynamic paradigm that changes over time depending

on new information and experiences (Schmidt & Karsten, 2015).

Previous studies (Schmidt &Karsten, 2015; Kraft et al.,2005; Kulviwat et al., 2014;

Ben-Ami et al., 2014;) had investigated this concept of self-efficacy in various

situations and have concluded that it is indeed a significant factor in those situations.

Thus, in this study, self-efficacy is in relation to the consumers’ judgement on how

well they will be able to cope with their lifestyles even with the implementation of

GST. This might include their perception of the costs and benefits involved as they

will have to set behavioral strategies. (Raghuram, 2003; Beauregard,2012) Self-

efficacy is studied as an antecedent to perceived behavioral control.

Self-efficacy was measured by previous research by enabling respondents to judge

their own capabilities to organize and execute their intention. (Kulviwat, 2014; Ben-

Ami, 2014) In this study, hence, this variable is measured by the individual’s

judgement of their ability to manage their lifestyles and finances with GST in place.

39

2.7 Theoretical Framework

In order to have a scientific and systematic research of this subject matter, a

theoretical framework has been formed. The theoretical framework is developed

using the Theory of Planned Behavior model as the underlying foundation. The

dependent variable of the framework is Level of Acceptance of GST. The

independent variables are based on the components of Theory of Planned Behavior

which are Attitude, Subjective Norms and Perceived Behavioral Control. Following

that, antecedent variables are determined and included to enhance the value of the

research. Antecedent variables are included only for Attitude and Perceived

Behavioral Control. The antecedent variables for attitude are Rule Observance

Behavior and Perception of GST Fairness whilst the antecedent variables to

perceived behavioral control are Level of GST knowledge and Self-efficacy.

40

Figure 2.9 Theoretical Framework

41

2.8 Factors Affecting Level of Acceptance of GST

Reading the literature, there are 3 main streams of literature that can be used for the

basis of this study. There are studies examining factors affecting compliance of GST,

factors affecting level of tax compliance and factors affecting level of acceptance of

products or systems, other than tax or GST. There are also a few other studies which

use the latent variables in this study in a different context. These studies have also

been used to support the hypotheses development involving the antecedent variables

and independent variables.

42

Table 2.1: Key Studies on Malaysian Goods and Services Tax

Author/Year Title Location Theory IV DV Methodology Findings

ZainolBidin and

FaridahwatiMohdSha

msudin/

2013

Using Theory of

Reasoned

Action to

Explain

Taxpayer

Intention to

Comply with

Goods and

Services Tax

(GST)

Malaysia Theory of

Reasoned

Action

1) Attitudes

2) Subjective

Norms

Compliance

with GST

Structured

questionnaires

that were

distributed to a

sample of 103

taxpayers

(corporates) with

the assistance of

the Department of

Royal Malaysian

Custom

Department in

Kedah and Perlis.

1) Subjective norms

and attitude

influenced the

manufacturers’

intention to

comply with

GST.

2) Subjective norm

was found to be

the strongest

factor for

successful GST

implementation

3) R square = 28%

NordianabintiRamli/

2012

The Perception

of Taxpayers

toward Goods

and Services

Tax (GST)

Implementation

Malaysia Theory of

Planned

behaviour

1) Attitude

2) Subjective

norm

3) Perceived

Behavioural

Control

4) Law &

enforcement

Intention to

comply with

GST

Structured

questionnaires

that were

distributed to a

sample of 150

with the

assistance of the

Department of

Royal Malaysian

Custom

Department in

Kedah and Perlis.

1) Subjective norms

and law &

enforcement

influence

intention.

2) Subjective Norm

is the strongest

factor.

3) R square = 28%

Mohd Rizal Palil and

MohdAdha Ibrahim /

2011

The Impacts Of

Goods And

Services Tax

(GST) On

Middle Income

Malaysia Theory of

Planned

behaviour

1) Demographics

2) Respondents’

readiness,

acceptance

Intention to

comply with

GST

Data was

collected through

a structured

survey among

middle income

1) This study

suggested that

many

respondents were

worried on their

43

Author/Year Title Location Theory IV DV Methodology Findings

Earners In

Malaysia

andperceptions.

3) Respondents’

behaviour

toward the

implementation

of GST

earners (RM

2000- RM 4000)

from various

organizations

including

government and

private sectors

from various

locations in Kuala

Lumpur,

Malaysia.

purchasing

power

particularly

among the

middle income

earners.

2) The study

provides the

government with

a number of

measures to be

undertaken for a

smooth process.

44

2.8.1. Studies on Goods & Services Tax

Table 2.1 shows the studies that have been conducted on GST which were referred to

in the hypothesis development of this study. As can be seen the studies conducted

are focusing on corporate tax payers who need to comply with the collection of GST

for the government. The Theory of Planned behavior is used in those studies and

hence can contribute significantly to the hypothesizing of latent variables’

relationships in this study.

2.8.2 Studies on Tax compliance

Studies on tax compliance have used Theory of Planned Behavior in the Western

context (Jones, 2009; Bobek & Hatfield, 2003) and Asian context (Damayanti, 2012;

Saad, 2010). Previous studies (Harris, 1989; Hanno &Violette, 1996; Bobek, 1997;

Tan & Chin Fatt, 2000; Richardson & Sawyer, 2001; Seidl & Traub, 2001; Loo et

al., 2007; Hai & See, 2011; Thomas, 2012; Devos, 2014; Boonyarat, 2014; Schmidt

& Karsten, 2015) all have dwelled in the various factors that affect tax compliance

amongst corporates and individuals. These studies also look into the various factors

that affect attitudes and perceived behavioral control towards tax compliance.

Amongst the common factors looked at were tax knowledge, tax fairness and self-

efficacy.

2.8.3 Studies on Intention to Accept/Level of Acceptance

As this study looks at the acceptance of GST, previous studies on acceptance, usage

and adoption were analyzed. Previous studies use the Theory of Planned Behavior to

45

study the acceptance of various subject matters such as technology, employee union,

systems and online applications (Teo & Lee, 2010; Vijayasarathy, 2003; Dawkins &

Frass, 2005; Sarina et al., 2007) Theory of Reasoned Action is also used in the study

of acceptance of computer usage (Shimp& Kavas, 1984). Through these studies,

relationships between attitude, subjective norms and perceived behavioural control

on acceptance can be examined and used to develop the hypotheses of this study.

Also, previous research (Kulviwat, 2014; Chiou, 1998) discussed the antecedents to

perceived behavioural control namely knowledge level and self-efficacy in the

context of technology and product acceptance.

2.8.4 Other Studies

Other studies in different contexts were examined in order to support the antecedents

to attitude and perceived behavioural control. Previous studies (Guthrie &

Schwoerer, 1994; Droomers et al., 2004; Raghuram et al.2003, Beauregard, 2012;

Ben-Ami et al., 2014) studied the relationship between self-efficacy and other factors

including perceived behavioural control. This is highly relevant to this study.

Trevino (1986) and Sarina et al., 2007) looked at the antecedents to attitude which

was significant in the formation of the hypotheses pertaining attitude in this study.

46

2.9 Hypotheses Development

Hypotheses development is done by first examining the relationship between the

antecedent variables, independent variables and dependent variable based on

previous literature.

2.9.1 Rule observance behavior and attitude

Rule observance behavior points to the obedience to authority whilst attitude

displays favour or disfavor to behave in a certain way. Being obedient to the

authorities reflects the belief that the law is above all. Goods and services tax is in

accordance to the laws of Malaysia and enforced by the Royal Malaysian Customs

Department. Hence, this rule observance behavior will influence a consumer’s

attitude towards GST (Sarina et al., 2007; Trevino, 1986).

The study hypothesizes that:

H1: Rule Observance Behavior will positively affect attitude to accept GST

2.9.2 Perception of Fairness of GST and attitude

Perception of GST fairness in this study only looks at general fairness and exchange

fairness. Previous study, mainly Saad (2010) advocated that positive fairness

47

perceptions are the antecedent of a positive attitude. This would mean that taxpayers,

or in this case, consumers with positive perceptions on the fairness of the GST

system are more likely to have positive attitudes towards the tax system and

consequently encourage them to accept it. According to Strumpel (1986), positive

attitudes resulting from perceived fairness of the tax system play an important part in

the level of tax compliance within a nation (Tan & Chin-Fatt, 2000). In the 1960s

itself, Strumpel’s fiscal psychology identified ‘willingness to cooperate” as a

variable in their model. This variable relates individual’s perception of tax fairness

and attitudes towards GST (Devos, 2014).

Tax burdens which are perceived as generally fair by taxpayers is an indication of

the taxpayers’ satisfaction with the current tax system which will result in enhanced

positive attitude towards the system. (Thomas, 2012; Seidl & Traub, 2001)

Therefore the study hypothesizes that:

H2: Perception of GST fairness will positively influence attitude to accept GST

2.9.3 Level of GST Knowledge and Perceived Behavioral Control

Knowledge comes in handy because individuals have an increased perception of

uncertainty when faced with a situation of change. (Saad, 2010; Palil,2010). An

understanding of the events causing the change directly impacts the level of stress

experienced by those involved in change efforts and the extent to which they support

48

change (Tetrick and LaRocco, 1987).Given evidence that tax knowledge affects

understanding of taxpayers, an obvious next that has been raised by previous

researchers (Saad, 2010; Harris, 1989) is whether enhancement of tax knowledge

will increase perceived behavioural control. In order to enhance the perceived

behavioral control in purchasing the right products, an individual’s high objective

product knowledge becomes an important factor of consideration (Chiou, 1998). The

same concept can be applied in this context. To understand the events that had

caused the implementation of GST, consumers will need to have the necessary tax

knowledge be it technical or practical.

Perceived behavioral control deals with how taxpayers perceive relative easiness and

difficulty in complying with tax obligations. As taxation is inherently a complicated

matter, it is more likely that taxpayer’s control over non-compliance with tax

obligations is influenced by resources. Based on this argument, it is appropriate to

investigate the impact of tax knowledge (resources) on perceived behavioral control.

Therefore, it is hypothesized that:

H3: Level of GST knowledge positively influences perceived behavioral control

2.9.4 Self-efficacy and Perceived Behavior Control

As a comparatively new concept in academic research, self-efficacy is derived from

the social cognitive theory advanced by the social psychologist Albert Bandura

49

(1986, 1997). In his words, self-efficacy is a state of self-regulation according to

which individuals develop self-disciplined behavior and seek to improve their

performance.

From previous study, (Guthrie and Schwoerer, 1994; Kulviwat, 2014; Ben-Ami et

al.,2014) self-efficacy has been viewed as a criterion for the individuals in question

to determine if they had the skills or ability to deal with the challenges arising from

the behavior. Previous research has indicated that self-efficacy is positively

associated with an individual’s disposition to participate in tasks and cope effectively

when faced with task-related difficulties (Schmidt & Karsten, 2015). Droomers et al.

(2004) singled out self-efficacy as an antecedent to perceived behavorial control and

proved it to be significant.

Raghuram et al.’s (2003) study of telecommuters supports these propositions with

those higher in self-efficacy reporting better adjustment and greater use of

structuring behaviors (Beauregard, 2012).

Consumers who are confident with their ability to manage the challenging lifestyle

post-GST implementation will have a higher perception of their control over the

behavior in question. This will lead to acceptance of GST compared to consumers

with lesser level of self-efficacy (or level of confidence).

The study hypothesizes that:

H4: Self-efficacy will positively influence perceived behavioral control

50

2.9.5 Attitudes and Level of Acceptance

According to theory of planned behavior, attitudes are believed to have a direct

effect on behavioral intention, in this scenario, the level of acceptance. Janzen (1988)

has named attitude as an important criteria in predicting and describing human

behavior. In the 1960’s, the “tax mentality” concept was introduced by Schmolders.

This concept concluded that the more positive a taxpayer’s attitude towards paying

tax, the greater their inclination to pay tax (Devos, 2014). Many researchers

following that have found attitudes to be a significant predictor of behavioral

intention (Jones, 2009; Teo & Lee, 2010; Hai & See, 2011; Bidin & Shamsudin,

2013) Beliefs of individuals influence their attitude towards an outcome which goes

on to influence their intention to perform the behavior. (Vijayasarathy, 2003) In the

context of taxation, Hanno and Violette (1996) used TRA as a theoretical basis and

reported that attitudes had a significant relationship with tax compliance intention.

In another study, Loo et al., (2007) also reported that attitudes towards tax system

positively influenced compliance behavior. They hypothesized that a positive

attitude towards the tax system would encourage taxpayers to comply. This can also

be translated as the positive attitude towards a tax system being able to encourage

consumers to accept the tax system itself. The same concept was advocated in an

Asian context by Damayanti (2012) in Indonesia and Boonyarat et al. (2014) in

Thailand.

51

Therefore, this study hypothesizes that taxpayers with positive attitudes towards the

tax system will increase their level of acceptance of GST.

H5: Attitudes towards GST are positively related to level of acceptance of GST.

2.9.6 Subjective Norms and Level of Acceptance

According to theory of planned behavior, subjective norms also influence behavioral

intention directly. In previous studies, many have found a significant effect of

subjective norms on behavioral intention (Hanno&Violette, 1996; Shimp & Kavas,

1984; Hai & See, 2011).An assessment of factors affecting compliance from 1986 to

1997 reveals agreement with peers as a significant influencer (Richardson & Sawyer,

2001).

In the context of taxation, Hanno and Violette (1996) found a significant and positive

effect of subjective norms on behavioural intention. Similarly, Bobek (1997) found

that the effect of subjective norm on compliance behaviour in a business simulation

scenario was significant. A comparative study in Australia, Singapore and the US by

Bobek and Hatfield (2003) also concluded that subjective norm is an influential

factor in explaining tax compliance behaviour. In an Indonesian taxation system

study by Damayanti (2012), subjective norm was the most influential factor for tax

compliance reflecting the collectivist culture in her country.

52

In the context of GST acceptance behavior, this study expects that subjective norms

will positively influence consumers’ level of acceptance of GST. Therefore, the

study hypothesizes that:

H6: Subjective norms are positively related to level of acceptance of GST

2.9.7 Perceived Behavioral Control and Level of Acceptance

Perceived behavioral control reflects an individual’s perception on the ease or

difficulty in performing a particular behavior (Ajzen, 1991). An individual with high

perceived behavioral control will be more likely to perform the behavior in context

than an individual with lower perceived behavioral control. A relevant illustration

would be that individuals who have high perceived behavioral control over

performing a daily physical exercise are more likely to do the exercise compared to

those with lower perceived behavioral control (Ajzen, 2006). Perceived behavioral

control has significant influence on intentions to behave based on assumption that

perceived behavioral control by an individual will have suggestions on the person's

motivation (Ajzen, 2002). In tax compliance behavior research, when a taxpayer

believes that he or she can successfully complete and file the tax return forms with

Inland Revenue without any mistakes, the person seems to have a high perceived

behavioral control and is more likely to comply with their tax obligations (Saad,

2010).

53

In this study, perceived behavioral control is the individual’s assessment of the

barriers they are likely to face and their ability to overcome those barriers when

accepting GST. The higher the individual’s PBC, the more likely the individual is to

accept GST (Dawkins & Frass, 2005; Damayanti, 2012). Bobek and Hatfield (2003)

have concluded that the behavioral control influences intentions based on the notion

that perceived behavioral control by an individual will have effects on the person's

motivation.

Therefore, the study hypothesizes that:

H7: Perceived behavioral control is positively related to the level of acceptance

of GST

2.10 Summary

To recap, this study is using the Theory of planned behavior to examine the factors

that influence consumers’ acceptance of GST which will be implemented on 1st April

2015. The theoretical framework of this study is made up of four antecedent

variables, 3 independent variables and one dependent variable.

The dependent variable is the level of acceptance of GST measured through the

intention to accept GST. The antecedent variables are Rule Observance Behavior,

Perception of GST fairness, Level of GST knowledge and Self-efficacy.

54

The independent variables are Subjective Norms of consumers, Attitude of

consumers and Perceived Behavioral Control of consumers. These were in line with

the components of the Theory of Planned Behavior.

55

CHAPTER 3

RESEARCH METHODOLOGY

3.0 Introduction

This chapter illustrates the methodology of the study which will comprise of research

design, type of study, population and unit of analysis, sampling size, sampling

method, data collection methods and data analysis methods.

3.1 Research Design

This is a quantitative research and thus, the research will be using primary data

obtained by means of questionnaire answered by selected respondents. The research

approach here is based on the deductive method as we are applying an established

theory in a current existing issue (Sekaran, 2003) In the deductive approach, theory

and hypotheses generation is the first step followed by operationalization of the

concepts in the theory and scientific assessments of the concepts (Lancaster, 2005)

In the next section, we will explore in detail the population of the study, sample size,

sampling method and measurements of variables in the questionnaire.

3.1.1 Type of Study

The study in essence is a correlational study as it examines the relationships between

the before-mentioned antecedent variables, independent variables and dependent

56

variable. Its aim is to study the extent of influence the antecedent variables have on

the independent variables and consequently the impact of the independent variables

on the dependent variable (Haron et al., 2011).

This study can also be considered as an exploratory study as it is the first to explore

the factors influencing acceptance of GST of consumers in Malaysia using the

Theory of Planned Behavior since the official announcement of its implementation

on 1st April 2015. Exploratory studies come in handy when researchers are

interested in examining possible relationships in the general aspect and allow the

method and the data to determine the type of relationship between the variables (Hair

et al., 2006).

The study setting is non-contrived since its respondents are to answer on issues that

are already existent in their natural environment. There are no man-made or lab

experiments.

The time horizon of this study is cross-sectional as it will be conducted over a period

of three months, in order to answer the research questions in place (Haron et al.,

2011).

3.1.2 Population and Unit of Analysis

As the GST is a consumption tax and not income tax, the study aims at

understanding the acceptance level of consumers in Malaysia. This covers a huge

population of Malaysia as consumers refer to any individual who purchases goods or

57

services for personal, domestic or household purpose, use or consumption. Hence, to

conduct a more specific study, the population of this study has been narrowed down

to Malaysian consumers who are currently pursuing their Masters in Business

Administration.

This is because MBA students are highly representative of the categories of

individuals portrayed in the Merdeka Center survey with high levels of unacceptance

of GST. The Merdeka Center survey had portrayed that there was a higher level of

unacceptance amongst private sector employees, self-employed individuals and

those involved business. Moreover, the age groups comprising of those in their

twenties and thirties showed a higher level of unacceptance. (Merdeka Center, 2015)

Majority of MBA students fall in those categories of age and occupation (Tay,

2001).

Besides that, the research requires the respondents to have a clear understanding of

the mechanism of GST, and it is assumed that MBA students would have the basic

understanding required to attempt the technical questions in the questionnaire. It

would be interesting to study if the assumption is in fact true, that MBA students do

have the basic knowledge on the mechanisms of GST. Without this control, there

will be too many variations in the respondents affecting the findings of this research.

With these focused respondents, the study will also be able to test common

assumption that the lack of GST acceptance is due to lack of knowledge (Borneo

Post Online, 2015; Jalil, 2015; GST Malaysia Info, 2014)

58

The success of the research can only be ensured if the sample taken is highly

representative of the population targeted. To obtain a representative sample of the

population mentioned, MBA students from Universiti Sains Malaysia, Universiti

Kebangsaan Malaysia, Universiti Malaya and Universiti Putra Malaysia were

targeted as all these public universities are Research Universities since 2006 and

have established business schools. (Ministry of Education, 2012). Although

Universiti Teknologi Malaysia is also a Research University, it was not included in

this list as it had only joined the research university cluster in 2010. (Ministry of

Education, 2012) Hence, to maintain the similarity within the selected universities

only the four first research universities were selected.

Thus, the unit of analysis is Malaysian individuals who are consumers in general and

are currently pursuing their Masters in Business Administration in either USM,

UKM, UM and UPM. Consumers as defined before are those who acquire goods or

services for personal, domestic or household purpose, use or consumption. Hence, all

respondents are treated as consumers.

3.1.3 Sample Size

The minimum number of respondents is to be at least ten times the total number of

variables to be analyzed.(Hair et al, 2014) Thus, in this case, it would mean that

since there are 8 variables to be examined, a sufficient sample size would be, 8*10 =

80. As the rate of response is usually 20%, at least 400questionnaires must be

distributed (Voorhis & Morgan, 2007).

59

3.1.4 Sampling Method

The sampling technique used is non-probability sampling; convenience sampling,

namely snowball sampling. Convenience sampling is a non-probability sampling

technique and respondents are selected because of the convenient accessibility and

proximity to the researcher (Haron et al, 2011)

Snowball sampling is when a group of people recommend participants for a study

based on personal acquaintances. Those participants then recommend other

participants hence building up like a snowball rolling down a hill. If the researcher is

opting for respondents who have to meet certain criteria to participate, then snowball

sampling can be used to ease data collection. It is known as a purposive sampling

and if executed correctly can approximate a random sample (Cohen & Arieli, 2011)

3.1.5 Data Collection Techniques

An online questionnaire was designed using the survey monkey online template. Link

to the online survey was sent out through email with a cover letter to one contact

person in each university-UKM, UPM and UM.As for USM, the email link was sent

out personally by the researcher. Details of the contacts are as follows:

60

Table 3.1: Profile of Contacts

University Number of Contacts Contacts’ profile

UPM 1 2nd

Year MBA student, part time

UM 1 3rd

Year MBA student, full time

UKM 1 3rd

Year MBA student, full time

The cover letter explained the objective of the study and assured confidentiality of

the responses. The cover letter is attached in Appendix D. They were then requested

to send out the email to their MBA batch mates. A period of two weeks was given

for the online survey link to be distributed. Reminder emails were sent after two

weeks to the contacts who in turn sent out reminder emails to their contacts to ensure

that the data collection fulfills the sample size needed.

An online survey is the most effective method to reach out to respondents of

different geographical areas. Hence it is the most logical instrument of data

collection for respondents from UKM, USM, UM and UPM who vary in their

geographical locations.

3.1.6 Survey Instrument

The instrument or tool used for this study is questionnaire. The questionnaire was

divided into FIVE sections, Section A to Section E in order to systematically collect

data for the measurement of the latent variables of the theoretical framework. There

are a total of 67 items to be answered in the questionnaire. Section D and E were

61

self-developed by the researcher to obtain relevant information for this research.

Detailed discussions of the sections and its construct will be discussed in the next

sections.

3.1.6 Questionnaire Design

A set of the questionnaire is attached in Appendix D.

The questionnaire was designed with the intention to gather relevant data for the

purpose of the study. Hence, the questionnaire was constructed with the theoretical

framework as a base, ensuring that each latent variable in the theoretical framework

is measured scientifically.

The questionnaire consists of FIVE sections, Section A to Section E, and clear

instructions on how to attempt the questions are stated in each section. This will

assist the respondents in understanding what is required from them.

A Likert scale is highly useful in ranking responses for each indicator of a latent

variable. It is ordinal in nature and is usually based on a five-point or seven-point

scale. (Likert, 1932) Previous research which measured the perceptions of the latent

variables in this study used the Likert Scale. (Saad, 2010; Sarina; 2007; Ben-Ami

2014), For the purpose of this study, a five-point Likert-type scale (Strongly

Disagree, Disagree, Neutral, Agree and Strongly Agree) was used for measurements

in Section B.

62

Section A of the questionnaire was entitled Demographic Profile and aimed at

gathering relevant demographic characteristics of the respondents to be used for

profiling of the respondents. The main aim is also to ensure the respondents fulfill

the criteria of the unit of analysis mentioned, they are; Malaysians currently pursuing

Masters in Business Administration.

Section B measured the antecedent variables, independent variables and dependent

variable based on a 5 point Likert Scale from (1) strongly disagree to (5) strongly

agree. Section B has 7 subsections which represent three antecedent variables; Rule

Observance Behaviour, Perception of GST Fairness and Self-Efficacy, three

independent variables; Attitude, Subjective Norms and Perceived Behavioural

Control and the dependent variable; Level of Acceptance.

Section C and Section D was fully dedicated to the measurement of Level of Tax

knowledge pertaining Goods and Services Tax (GST). Section C comprised of ten

technical questions which required the respondents’ agreement on the statements

given by circling YES if they agree and NO if they disagree.

Section D was made up of 5 short scenarios. The scenarios depicted real-life

situations and the respondents were required to identify the GST supply category and

the corresponding tax rate. The respondent had to tick the box with the right answer

for Part A and Part B for each scenario. This is to evaluate the respondent’s practical

knowledge of GST.

63

Section E consists of open-ended questions which aimed at gathering valued

opinions and feedbacks of respondents. This section would be pivotal for Chapter 5

when the researcher provides recommendations based on the findings from Chapter

These open-ended questions provide the respondents with an avenue to give relevant

and honest opinions and views.

Table 3.2 Summary of Questionnaire Sections

Section Purpose No of

measurement

items

A Demographic Profile 12

B Measurement of Antecedent variables, Independent

variables and Dependent variable using Likert scale

30

C Level of Tax Knowledge through technical questions 10

D Level of Tax Knowledge through scenario based

questions

10

E Open ended questions on opinions 5

Total 67 Items

3.2 Measurements of Variables

This section will indicate the source of the measurements used for the various latent

variables in the various sections of the questionnaire.

3.2.1 Measurement of Antecedent Variables and Independent Variables

Measurement of attitude has been adapted from Bhattacherjee (2000) and Loo et al

(2007) with modifications made to the content to suit the study of acceptance of

GST.

64

Attitude has two antecedent variables; Rule Observance Behaviour and Perception of

fairness of GST. Measurements of Rule Observance Behaviour was adapted from

Jeffrey &Weatherholt (1996) whilst perception of tax fairness was adapted from

Saad (2010) using the instruments of measurements for general fairness and

exchange fairness and customized to GST context.

Perceived behavioral control has two antecedent variables; self-efficacy and level of

tax knowledge. Measurements for Perceived Behavorial Control were adopted from

Ajzen (2013) modified for the GST context. For self-efficacy, measurements were

adapted from Kulviwat (2014) and Ben-Ami (2014) with adjustments to suit the

subject matter of GST. The measurement of level of tax knowledge will be discussed

in detailed in the next section.

Subjective Norm was adapted from Bhattacherjee(2000).The content was also

modified to suit this study and broadened to include peers and family influences.

3.2.2 Level of Tax of Knowledge Scoring

Level of tax knowledge measurements were conducted through two sections,

technical knowledge and practical knowledge. The technical knowledge concept was

adapted from Saad (2010), however, the items in the questionnaire were modified to

suit the study of acceptance of GST which means the questions have to test

consumers' technical knowledge on GST. The construct of these questions were

based on Royal Malaysian Customs Department’s Frequently Asked Questions

65

section on GST. A Yes or No question format was used for the tax knowledge quiz.

For each correct answer, one point will be given, with maximum marks any one

person can obtain being set at ten for the ten questions.

For the practical knowledge, real-life scenarios were self-developed for the

respondent to answer. For each scenario, the respondent chooses the correct GST

supply category- standard, zero or exempt (Part A) and the corresponding tax rate –

6%, 0% or none (Part B). One mark is given for each correct answer in Part A and

Part B with the maximum marks any one person can obtain being 10 marks, 5 marks

each for Part A and Part B.

For the level of tax knowledge section, the correct answers of the respondents are

calculated and a grade over 100% is given. For instance, if a person answers 5

questions correctly in Section C and 5 questions correctly in Section D, his/her total

marks would be 10 out of 20. The total marks for the Section C and Section D is 20.

Hence, his/her marks would be 10/20 * 100% = 50%. Thus continuous data is

collected for this section and analyzed.

3.2.3 Measurement of Dependent Variable

The dependent variable of level of acceptance of GST is measured using instruments

developed on our own based on the adaptations from Bhattacherjee (2000). It is on

5-point Likert scale with 2 items. As such a study had not been conducted,

measurements of this latent variable could not be validated based on the review of

literature.

66

In this study, the context of the adapted questions were changed to suit the subject

matter. For instance for the questionnaire items for the dependent variable on level of

acceptance, ‘I intend to accept GST’ is used instead of the original version of ‘I

intend to use the e-commerce services’ as stated in Bhattacherjee (2000). Another

example would be that for perception of GST fairness, in Saad (2010) it was stated

that ‘I believe everyone pays their fair share of income tax under the current income

tax system’. However, for the purpose of this study adaptations were made and the

question was ‘I believe everyone pays their fair share of tax under the proposed GST

system’.A summary of the measurement of variables is presented in Table 3.4.

Table 3.3 Sources of measurement of variables

Section Variable No of

items

Source

A Demographic Q1- Q12 Self-developed

B Antecedent Variables to Attitude

(i) Rule Observance Behavior Q1-Q5 Adapted- Jeffrey

&Weatherholt (1996)

(ii)Perception of fairness Q1-Q6 Adapted -Saad (2010)

Independent Variables

(iii) Attitude Q1 – Q2

Q3 – Q5

Adapted–Bhattacherjee(2000)

Adapted- Loo et al (2007)

(iv) Subjective Norms Q1 – Q5 Adapted- Bhattacherjee(2000)

Antecedents to Perceived Behavioral Control

(v)Self-Efficacy Q1- Q2

Q3- Q4

Adapted Kulviwat (2014)

Adapted Ben-Ami (2014)

(vi)Perceived behavioral

control

Q1-Q3 Adapted – Ajzen (2013)

Dependent Variable

(vii) Behavioral Intention Q1-Q2 Adapted – Ajzen (2013)

C Antecedents to Perceived Behavioral Control

(viii)Level of Tax Knowledge

– Technical Questions

Q1-Q2

Q3-Q10

Adapted- Saad(2010);

RMCD GST FAQ(2014)

D (ix) Level of Tax Knowledge

– Scenario-based Questions

Q1-Q10 Self-developed

E (x) Open-ended questions Q1-Q5 Self-developed

67

3.3 Pre Test

To determine the consistency and reliability of the questions in measuring the

variables, a pre-test is conducted. The pre-test is also aimed at assessing the

respondents’ understanding of the questions in the questionnaire designed.

Five USM MBA students and 2 Academicians were selected to read through the

questionnaire and a discussion on their understanding of the questions was held.

Based on their feedback, some questions were restructured (without affecting the

content) and supportive information was added for clarity. An example of this was

for the behavioral intention measurement questions where a few of these respondents

did not understand the meaning of acceptance of GST. Hence, the researcher added

in an informative line explaining that acceptance of GST refers to the level of

agreement of the respondent with GST despite it being a mandatory implementation

and not the act of paying GST.

Following that, two experts in the area of Goods and Services Tax, Partners of

Stephen & Co (A MIA registered Audit & Tax firm were approached to help assess

Section C and Section D on level of tax knowledge. They were able to understand

and were in agreement with the questions asked, thus no changes were made.

68

3.4 Pilot Test

Finally, after fine-tuning of the questionnaire, 32 respondents were selected for the

pilot study based on convenience sampling. Out of the 32 respondents, 29 were USM

students, 1 was an UKM student, 1 was a UM student and 1 was a UPM student.

Besides analyzing their responses for consistency and reliability, the clarity of the

questionnaire was also confirmed with the respondents. All the 32 respondents

confirmed that they fully understood the questions when answering them.

These pilot study respondents were clearly identified and when the questionnaire

was distributed for the collection of actual data, extra precaution was taken to make

sure that the online questionnaire was not attempted by these individuals.

The completed questionnaires were then analyzed using the PLS-SEM version 3 to

obtain the Cronbach Alpha values. These values are indicators of the reliability and

internal consistency of the variables.

The framework will be analyzed in three models for a more systematic study of the

reliability and validity of all latent variables as well as their relationships. With the

theoretical framework drilled down to the antecedents, a more conducive discussion

can be made on the antecedents to attitude and perceived behavioural control. Details

of each model are presented in the table below:

69

Table 3.4 Details of Models for analysis

Model Model Name Latent Variables

1 Antecedents to Attitude to

accept GST

1. Perception of GST fairness

2. Rule Observance Behaviour

3. Attitude to accept GST

2 Antecedents to Perceived

Behavioural Control

1. Self-efficacy

2. Level of GST knowledge

3. Perceived Behavioural Control

3 Whole Model 1. Perception of GST fairness

2. Rule Observance Behaviour

3. Self-efficacy

4. Level of GST knowledge

5. Attitude to accept GST

6. Perceived Behavioural Control

7. Subjective Norms

8. Level of Acceptance

Hence, as there are three models for this study, the Cronbach’s Alpha was measured

for each variable in all three models. The value of the Cronbach’s Alpha has to be

above 0.7 for the variable to be accepted as reliable and consistent. (Hair et al.,

2014).

The table in the next page presents the values for the variables in all three models.

70

Table 3.5 Cronbach’s Alpha for pilot study

Model 1

Variables Cronbach’s Alpha

Attitude 0.926

Perception of Tax Fairness 0.936

Rule Observance Behaviour 0.894

Model 2


Level Of Tax Knowledge 1.000

Self-Efficacy 0.833

Perceived Behavioural Control 0.878

Model 3


Attitude 0.898

Level of Acceptance 0.868

Level of Tax Knowledge 1.000

Perceived Behavioural Control 0.878

Perception of Tax Fairness 0.920

Rule Observance Behaviour 0.848

Self-Efficacy 0.833

Subjective Norms 0.885

As can be seen from the table, all the Cronbach Alpha values are above 0.7

indicating that the variables are consistent and can be reliably used for the purpose of

this study. For Level of Tax Knowledge the Cronbach Alpha is valued at 1 because it

is measured by the total marks obtained through 20 technical and practical questions.

On the other hand, the other variables are measured by 2-6 question items on a

Likert-scale.

Through this pilot study, we will be able to avoid inconvenience to respondents as

well as minimize the removal of questions through factor and reliability analysis of

the actual data collection.

71

3.5 Data Analysis

All data collected has to be analyzed to attain the research objectives, answer the

research questions and accept or reject the hypotheses developed. The data were

analyzed using two statistical softwares; IBM Statistical Package for Social Sciences

(SPSS) version 22 and Smart Partial Least Squares (SmartPLS) in Structural

Equation Modelling (SEM) version 3.0.

SPSS will be used to analyze and summarize the respondents’ demographic profile

whereas SmartPLS will be used to examine and conclude on the hypotheses

developed in the earlier chapter based on the theoretical framework and its variables.

The data will be examined in 3 models as explained in the above section. SPSS is

also used in Chapter 5 to further examine the demographic profiles to investigate

future research potentials.

3.5.1 Structured Equation Modeling (SEM)

SEM is a relatively new approach for testing multivariate models with empirical

data. It incorporates both factor analysis and path analysis indicating that it deals

with concepts and observed variables. The concept of SEM has been widely used for

the past 20 years in anthropology, education, political sciences and economics.

(Wold, 1980; Fornell&Bookstein, 1982) SEM enables the integration of both

observed variables, also known as indicators, and latent variables, also known as

constructs. Constructs cannot be measured directly and has to be measured through

72

one or more observed variables or indicators. (Rosipal & Kramer, 2006) In a

nutshell, as a second generation data analysis technique, SEM allows researchers to

develop unobserved Latent Variables, model relationships among Latent Variables,

identify observed variables for each latent variable and statistically test theoretical

assumptions against empirical data (Hair et al., 2014)

3.5.2 Covariance-based SEM (CBSEM) against Variance-based SEM with

Partial Least Squares (PLS)

Covariance-based SEM (CBSEM) is the most well-known SEM technique. This

technique aims to minimize the differences between the covariances between the

samples and the theoretical model. The statistical objective here is to obtain

goodness-of-fit but this on its own does not imply a good model. (Rosipal& Kramer,

2006)

In comparison, the variance-based partial least squares technique aims at obtaining

determinate values of the latent variables for analytical purposes. Using the scores of

all latent variables, the PLS algorithm maximizes the proportion of variance in the

dependent variable that is explained by the predictor variables. Hence, this indicates

that VB-SEM with PLS is developed to maximize prediction and not goodness-of-

fit. (Rosipal & Kramer, 2006)

For this study, the software of SmartPLS 3.0 is used to apply variance-based SEM

with PLS analysis. With the collected data, the PLS algorithm is employed to

73

estimate the path coefficients and other relevant parameters in order to maximize

explained variance or minimize unexplained variances. (Hair et al., 2014) Following

that, bootstrapping is conducted to examine the coefficient for the significance of the

path modelling. In this step, random samples with replacement are extracted from the

data and used to estimate the path model numerous times under slightly changed data

patterns. This resampling approach is used to approximate the t-value which is

highly relevant information to the study. (Hair et al., 2014) Significance of t-values

is as depicted:

i) t-value of 1.645 is significant at a significance level of 5% (α = 0.05, one-

tailed)

ii) t-value of 2.33 is significant at a significance level of 1% (α = 0.01, one-

tailed)

One-tailed test is used for this study as the hypotheses are all directional indicating

clearly that is a positive relationship between the antecedent variable and the

independent variable as well as between the independent variable and the dependent

variable.

Amongst the justifications of using VB-SEM with PLS for this research are (Wold,

1980; Fornell& Bookstein, 1982):

i) The model of this study has formative constructs

ii) The data is not normally distributed

iii) This is an exploratory study

74

iv) The model is based wholly on one theory, the Theory of Planned Behaviour

3.5.3 Statistical Analysis

IBM SPSS Statistics version 22 is used here to assist with the analysis of other data.

It is a comprehensive statistical tool for analyzing data in order to generate reports,

charts, trends, plots and statistic which are relevant to the user. It is highly capable of

handling a large amount of data and transactions making it a common statistical tool

in the practical world. (Hair et al., 2014) In this research, SPSS is used for the

demographic analysis.

3.5.4 Demographic Analysis

An analysis on the demographic characteristics of the respondents is conducted. The

summary of the analysis enables the researcher to be aware of the proportion of

respondents in the various demographic categories, such as age, mode of study,

institute of education, position and income level.

3.5.5 Common Method Bias

Common method bias takes place when measurements of all latent variables;

antecedent, independent and dependent are gathered from different sources.

Common method bias can occur when using similar scales with the same number of

response options. (Podsakoff et al., 2003) Variance resulting from the measurement

75

method will lead to systematic errors and distort the relationship amongst the

theoretical constructs (Chandra et al., 2011). Common method biasness is possible in

this research as all variables are measured in one self-administered questionnaire

conducted during the same period.

Principal Component Analysis is conducted on all question items of the variables in

the research under the Harman’s single-factor test of common method biasness. This

analysis will enable the researcher to determine if any one general factor explains

majority of the variance in the model (Podsakoff et al., 2003). Common method bias

is not present when the first component accounts for less than 50% of the total

variance and the last, largest component accounts for more than 50% of the total

variance. When common method bias is not present, it indicates that the

questionnaire is authentic, scientific and there is no ambiguity in the questionnaire

3.5.6 Descriptive Analysis

Before proceeding to the in-depth analysis of the relationship between the constructs,

descriptive analysis is conducted. Through descriptive analysis, the mean and

standard deviation of all the questions can be obtained. This will enable the

researcher to understand the overall response of the respondents towards the various

question items. (Podsakoff et al., 2003) For the purpose of this study, this analysis

will answer the first research question on the acceptance level amongst consumers

with GST.

76

Correlation between the various constructs was calculated to determine if

multicollinearity exists. Multicollinearity is present when two or more predictor

variables are highly correlated which will affect the accuracy of the model.

(Podsakoff et al., 2003)

3.5.7 Goodness of Data Test

The goodness of the data is determined by examining the data for reliability and

validity. A measure may be correct but not consistent which is why both reliability

and validity testing must be conducted. The quality of the results of this study is

highly influenced by the results of these tests. (Sekaran, 2000)

3.5.8 Factor Analysis

Factor Analysis is conducted to examine and outline the underlying structure

amongst the constructs in the analysis. It is a data reduction method where an

attempt is taken to reduce the number of original factors to a smaller set of factors

that measure the construct without loss of information (Hair et al., 2010) Factor

Analysis is conducted for the following reasons:

1) To determine the correlation between the many questions used to measure a

variable. The more correlated the questions are, the better they measure the

variable together.

77

2) It is a dimension reduction method in which it removes questions in the

variable measurement which is not appropriate.

3) It is used to measure the validity of subjective measurement, latent variables.

There is confirmatory factor analysis (CFA) and exploratory factor analysis. CFA is

used to validate the factor structure of a set of observed variables. The correlations

between the factors are validated to ensure that the constructs are conforming to the

pre-established theory. Confirmatory Factor Analysis leads to the validity testing of

the constructs. (Hair et al., 2010)

3.5.9 Validity Analysis

Validity is defined as the capability of a scale to measure what it is supposed to

measure (Zikmund, 2003). It is important as mentioned in the goodness of data

section to ensure the quality of the results obtained from these measurements. The

construct validity is examined to determine the quality of the results achieved from

the measurements undertaken fitting the theory presented in the research. (Sekaran,

2000) The management theory selected should support the framework. Construct

validity is examined through convergent validity and discriminant validity.

Convergent validity tests if the measures of the same construct are highly correlated.

This is done by looking at the main outer loadings of each question item within the

construct, the Composite reliability and the Average Variance Extracted (AVE)

value. The main outer loadings and AVE must be above 0.5 whilst the composite

reliability value must be higher than 0.7 in order to confirm the presence of

convergent validity. (Hair et al., 2014)

78

Discriminant validity concludes that the measures of a construct are not highly

correlated with other constructs. It is employed to ensure that the constructs which

are not supposed to be related to each other are in fact not related. (Hair et al., 2014)

Discriminant validity is examined through a construct where none of the inter-

construct correlations are greater than the square root value of the AVE of the

constructs. The discriminant validity table can be seen in Chapter Four. (Fornell &

Laracker, 1981; Hair et al., 2014)

3.5.10 Reliability Analysis

Reliability is the extent to which measures are influenced by random error and

produce consistent results. (Zikmund, 2003) Reliability is also referred to as internal

consistency within the question items of the variables. Higher reliability values

indicate that the question items are consistently measuring the same latent concept.

(Shin, Collier & Ilson, 2000)

Reliability is measured using Cronbach’s Alpha which is designed to examine the

average correlation of question items to assess its reliability. (Santos, 1999)

Composite reliability also works on the same concept as Cronbach’s Alpha. For both

Cronbach’s Alpha and Composite reliability, the value has to be more than 0.7 in

order for the scale to be maintained.

79

3.5.11 Goodness-of-Fit (GoF)

The Goodness-of-Fit (GoF) statistical model examines how well the model fits into a

set of constructs. (Tenenhaus, Vinzi, Chatelin&Lauro, 2005; Maydeu-Olivares &

Garcia-Forero, 2010) It is determined through the following formula:

GoF= √ (AVE x R2)

The GoF value is between 0 and 1 and the higher the value, the more accurate the

path model estimates would be. (Abd-El-Fattah &Fakhroo, 2012) GoF values can be

used to determined effect sizes and any value above 0.36 has a large effect size.

(Cohen,1988: Abd-El-Fattah & Fakhroo,2012)

3.6 Summary

In summary, this chapter explained that this is a quantitative research and that the

research instrument is a self-administered online questionnaire. The unit of analysis

of this research is Malaysian Masters in Business Administration (MBA) students

from USM, UKM, UM and UPM.

The sample size has to be a minimum of 80 respondents and each section of the

questionnaire and its measurement operationalization was discussed. Sources of the

question items of each construct were also discussed. The pilot study included clarity

testing, expert’s review and reliability testing on responses by 32 respondents.

80

The data collected will be analyzed using IBM SPSS Statistics and SmartPLS

version 3.0 (VB-SEM with PLS) in 3 models. These two statistical tools were

examined and the various analyses that will be conducted were discussed. Amongst

them were demographic analysis, descriptive analysis, common method bias, validity

testing, reliability testing and goodness-of-fit measurement. The results of the

abovementioned analyses will be presented in the next chapter.

81

CHAPTER FOUR

RESULTS

4.0 Introduction

This chapter sets out to present the results of data analysis. A comprehensive data

analysis is performed in order to determine the dimensionality of items under the

respective constructs, the reliability of the instruments, the measure of fitness for the

proposed model, the goodness of a fitted model, and the appropriate statistical

analysis procedures to address the research questions as well as to test the

corresponding research hypotheses. Hence, the organization of this chapter is as

follows:

1. The demographic variables are examined to provide a clear description of the

sample characteristics.

2. To measure the goodness of data, validity and reliability testing is conducted.

3. Validity testing involved Confirmatory Factor Analysis where convergent

and discriminant validity is tested. The required data for this analysis can be

obtained from the PLS Algorithm step in SmartPLS 3.0

4. Structural Equation Modeling (SEM) statistical technique will be used to test

the hypotheses developed from the theoretical framework.

5. The Global fit or Goodness of fit value calculated to confirm suitability of

model.

82

4.1 Response Rate

Table 4.1: Summary on the Response Rate

University Number of MBA

students in batch

Number of

respondents

Unusable Usable

responses

Response

Rate

USM 212 94 8 86 40.57%

UPM 145 53 2 51 35.17%

UM 194 46 3 43 22.16%

UKM 200 51 3 48 24.00%

TOTAL 751 244 16 228 30.35%

As mentioned in the third chapter, the unit of analysis of this study is Malaysian

consumers pursuing their Masters in Business Administration in four universities;

USM, UPM, UKM and UM. The researcher had sent to one contact in UM, UKM as

well as UPM and 200 contacts in USM. These contacts were then requested to send

the link of the online questionnaire to their batch mates. Hence the response rate was

determined in reference to the total number of MBA batch mates for the contacts in

each university. Table 4.1 has concluded that the total response rate is 30.35%

whilst a breakdown displays that USM had 40.57% response rate, UPM had 35.17%,

UM had 22.16% and UKM had 24%. This is clearly explained by the fact that the

researcher had more convenient access to USM students as compared to the other

four universities.

83

4.2 Profile of Respondents

This section displays a summary of the respondents and their socio-demographic

characteristics. The unit of analysis of this study is MBA students from USM, UPM,

UKM and UM.

Table 4.2: Socio-Demographic table – Respondent’s profile

Demographic Categories No. of

Respondents

Percentage

(%)

Age

18-25 9 3.9

26-40 210 92.1

41-60 9 3.9

Total 228 100.0

Gender

Male 106 46.5

Female 122 53.5

Total 228 100.0

Mode

Full Time 32 14.0

Part Time 196 86.0

Total 228 100.0

Institution

Universiti Sains Malaysia 86 37.7

Universiti Kebangsaan

Malaysia 48 21.1

Universiti Malaya 43 18.9

Universiti Putra Malaysia 51 22.4

Total 228 100.0

84

Table 4.2: Socio-Demographic table – Respondent’s profile (continued)

Demographic Categories No. of Respondents Percentage (%)

Employment Not Applicable (Full time) 32 14.0

Self-Employed 8 3.5

Owner of Business 5 2.2

Working in an

Organization 181 79.4

Unemployed 2 9

Total 228 100.0

Position

Not Applicable (Full time) 32 13.6

Top Management 3 1.3

Finance/ Accounts

Manager 7 3.5

Marketing/ Sales Manager 24 10.5

Technical/ Operations

Manager 22 9.6

Others: Executives 92 40.4

Others: Teachers 7 3.1

Others: Professionals 41 18%

Total 228 100.0

Income

Not Applicable (Full time) 32 13.6

RM 1000 - RM 5000 126 55.3

RM 5001- RM 10000 34 15.4

RM 10001 - RM 15000 33 14.5

> RM 15000 3 1.3

Total 228 100.0

Table 4.2 provides a clear insight of the respondents based on the demographic

questions posed in the questionnaire. Three filter questions were in place to ensure

the respondents match the unit of analysis. The questions were:

1) Question 3 – Nationality

2) Question4 – Current Level of Education

85

3) Question 5 – Please indicate the programme you are enrolled in

Only respondents who were Malaysian and currently pursuing Postgraduate

education in Masters of Business Administration were accepted.

From the table, we can notice that most of the respondents are between the ages of

26 to 40. This is crucial information as we are able to first understand the age group

which is pursuing MBA in this country. More importantly, it is consumers within

this age group that are highly resistant against GST as depicted in Merdeka Centre

Survey. Hence, it is interesting to be able to gather their current, up-to-date opinion

on the same matter.

We can also gather that we have a slightly higher number of female respondents than

male. The respondents also comprise more of part time students are mostly in other

positions besides the ones mentioned in the position list. Amongst the responses

under the category ‘Others’ were technical executives, human resource executives,

engineers, IT executives, pharmacists, teachers and auditors. Higher number of

respondents fell in the RM 1000- RM 5000 income level category.

The findings on the demographic profiles further enhance our study because majority

of our respondents can be categorized as the middle-income group based on their

income level and employment status as well as position. Protests by the general

public and many articles pertaining GST have expressed the concern that the middle-

income group will be the one hit the worst by GST in Malaysia, at least in early

stages. (Zamhari, 2014) With respondents loading on this middle-income group, the

86

results of this research would give us a crystal clear understanding of their opinions

and thoughts in relation to GST making the findings of this research very significant

and applicable.

4.3 Descriptive Statistics and Multicollinearity

Table 4.3 Descriptive Statistics and Correlation Matrix (n=228)

Correlations

Variable

Scale Mean

Standard

Deviation LTK

Ave

ROB

Ave

PTF

Ave

ATT

Ave

SN

AveS

E

Ave

PBC

Ave

LA

VIF

LTK 0-100 76.34 14.33 1 1.16

Ave ROB 1-5 3.66 0.70 .190

** 1

1.41

Ave PTF 1-5 2.86 0.84 .209

**

.449*

*

1 1.95

Ave ATT 1-5 3.14 0.95 .306

**

.500*

*

.637**

1

3.50

Ave SN 1-5 3.00 0.70 .106

.495*

*

.630**

.676

** 1

2.53

Ave SE 1-5 3.24 0.72 .041

.233*

*

.302**

.475

**

.270**

1

1.43

Ave PBC 1-5 2.91 1.02 .143

*

.437*

*

.605**

.783

**

.706**

.488*

*

1 3.55

AveLA 1-5 2.94 1.08

.375*

*

.526*

*

.651**

.824*

*

.64

2**

.371**

.756*

*

1

**. Correlation is significant at the 0.01 level (1-tailed).

*. Correlation is significant at the 0.05 level (1-tailed).

Table 4.3 shows descriptive information on all the latent variables of this research

model. This descriptive information was gathered using the average values of the

latent variables explaining why the variables have the prefix Ave before their

abbreviation, for instance; AveATT which stands for Average Attitude. The

descriptive information includes the scale on which the variable was measured, the

mean of the responses for each variable based on the measurement scale and the

standard deviation. The correlation matrix is also attached followed by the VIF

(Variation Inflation Factor) values.

87

The mean of responses for Level of GST knowledge was 76.34 on a scale of 1 to

100. As the level of tax knowledge is being measured by the score the respondent

obtains out of 20 questions, this mean indicates that the level of tax knowledge

amongst the respondents is high. This confirms the assumption held earlier that

MBA students are expected to have a sound basic knowledge of the Goods and

Services Tax. The standard deviation of 14.33 is small as it is less than 20% of the

maximum point of the scale. It implies that the respondents are agreeing amongst

themselves, confirming once again that MBA students are in fact, in general,

knowledgeable in the technical and practical aspects of GST.

For Rule Observance Behaviour, its mean is 3.66 indicating that the respondents are

those who have the higher tendency to observe and obey rules, believing that rules

will lead the improvement. The standard deviation is low at 0.70 indicating a high

level of agreement in the respondents’ responses.

For Perception of GST Fairness, the mean is quite low at 2.86 indicating

disagreement amongst the respondents. Hence, it can be concluded that the

respondents do not view the Goods and Services taxation system and the

management of taxes collected as a fair system. The standard deviation of 0.84,

being below 1, once again indicates that the respondents are in consensus of this

opinion.

For Attitude, the mean of 3.14 indicates borderline positive feelings amongst

respondents with reference to GST. As an overview, the respondents are only in

88

slight agreement that GST is a wise decision and they feel slightly positive about it.

The standard deviation of 0.95, although being slightly high, still indicates that there

are no significant variations in the responses of participants.

The subjective norms mean is at 3.00 indicating absolute neutrality. Respondents are

neutral on perceptions that their peers and family members would affect their own

views and decisions. This is highly appropriate to the respondents of this study as

they are MBA students who are believed to have a higher intellect hence not relying

on the opinions of others to make decisions. The standard deviation of 0.70 reiterates

that the respondents are collectively disagreeing on this matter.

The mean for Self-Efficacy is 3.24 which indicates that the respondents have only

slightly high levels of self-efficacy. The respondents are not very confident that they

have the required capabilities to manage through the GST. A low standard deviation

for Self-Efficacy confirms that the respondents are agreeing amongst themselves.

The mean for Perceived Behavioural Control is slightly low at 2.91. This indicates

that the respondents feel that there are internal and external constraints on them with

the implementation of GST. A higher standard deviation of 1.02 for PBC displays

that there were significant variations between opinions and that this opinion was not

generalized.

Finally, the mean of Level of Acceptance answers our first Research Question on the

current level of acceptance of GST amongst consumers. Respondents have tilted

towards the disagreement of accepting GST with the mean of 2.94. However the

89

standard deviation is above 1 at 1.08 indicating that there are significant variations

between opinions.

Moving on to the correlation matrix, it is vital in determining the correlation between

the latent variables to ensure multicollinearity does not exist. Multicollinearity will

affect the accuracy of the model and its measurements. A high correlation should

only be present between independent variables and dependent variable and in this

case, antecedent variables and independent variables. This is because a high

correlation between predictor and dependent variables strengthens the relationship

and the pathway between them. From the correlation matrix, it is clear that Attitude

has the highest positively significant correlation with Level of Acceptance of GST (r

= 0.824, p<0.01) followed by Perceived Behavorial Control (r = 0.756, p<0.01).

However, predictor variables should have low correlations in order to eliminate the

multicollinearity effect. The common threshold is that the correlation value between

two predictor variables should not exceed 0.7. From the table above, Attitude and

Subjective norms have a high correlation with Perceived behavioural control, 0.783

and 0.706 respectively. In order to test for multicollinearity, the Variance Inflation

Factoris is tested for the predictor variables. A VIF value of less than five eliminates

any possible existence of multicollinearity. (Groebner et al, 2014) As can be seen

from the table, all VIF values are below 5 confirming that multicollinearity does not

exist in the data set of this research.

90

4.4. Common Method Bias using Principal Component Analysis

Using the principal component analysis, the authenticity of the data is examined.

This test must be conducted in view of the fact that the questionnaire is constructed

based on inputs from the literature of a variety of authors. According to (Podsakof et

al., 2003), Common method bias (CMB) is the bias of measurement error variance

that can affect the validity of relationships assumed between the indicators of

different variables. Harman’s one-factor (or single-factor) test is used here by

running factor analysis with all the question items of the model.

In the current study, 30 components were extracted for 30 question items of the

model using principal component analysis in the first factor run (Appendix H-2).

The indicator on non-CMB is the first and last value of the Rotation Sums of

Squared Loadings Cumulative % in Table 4.4. The first component should be less

than 50% and the last component should be more than 50%. (Podsakof et al., 2003)

In this case, the first seven components significantly explain the variance in the data

and have Eigen value of more than one which is why the cumulative loadings stop at

7th

component. To add on, component 1 is 24.022% which is less than 50% and the

last component, component 7 is 76.110%. These results confirm the authenticity of

the data and that it is scientific in such a manner that the respondents fully

understand what the question is trying to convey.

91

4.5 Data analysis and results

4.5.1 Goodness of measures

The two important criteria used for testing goodness of measures are validity and

reliability. According to Sekaran and Bougie (2010), reliability is an assessment of

how dependable weighing instrument measures any concept it is weighing,

meanwhile validity is an examination of how fit an equipment which is built

measures the certain concept it is intended to weigh. As this study consists of 3

models, the validity and the reliability of each model will be examined separately.

Table 4.4 Total Variance Explained

Component

Initial Eigenvalues

Extraction Sums of Squared

Loadings

Rotation Sums of Squared

Loadings

Total

% of

Variance

Cumulative

% Total

% of

Variance

Cumulative

% Total

% of

Variance

Cumulative

%

1 12.941 41.746 41.746 12.941 41.746 41.746 7.447 24.022 24.022

2 2.671 8.616 50.362 2.671 8.616 50.362 4.108 13.251 37.273

3 2.342 7.554 57.916 2.342 7.554 57.916 3.776 12.182 49.455

4 1.871 6.037 63.953 1.871 6.037 63.953 3.061 9.875 59.330

5 1.491 4.809 68.762 1.491 4.809 68.762 1.996 6.439 65.769

6 1.222 3.941 72.703 1.222 3.941 72.703 1.911 6.164 71.932

7 1.056 3.407 76.110 1.056 3.407 76.110 1.295 4.177 76.110

Extraction Method: Principal Component Analysis.

92

4.5.2 Pathway of Each Model Construct

MODEL 1

Figure 4.1: PLS Pathway for Model 1

MODEL 2

Figure 4.2 : PLS Pathway for Model 2

93

MODEL 3

Figure 4.3: PLS Pathway for Model 3

4.5.3 Construct validity

Construct validity shows on how the results acquired from the use of the measure

fit the theories around which the test is developed (Sekaran&Bougie, 2010).

Construct validity is examined through two methods; convergent and discriminant

validity.

4.5.4 Convergent validity

Next, the convergent validity was tested which is the degree to which multiple items

which evaluate the same concept are agreeable. Hair et al. (2010) suggested that the

94

factor loadings, composite reliability and average variance extracted were used to

examine convergence validity. A model with main loadings above 0.5, composite

reliability of above 0.7 and Average Variance Extracted (AVE) of above 0.5 indicate

a highly valid model. Composite reliability values depict the degree to which the

construct indicators denote the variable (Hair et al., 2010). Composite reliability acts

as a proxy to Cronbach’s alpha in order to test the internal consistency of these

variables. The average variance extracted (AVE) weighs the variance held by the

indicators relative to measurement error, and it should be higher than 0.50 to endorse

using a construct (Barclay et al., 1995). These values will be displayed in the

measurement model table which is obtained after running the PLS Algorithm in

SmartPLS 3.0.

4.5.5 β value and R Square

The β value appears on the pathway between the antecedent variables to independent

variable or independent variables to dependent variable. This β value indicates the

strength of the relationship of the two variables being connected by the pathway. The

higher the β value, more weightage is given to the relationship.

PLS provided statistics indicating how well the model predicted the relationships

stated in the hypothesis. PLS also provided an indication of the predictive power of

this research model, or the squared multiple correlation (R2) value for each

endogenous variable. (Rosipal& Kramer, 2006) The R2 value found in PLS is

equivalent to the R2 value in a multiple regression model, which represents the

95

amount of variance explained by the independent variables confined within the

model (Barclay et al., 1995). It is also known as the coefficient of determination with

a higher value indicating the model to have a higher predictive power.

4.5.6 Discriminant validity

Discriminant validity signifies the degree to which one construct is dissimilar from

all other constructs in the instrument. The method for examining adequate

discriminant validity is in line with the Fornell-Larcker criterion. (Fornell&Larcker,

1981) Discriminant validity was assessed by ensuring that the diagonal elements

were significantly higher than the off-diagonal values in the corresponding rows and

columns. Items should load more strongly on their own constructs in the model, and

the average variance shared between each construct and its measures should be

greater than the variance shared between the construct and other constructs (Hair et

al., 2014). Discriminant validity is adequate when constructs have an AVE loading

greater than 0.5, which means that at least fifty percent of measurement variance was

captured by the construct (Hair et al., 2014). In the matrix, the diagonal elements in

bold are the square root of the AVE and the off- diagonal elements are the

correlations between constructs.

4.5.7 SEM Model

This model obtained from the bootstrapping function of SmartPLS provides the beta-

coefficient values, standard error values, which in turn will provide the t-values and

96

significance level. These values are highly significant for testing and supporting the

hypotheses developed. The threshold for t-values and its significance is as follows:

iii) t-value of 1.645 is significant at a significance level of 5% (α = 0.05, one-

tailed)

iv) t-value of 2.33 is significant at a significance level of 1% (α = 0.01, one-

tailed)

4.5.8 Results of each model

For each model we will present the findings for Convergent Validity, Discriminant

Validity, R square, Reliability and the Hypothesis testing results.

97

(i) Model 1

TABLE 4.5 Measurement model of VB-SEM – Model 1

MODEL

VARIABLE

LATENT

VARIABLE

INDICATOR

QUESTION

ITEMS

MAIN

LOADING

COMPOSITE

RELIABILITY AVE

IV1

RULE

OBSERVANCE

BEHAVIOUR

ROB1 0.897

0.9283 0.7220

ROB2 0.905

ROB3 0.825

ROB4 0.808

ROB5 0.809

IV2

PERCEPTION

OF GST

FAIRNESS

PTF1 0.897

0.8842 0.5708

PTF2 0.837

PTF3 0.501

PT4 0.918

PT5 0.689

PT6 0.592

DV ATTITUDE

ATT1 0.932

0.9369 0.7493

ATT2 0.925

ATT3 0.752

ATT4 0.819

ATT5 0.886

This table displays that the main loading for all question items for the variables in

model 1 is above 0.5 allowing the question items to be retained. The composite

reliability for all three variables are more than 0.7 and the AVE value is more than

0.5. These results confirm that this model has convergent validity as well as internal

consistency.

98

TABLE 4.6: Discriminant Validity – Model 1

Similar latent variables should have higher correlation whilst different latent

variables should have less correlation. It is a symmetrical table and hence will only

either have above or below value. The measurement model validated adequate

discriminate validity, since the diagonal loadings were significantly greater than the

off-diagonal loadings in the corresponding rows and columns.

Figure 4.4: PLS output for Beta-value and R square value (Model 1)

Latent

Variable ATT PTF ROB

ATT 0.866

PTF 0.668 0.756

ROB 0.510 0.462 0.850

99

The β value indicates the strength between the two latent variables, both here show

significant β values as they are above 0.2. (Hair et al., 2010) The R square is 49.9%

which is significant. 49.9% of changes in the attitude of the respondent is explained

by their rule observance behavior and their perception of tax fairness. .

Table 4.7 Summary of PLS Results – Direct Effects (Model 1)

As the t-statistic values for both Hypothesis 1 and 2 are 3.853 and 10.280

respectively, the hypotheses are significant at 1% significance level. Concurrently,

the p value for both hypotheses is 0.000.

Hypothesis Path β value Standard Error t Statistics p Value Decision

1 ROB -> ATT 0.256 0.067 3.853 0.000 Supported

2 PTF -> ATT 0.550 0.054 10.280 0.000 Supported

100

Figure 4.5: PLS output for testing population regression coefficients (beta values) –

Model 1

101

(ii) Model 2


MODEL

VARIABLE LATENT VARIABLE

INDICATOR

QUESTION

ITEMS

MAIN

LOADING

COMPOSITE

RELIABILITY AVE

IV1 SELF-EFFICACY

SE1 0.867

0.941 0.699 SE2 0.824

SE3 0.854

SE4 0.799

IV2 LEVEL OF TAX

KNOWLEDGE NOT APPLICABLE 1.000 1.000

DV

PERCEIVED

BEHAVIOURAL

CONTROL

PBC1 0.950

0.903 0.842 PBC2 0.870

PBC3 0.931

This table displays that the main loading for all question items for the variables in


reliability for all three variables are more than 0.7 and the AVE value is more than

0.5. These results confirm that this model has convergent validity as well as internal

consistency. For Level of Tax knowledge, the values are not applicable as it is

measured by one absolute value and not a series of question items.

TABLE 4.9: Discriminant Validity- Model 2

Latent

Variable LTK PBC SE

LTK 1.000

PBC 0.151 0.917

SE 0.043 0.514 0.836

102

The measurement model confirmed adequate discriminate validity, since the

diagonal loadings were significantly greater than the off-diagonal loadings in the

corresponding rows and columns.


The β value indicates the strength between the two latent variables. The value

between self-efficacy and perceived behavioral control is more than 0.2 indicating

significant strength whilst tax knowledge and perceived behavioral control is less

than 0.2 indicating less significant relationship (Hair et al., 2010)

The R square is 28.1% which is indicates that 28.1% of changes in the perceived

behavioural control of the respondent is explained by their self-efficacy and level of

tax knowledge.

103


Hypothesis 3 has a t-statistic value of 2.272 which is more than 1.65 but less than

2.33 making the hypothesis supported at the 5% significance level. Hypothesis 4 has

a t-statistic value of 10.117 which is way above 2.33 confirming that the hypothesis

is supported at the 1% significance level .

Figure 4.7: PLS Output for testing population regression coefficients (Model 2)

Hypothesis Path β value Standard Error t Statistics p Value Decision

3 LTK -> PBC 0.129 0.057 2.272 0.024 Supported

4 SE -> PBC 0.509 0.050 10.117 0.000 Supported

104

(iii)Model 3

Table 4.11 displays that the main loading for all question items for the variables in


reliability for all eight variables; antecedent, independent and dependent are more

than 0.7 and the AVE value is more than 0.5. These results confirm that this model

has convergent validity as well as internal consistency. For Level of Tax knowledge,

the values are not applicable as it is measured by one absolute value and not a series

of question items.

One question item was dropped, which was SN 5 for Subjective norms as it had a

main loading of 0.482 which was less than the cut-off value of 0.5.

105

MODEL

VARIABLE LATENT VARIABLE

INDICATOR

QUESTION ITEMS MAIN LOADING

COMPOSITE

RELIABILITY AVE

IV1 ATTITUDE

ATT1 0.931

0.937 0.750

ATT2 0.923

ATT3 0.748

ATT4 0.825

ATT5 0.889

IV2 SUBJECTIVE NORMS

SN1 0.850

0.856 0.602 SN2 0.676

SN3 0.669

SN4 0.884

IV3 PERCEIVED BEHAVIOURAL

CONTROL

PBC1 0.950

0.941 0.842 PBC2 0.873

PBC3 0.929

DV LEVEL OF ACCEPTANCE OF GST LA1 0.961

0.954 0.911 LA2 0.948

IV1 SELF-EFFICACY

SE1 0.867

0.941 0.699 SE2 0.824

SE3 0.854

SE4 0.799


106

Table 4.11 Measurement Model of VB-SEM – Model 3 (continued)

AV2

PERCEPTION OF TAX FAIRNESS

PTF1 0.897

0.8842 0.5708

PTF2 0.837

PTF3 0.501

PT4 0.918

PT5 0.689

PT6 0.592

AV3 SELF-EFFICACY

SE1 0.867 0.941 0.699

SE2 0.824

SE3 0.854

SE4 0.799

AV4 LEVEL OF TAX KNOWLEDGE NOT APPLICABLE 1.000 1.000

AV1 RULE OBSERVANCE BEHAVIOUR

ROB1 0.897

0.9283 0.7220

ROB2 0.905

ROB3 0.825

ROB4 0.808

ROB5 0.809

107

Table 4.12 Discriminant Validity – Model 3

The measurement model confirmed adequate discriminate validity, since the

diagonal loadings were significantly greater than the off-diagonal loadings in the

corresponding rows and columns

LATENT VARIABLE ATT LA LTK PBC PTF ROB SE SN

ATT 0.866

LA 0.833 0.955

LTK 0.311 0.372 1.000

PBC 0.793 0.772 0.150 0.918

PTF 0.666 0.666 0.216 0.607 0.756

ROB 0.509 0.533 0.192 0.452 0.461 0.850

SE 0.486 0.390 0.043 0.513 0.319 0.249 0.836

SN 0.698 0.683 0.127 0.757 0.619 0.478 0.304 0.776

108


The Beta value indicates the strength between the two latent variables that are

connected through a path. The beta value here for all pathways are significant as a

common indicator of significance is a value of 0.2 and above. (Hair et al., 2010)

The R square is 73.2% which is very significant as it implies that 73.2% of changes

in the respondent’s level of acceptance of GST is explained by their attitude,

subjective norms and perceived behavioural control with the antecedent variables of

perception of tax fairness, rule observance behaviour, self-efficacy and level of tax

knowledge present. This is most probably the case because the Theory of Planned

Behaviour which was the theory supporting the framework of this model is a well-

LA

109

established and concrete theory applicable in any context, even in such an

unexplored context of Malaysian GST acceptance.


Hypotheses 5 and 7 both have t-statistic values which are way above 2.33 which

indicate that both the hypotheses are supported at 1% significane level. The p value

is in fact 0.000 and 0.001 respectively, concurring the significance level stated.

However, Hypotheses 6 is not supported as the t-statistic value is below 1.65 which

is the minimum threshold.

Figure 4.9: PLS Output for testing population regression coefficients – Model 3

Hypothesis Path Beta value Standard Error t Statistics p Value Decision

5 ATT -> LA 0.570 0.061 9.379 0.000 Supported

6 SN -> LA 0.100 0.062 1.625 0.105 Not Supported

7 PBC -> LA 0.243 0.076 3.207 0.001 Supported

LA

110

4.5.9 Reliability Analysis

Cronbach’s Alpha has been widely used to measure the internal consistency of

variables in a model. The following tables have summarized the Cronbach Alpha

values for each variable. All the values are above the threshold value of 0.7

(Nunnally and Berstein, 1994). This confirms that all measurements of the variables

are reliable and have significant internal consistency.

Table 4.14: Cronbach’s Alpha and Composite Reliability for Model 1

Latent Variable Cronbach’s Alpha Number of indicator

items

Attitude 0.915 5

Perception of Tax Fairness 0.839 6

Rule Observance

Behaviour 0.903 5


Latent Variable Cronbach’s Alpha Number of indicator

items

Level of Tax Knowledge Not applicable 20

Perceived Behavioral

Control 0.907 3

Self-Efficacy 0.858 4

111


Latent Variable Cronbach’s Alpha Number of indicator items

Attitude 0.915 5

Rule Observance

Behaviour 0.903 5

Perception of Tax Fairness 0.839 6

Perceived Behavioural

Control 0.907 3

Level of Tax Knoweledge Not applicable 20

Self-Efficacy 0.858 4

Subjective Norms 0.781

5

Level of Acceptance 0.903 2

4.5.10 Summary on Validity and Reliability of Models

This study has met its measurement model requirements for reflective construct as

the results show:

1) The internal consistency reliabilities indicated by the composite reliability were

all at least 0.8 and exceeding minimal reliability criteria.

2) Strong evidence of convergent and discriminant validity was found as

112

a) the square root of the average variance extracted for each construct was greater

than 0.70 (i.e., AVE > 0.50) and greater than the correlation between that construct

and other constructs (without exception)

b) the factor structure matrix shows that all items exhibited high loadings (>0.50) on

their perspective constructs and no items loaded higher on constructs they were not

intended to measure.

Therefore, it was determined that this instrument had achieved acceptable levels of

validity and reliability. Overall, the measurement instruments exhibited sufficiently

properties to support valid testing of the proposed structural model.

Following that, the structural model was assessed and the hypotheses were tested

based on the t-values.

113

4.5.11 Summary of Hypothesis Testing

4.17 Summary of Hypothesis Testing

Hypothesis Statement Results

1 Rule Observance Behavior will positively

affect attitude to accept GST

Supported

2 Perception of GST fairness will positively

influence attitude to accept GST

Supported

3 Level of GST knowledge positively

influences perceived behavioral control

Supported

4 Self-efficacy will positively influence

perceived behavioral control

Supported

5 Attitudes towards GST are positively

related to level of acceptance of GST.

Supported

6 Subjective norms are positively related to

level of acceptance of GST

Not Supported

7 Perceived behavioral control is positively

related to the level of acceptance of GST

Supported

114

4.5.12 Goodness-of-Fit

The Goodness of Fit (GoF) for the present research is determined as below:

GoF = √ ( AVE x R2

)

This measurement will be done for all three models. The AVE values can be

obtained from the Measurement model table. The R2 value can be obtained from the

PLS output image.

Model 1 – Antecedent Variables to Attitude

Average AVE = (0.7220+0.5708+0.7493) / 3 = 0.6807

R2 = 0.499

GoF = √ ( 0.6807 x 0.499)

= 0.5828

Model 2 – Antecedent Variables to Perceived Behavorial Control

Average AVE = (0.699+0.842+1.000)/ 3 = 0.847

R2 = 0.281

GoF = √ (0.847 x 0.281)

= 0.4879

115

Model 3 – Full Framework

Average AVE = (0.7220+0.5708+0.699+0.842+1.000+0.750+0.602+0.911)/8 =

0.72614

Average R2 = (0.281+0.499+0.732)/3 = 0.504

GoF = √ (0.7261 x 0.504)

= 0.6049

The calculated values for Model 1, 2 and 3 are 0.5828, 0.4879 and 0.6049

respectively. All these values exceed the threshold value of 0.36 indicating that

goodness of fit is sufficient. All three models of this research are highly significant

as the constructs of the model fit in well.

4.6 Summary

In this chapter, necessary statistical tests were conducted and as a result the models

were proven to be valid and reliable with a significant explanatory power.

Hypotheses 1 to 7 were all accepted except for Hypotheses 6 concerning the

relationship between subjective norms and behavioral intention.

116

CHAPTER 5

DISCUSSIONS AND CONCLUSIONS

5.0 Introduction

This chapter will summarize the study and will discuss the findings of the study.

Implications of the study will also be discussed followed by discussions on the

limitations of the study and suggestions for future research. Finally the chapter will

end with a conclusion of the study.

5.1 Recapitulation of the Study Findings

The purpose of this study was to answer the following research questions:

1) What is the level of acceptance of GST amongst consumers?

2) What is the relationship between rule observance behavior and attitude of

consumer?

3) What is the relationship between perception of GST fairness and attitude of

consumer?

4) What is the relationship between the level of GST knowledge of consumers

and perceived behavioral control?

5) What is the relationship between self-efficacy of a consumer and perceived

behavioral control?

6) What is the relationship between attitude of consumers and their level of

acceptance of GST?

117

7) What is the relationship between subjective norms of consumers and their

level of acceptance of GST?

8) What is the relationship between perceived behavioral control of consumers

and their level of acceptance of GST?

This research is conducted in conjunction with the latest implementation of Goods&

Services Tax in Malaysia on 1st April 2015. Goods & Service Tax was a move made

by the government in order to regularize and systemize the revenue generation of the

government. From the government’s viewpoint, GST is the most logical transition

for the tax collection system in Malaysia especially since all neighboring countries

have already implemented GST.

However, the acceptance of Malaysian consumers with the concept of GST was

projected as low due to the numerous protests and the preliminary survey conducted

by the Merdeka Centre. Whilst some consumers are agreeable with the

implementation of GST, some are not agreeable. The government is in the

impression that the lack of acceptance is due to the lack of knowledge on the

mechanisms of GST and they have taken all measures to educate the general public

on GST (Borneo Post Online, 2015; Jalil,2015; GST Malaysia Info,2014)

This research is hence aimed at first gauging the level of acceptance amongst

consumers and even more interestingly, this study has focused its respondents to

MBA students. This augments the quality of the research because we are studying a

group of people who are assumed to have a sound knowledge in GST as it would be

highly related to their career and intellectual level.

118

The research seeks to identify the crucial factors that influence the acceptance of

GST amongst these specified consumers. What are the issues or elements that will

result in a consumer agreeing with the concept of GST or otherwise? In search of the

answers to this vital question, the Theory of Planned Behavior came in view as a

theory which can be used to conduct a study on the subject matter. Being a well-

established theory, used to support many research frameworks in numerous contexts,

the Theory of Planned Behavior formed the base to the theoretical framework of this

research. In addition, the components of the TPB were further examined and

antecedent variables to the Attitude and Perceived Behavioral Control factors were

included. These antecedent variables were crucial to understand the practical

implications of the Theory of Planned Behaviour. The antecedent variables lead to

the independent variables and these antecedent variables are concepts that can be

easily understood by consumers.

For instance, just saying attitude of consumers is theoretically viable however the

practical implication is limited. However, when we include antecedent variables and

prove that these antecedent variables lead to a positive attitude towards GST

implementation, the government and related authorities will be able to practically

address the issue. In this context, Rule observance Behavior and Perception of Tax

Fairness are antecedent variables to Attitude. From the findings of the research, the

related authorities will be able to engage in practical solutions to address consumers’

perception of tax fairness and rule observance behavior in order to improve their

119

attitude towards GST instead of not having a clear guideline on how to address the

issue.

In order to answer the research questions of the study, the framework hence consists

of four antecedent variables, three independent variables and one dependent variable.

The antecedent variables in this study are rule observance behaviour, perception of

tax fairness, level of tax knowledge, self-efficacy, the independent variables are

attitude, subjective norm and perceived behavioural control and the dependent

variable is the level of acceptance of GST. It is vital to note here that the acceptance

of GST refers to the level of agreement to the implementation of GST and not the act

of paying GST. These variables were then measured through an online questionnaire

with various sections designed to ensure a systematic data collection for each

variable. 228 MBA students took place from USM, UKM, UM and UPM with the

proportion of respondents being 86, 48, 43 and 51 respectively. The data collected

from these respondents were analyzed using the SmartPLS version 3 and IBM SPSS

Statistics to study the relationships between the various variables.

5.2 Discussion

In this section, the relationships between the antecedent variables, independent

variable and dependent variable and their respective implications are discussed based

on the research questions.

120

5.2.1 What is the GST acceptance level amongst Malaysians?

The acceptance level amongst Malaysian consumers who are MBA students was

determined through this study. The acceptance level was determined through the

intention of the consumer to accept GST as during the time of data collection GST

had not been implemented. The time period of data collection was in February 2015.

Hence, in this situation the behavioral intention component of the Theory of Planned

Behavior was the acceptance level of these Malaysian consumers. For level of

acceptance, the average score was 2.94 confirming slightly low level of acceptance

of GST amongst these consumers who were MBA students. This is because it is

below 3 which is the neutral point of the scale. Any mean value above 3 indicates

agreement and below 3 indicates disagreement (Boone & Boone, 2012).

In order to study the impact of the various demographic characteristics on the

respondents’ level of acceptance of GST, independent t-tests as well as one-way

Analysis of Variance (ANOVA) were conducted.

An independent t-test was conducted to compare whether two groups have different

averages in terms of their level of acceptance of GST. A one-way ANOVA is used

also to compare means but for three or more groups using the F-distribution. IBM

SPSS statistics software version 22 is once again employed for this analysis. For the

purpose of the t-test and one way ANOVA, each group should have at least 30

samples for an appropriate analysis.

121

Hence, t-tests were conducted on Gender and Programme Mode whilst one-way

ANOVA was conducted for Institution and Income. The findings of the tests are

presented in Tables 5.1 to 5.5 and the outputs of the analysis are attached in the

appendix.

Table 5.1 Summary on t-test for Gender and Mode of Study

Dimensions Groups N Mean Std Dev t Sig(2-

tailed)

Gender Male 105 3.0762 1.09145 1.736 0.084

Female 122 2.8279 1.05973

Mode of

Study

Full time 32 2.7031 1.08404 -1.373 0.171

Part time 196 2.9847 1.07406

As the significance level for both dimensions is more than 0.05, the differences in

mean between the groups are not significant. It can be concluded that the intention to

accept does not vary irrespective of gender and mode of study. Full time students

who are not working have the same level of intentions to accept as part time

students.

Table 5.2 Descriptive Table on Institutes and Level of Acceptance of GST

Dimensions Groups N Mean Std Dev

Institute USM 86 2.6977 0.99493

UKM 48 2.8542 0.96733

UM 43 2.8256 1.09590

UPM 51 3.5490 1.09661

Total 228 2.9452 1.07753

122

Table 5.3 ANOVA table of results for institution groups

Sum of

Squares df Mean Square F Sig.

Between Groups 24.877 3 8.292 7.782 .000*

Within Groups 238.688 224 1.066

Total 263.565 227

Significant at 1%

From table 5.3, it can be seen that there is a significant difference in mean of GST

acceptance amongst students in the various institutes; USM, UKM, UM and UPM.

From Table 5.2 it can be seen that UPM students have a much higher acceptance

level of the implementation of GST.

Table 5.4 Descriptive Table on Income Level and Level of Acceptance GST

Dimensions Groups N Mean Std Dev

Income Level No Income 34 2.7353 1.06767

1000 – 5000 126 2.9405 1.16894

5001 – 10000 35 3.1143 .68691

10001 – 15000 33 3.0000 1.06800

Total 228 2.9452 1.07753

Table 5.5 ANOVA table of results for income level groups

ANOVA

Sum of

Squares df Mean Square F Sig.

Between

Groups 2.601 3 .867 .744 .527

Within Groups 260.964 224 1.165

Total 263.565 227

123

From Table 5.5, it can be seen that there is no significant difference between the

different income level groups. This means an unemployed student and a student with

a monthly income have the same level of intention to accept GST. Income level does

not seem to influence their intention to accept GST.

5.2.2 What is the relationship of rule observance behavior and perceptions of

tax fairness with attitude of consumer?

The relationship of Rule Observance Behaviour and Perception of Tax Fairness with

Attitude of consumer is discussed under one section as these three latent variables

were examined in one model, Model 1 in chapter 4.

Attitude is the positive or negative feeling a consumer has regarding GST which is a

result of their belief system (Fishbein&Ajzen, 1975). It can be seen from the

descriptive statistics that attitude is 3.14 with a standard deviation lower than 1. This

points out that the respondents are in agreement as standard deviation is low and that

they have a slightly positive feeling towards the implementation of GST. This is

because on a Likert Scale, 3 marks neutrality and any value below inclines towards

low level of acceptance of GST and any value above inclines low level of acceptance

of GST. With effective measures taken, the attitude of consumers towards

acceptance of GST can be further enhanced.

124

Rule observance behavior refers to an individual’s tendency to conform to rules

(Trevino, 1986). In this study, rule observance behavior is treated as an antecedent

variable to attitude. The rule observance behavior mean was 3.66 with a low

standard deviation indicating that the respondents in general, were those who obeyed

rules and believe that rules bring about improvements. The results of this study

concluded that rule observance behavior positively affects attitude and that is why

Hypothesis 1 was supported. This is in line with previous research conducted that

tests the relationship between rule observance behavior and attitude (Sarina et al.,

2007; Jeffrey &Weatherholt, 1996).

In order to further enhance the rule observance behavior amongst Malaysians, the

authorities can opt to increase their understanding of certain rules and the purpose of

those rules. If consumers are able to understand the positive effects of the

implementation of certain rules, their rule observance behavior will increase

Perception of tax fairness refers to the opinion that the consumers have on the

fairness level of the taxation system (Saad, 2010). This was also taken as an

antecedent variable to attitude. The average response on the perception of tax

fairness was at 2.86 indicating that the respondents do not perceive the tax system as

fair. The low standard deviation confirms that all respondents are in agreement with

the low perception of tax fairness. This is a very important point to note as a low

perception of tax fairness can only be enhanced with the direct involvement of

government authorities. The consumers need to be transparently informed on the

allocation of the taxes they have paid and the returns they are obtaining from the tax

they have paid. The positive relationship between perception of tax fairness and

125

attitude, reflected in Hypothesis 2, was confirmed as significant. Hence, this study

inferences that the more positive the perception of tax fairness, the more positive the

attitude towards the acceptance of GST. Findings from previous studies have

concluded with the same positive relationship (Thomas, 2012; Seidl & Traub, 2001;

Saad, 2010; Tan & Chin-Fatt, 2000)

Rule observance behavior and perception of tax fairness are able to explain 49.9% of

variance in a consumer’s attitude as the R square for Model 1 is 49.9%. This is a

significant value and confirms that these two factors are appropriate antecedent

variables to attitude. The beta weight for perception of tax fairness is higher

(β=0.550) than the beta weight of Rule Observance Behaviour (β=0.256). This

explains that the perception of tax fairness is a critical factor to be considered in

influencing consumers’ attitude.

This explains the phenomenon that although the rule observance behavior of

respondents is high, the positive attitude is not equally high because of the low

perception of tax fairness. Consumers are not going to accept GST as they value the

fairness of the taxation system more than obeying the rules itself.

5.2.3 What is the relationship of Self – Efficacy and Level of Tax Knowledge

with Perceived Behaviour Control?

126

The relationship of Self – Efficacy and Level of Tax Knowledge with Perceived

Behaviour Control of consumers is discussed under one section as these three latent

variables were examined in one model, Model 2 in chapter 4.

Perceived behavioural control is the consumer’s discernment on how easy or

difficult it is to perform a specific act. (Ajzen& Madden, 1986) In this case, the

respondents’ responses to the questionnaire have inferred that they have a slightly

low perceived behavioural control at an average of 2.91. However, the standard

deviation of slightly higher than 1 shows that the level of agreement on the low

perceived behavioural control is not evenly spread across all respondents. This

slightly low value of perceived behavioural control indicates that these specified

consumers foresee internal and external constraints to them agreeing with the

implementation of GST.

Self-Efficacy plays a key role in enhancing the perceived behavioural control of an

individual as proven by Raghuram(2003) and Beauregard(2012). In this research,

self-efficacy refers to the judgments of respondents of their own capability in

managing their finances, consumption and expenditure as well as adapting to GST.

The mean self-efficacy amongst the respondents was 3.24 with a low standard

deviation giving an inference that the respondents in general slightly agreed that they

possess the self-efficacy required for the adaptation to the GST. This slight lack of

self-efficacy amongst the consumers should be a concern of the government

authorities as it will cause higher levels of debts and financial issues if consumers are

not assisted effectively.

127

The findings of this research have established that there is a positive relationship

between self-efficacy and perceived behavioral control. Hypothesis 3 has been

supported by the findings from the data analysis. The positive relationship between

self-efficacy and perceived behavioral control was also proven in previous studies,

(Raghuram,2003; Schmidt & Karsten, 2015; Kulviwat (2014). The self-efficacy of

respondents has to be enhanced in order to increase their perceived behavioural

control.

Level of tax knowledge for the purpose of this study was evaluated through two

sections, technical questions on the mechanisms of GST and scenario-based practical

questions. These two sections will be able to gauge the respondents’ level of

understanding in the subject matter of GST. The average score of respondents is

76.34% with a standard deviation of 14%. This is quite a positive outcome as it

states that, the respondents in general without any significant variances, have a sound

and good level of knowledge on GST. This has also proven that all the efforts taken

by the government to educate the public through seminars has had positive effects on

this group of respondents.

The relationship between level of tax knowledge and perceived behavioural control

has proven to be positive as hypothesis 4 has been accepted. Hence, an increase in

the level of tax knowledge will inevitably increase the consumer’s perceived

behavioural control. This is in line with previous research by Saad (2010) and Chiou

(1998).

128

With the abovementioned antecedent variables, the relevant authorities will have a

clearer idea on how to address this issue of lack of perceived behavioural control

amongst the consumers.

Self-efficacy and level of tax knowledge are able to explain 28.6% of variance in a

consumer’s perceived behavioural control as the R square for Model 2 is 28.6%.

Although lower than Model 1’s R square, it is still a noteworthy value and indicates

that these two factors are antecedent variables to perceived behavioural control.

(Filho et al, 2011) The beta weight for self-efficacy is significantly higher (β=0.506)

than the beta weight of Level of tax knowledge (β=0.148) concluding that self-

efficacy has higher influential power on the consumers’ perceived behavioral

control. This explains the phenomenon that although the level of tax knowledge of

respondents is high, the perceived behavioral control is not equally high because of

the lower level of self-efficacy amongst the respondents.

5.2.4 What is the relationship of Attitude, Subjective Norms and Perceived

Behavioral Control with Level of Acceptance?

The average scores for attitude, subjective norms and perceived behavioral control

were 3.14, 3.00 and 2.91 respectively. As the mean value for attitude is slightly

above 3 it indicates slight agreement. Subjective norm has a mean value of 3

indicating absolute neutrality. Perceived behavorial control is slightly low at 2.91.

The average score for level of acceptance of GST was 2.94 indicating slightly low

level of acceptance.

129

From the coefficient of determination value, it can be seen that attitude, subjective

norms and perceived behavioural control together explain 73.2% of the variation in

behavorial intention of the respondents. This is a highly significant value endorsing

that the Theory of Planned Behaviour is a fundamentally concrete theory which can

be used in any context.

From the findings of the data analysis, the relationship between attitude and

perceived behavioural control with behavioral intention has been proven to be

significant. The beta weight for attitude is significantly higher (β=0.570) than the

beta weight of perceived behavorial control (β=0.243) illustrating that attitude is the

better predictor of intention to accept GST. This critical information now

emphasizes the importance of cultivation of attitude in support of Goods and

Services Tax in order to increase the acceptance level amongst the consumers.

Hence, looking into the antecedent variables of attitude, the government should

allocate resources and probably undertake drastic measures to increase the

perception of tax fairness amongst the consumers. This in turn will cultivate a

positive attitude towards the concept of GST and ultimately result in a higher

acceptance level. Previous studies, Hanno and Violette (1996), Wan et al. (2012) and

Loo et al (2007) concluded in the same way.

The influence of perceived behavioral control is weaker but still important. Inclusion

of perceived behavioural control in the Theory of Reasoned Action has led to the

formation of Theory of Planned behavior, enhancing its explanatory power

extensively. (Taylor&Todd, 1995) As mentioned by Sparks et al. (1997), several

researchers have customized the measurements of perceived behavioural control to

130

the behavioral issue in concern in the particular study. In this case, the measurements

of the perceived behavorial control was tailor-made to suit the GST scenario and

antecedent variables were identified to enable easier understanding on the

operationalization of perceived behavioral control. Previous studies have found

similar findings (Bobek and Hatfield, 2003; Damayanti,2012).

In contrast, Hypothesis 6 which was testing the relationship between subjective

norms and behavioural intention was not supported. Subjective norm is the

individual’s perception that those who are important to him/her think he/she should

not perform the behavior in question. (Fishbein&Ajzen,1975) Hence, it is concluded

that subjective norms had no significant effect on the intention of respondents to

accept GST. This is in contrast with previous studies, (Damayanti, 2012; Bobek,

1997;Nordiana, 2012) but at the same time coherent with other studies which found

no relationship between subjective norms and intention to behave. (Mathieson, 1991;

Chau & Hu, 2001).This could be explained by the fact that the respondents were all

MBA students with a high intellectual level. Individuals with a high intellectual level

have a lower tendency of being influenced by peers and family on issues and

decision making as they are able make sound decisions based on their perceptions

and beliefs.

5.3 Summary of findings

An empirical research was conducted to determine the factors which influence the

acceptance level of consumers with Goods and Services Tax in Malaysia. The

Theory of Planned Behavior was employed to examine the significance of attitude,

131

subjective norms and perceived behavior control as factors influencing acceptance of

GST. However, only attitude and perceived behavorial control had a significant

effect on the behavioral intention of consumers to accept GST.

In addition to antecedent variables to attitude and perceived behavorial control were

also identified to enable easier practical application of the research. The antecedent

variables to attitude were rule observance behavior and perception of tax fairness

with the latter having a more significant influence compared to the former. With the

perceived behavior control of the consumers, its antecedent, self-efficacy had a

higher explanatory power than the other antecedent variable; level of tax knowledge.

These specific and clear findings will act as stepping stone for the relevant

authorities to understand the effective measures that can be undertaken to improve

the acceptance level amongst consumers.

5.4 Implications

The implication of this study is an important discussion as it will make sense of the

purpose of conducting this particular research. This section will also discuss the

impact and contribution of the outcomes of this research. Thus, the implications of

this research can be viewed from two major points; theoretical implication and

practical implication.

5.4.1 Theoretical Implication

132

This study has used Theory of Planned Behavior to examine the perceptions of

consumers with regards to GST. This study attempted to study the factors

influencing the consumers’ acceptance of Goods and Services Tax in Malaysia. Most

studies pertaining GST in Malaysia has been related to tax collectors; companies and

their compliance level. (Bidin & Shamsudin, 2013; Nordiana, 2012; Palil & Ibrahim,

2011) Other studies have examined the awareness and acceptance level towards

GST. This study will however be the first to examine the factors affecting the

acceptance level using an established theory such as TPB. This research attempts to

enhance the academic awareness in the field of Goods & Services Tax in Malaysia.

An honest effort has been made to apply a well-established theory to a local and very

current context to further strengthen the validity of the theory. In addition, this study

has also examined the antecedents to Attitude and Perceived Behavioral Control.

With these antecedents, the explanatory power and applicability of the research

enhances multiple times.

Using Theory of Planned Behavior, the model was able to explain 73.2% of the

variance in the intention of consumers to accept GST. This significant statistical

indicator confirms that this model is also appropriate in predicting the Malaysian

consumers’ acceptance intention of the newly implemented Goods & Services Tax.

Previous study by Nordiana (2012) which used Theory of Planned Behavior to

examine the compliance behavior amongst corporate taxpayers reported a coefficient

of determination of 22%. This study on consumers has a higher and more significant

coefficient of determination which indicates the significance of the independent

variables on the dependent variable.

133

5.4.2 Practical Implication

Goods and Services Tax is a consumption tax and hence will ultimately have a

significant impact on consumers. Thus, a study which is consumer-centric will

provide significant practical implications which will be able to solve the issues faced

at this point of time. The outcome of the research has confirmed a slightly low level

of acceptance of GST despite the efforts taken by the government. This definitely

indicates that a new approach has to be undertaken to address this issue and this

research will be able to provide an idea on the current mindset of the consumers.

As Rule Observance behavior has a significant impact on consumers’ attitude toward

Goods and Services Tax, effort should be taken to further enhance this behavior

amongst consumers. Currently, the government is emphasizing on the eradication of

corruption which will lead to an improvement in the compliance of rules by citizens

and authorities. This is because enforcement will be enhanced by the authorities.

However, citizens can always find a way around rules if this rule observance

behavior is not embedded in them. The relevant authorities have to take more

effective steps to ensure rule compliance. According to Burgess (1996), following

rules is a mindset and in order to influence it, steps have to be taken in the early

years of an individual’s life. Hence, school children must be exposed to the various

rules in place and most importantly be aware of the reasoning and justifications of

those rules. They must be taught to appreciate these rules which will lead to higher

rule observance behaviors. For the immediate scenario, the rules' reasoning and the

134

justification for abiding by those rules should be focused on in the strongest possible

way. Burgess (1996) has suggested that we should emphasize on the negative

consequences for other people if rules are not obeyed. The government can even go

further to introduce a merit system to motivate citizens to obey rules. A simple

example that can be considered would be that the local authorities can commit that

for every vehicle which does not receive a single summon in three years is entitled to

free renewal of road tax on the 3rd

year. Once there is a strong enough reason, rule

observance behaviour of citizens will be enhanced. (Burgess, 1996)

This research has indicated perception of tax fairness is low amongst consumers and

it is the factor that significantly influences attitude of consumers. Once consumers

view themselves as victims of fiscal inequity, they lose trust in the fairness of the

taxation system. (Siahaan, 2012) Thus, the government has to undertake effective

measures that enable the consumers to clearly view the uses of the taxes collected as

well as to be able to obtain benefits from the country in return of the tax paid. From

the open-ended questions, it can be gathered that the distribution of ‘Bantuan Rakyat

1 Malaysia’ (BRIM) was not viewed as a fair distribution of the tax collected as the

respondents felt that the tax collected should be used to enhance medical, education

and welfare services for all. The concept of BRIM was also not applauded as it is a

very short term solution for a change which is going to be permanent;

implementation of GST. Many of the respondents suggested that further reductions

in income tax rates would be better as currently the high-income and middle-income

earners are highly affected as they have to pay both income and consumption tax.

The low-income earners will be benefiting as the GST is multiple-rate system and

135

hence a lot of daily grocery items highly used by the low income earners has been

stated as zero-rated. Other suggestion in contrast to BRIM also included increasing

job opportunities, subsidies, education assistance, business start-up assistance and

technical skills providers. The government has to focus on transparent distributive

justice-the fair allocation of resources to ensure the consumers are able to reap basic

benefits for the tax that they are paying. (Boonyarat et al, 2014) The transparency of

the collection of income tax and its allocations had led the Jordanian taxation system

to be viewed as fair by its citizens. (Khasawneh, 2008)

The government has been harping on educating the public on the mechanisms of

GST and as a result of it the knowledge level was concluded as high in this research.

Now that they have educated the current consumers, the education ministry should

be proactive and move towards educating school children and undergraduates on

GST. This is so that the concept of GST is already in-built and they are able to

understand and acknowledge it as a necessity for the government. These individuals

are going to be the next generation of consumers and embedding this knowledge in

them will better prepare them to deal with GST. With all these knowledge being

imparted to these future consumers, they will not perceive anymore constraints in

accepting GST.

For perceived behavioral control, it was learnt that a higher weightage should be

placed on the self-efficacy levels of the consumers. Hence, although their knowledge

of GST and its implications were high, consumers were not confident that they will

be able to manage the situation especially their finances once GST has been

implemented. In this case, the government should look into improving the self-

136

efficacy levels of consumers. This can only be improved if the consumers have the

self-confidence that they possess the required skills and abilities to manage their

finances and expenditure with the implementation of GST. It is a perception that they

make of themselves and can only be enhanced by changes in their own personal

perceptions. Thus, besides being given knowledge on the mechanisms and purpose

of GST, the government can opt to utilize the given budget of RM 250 million on

educating consumers in managing their consumption and finances. A very practical

example would be that, when a consumer understands the basic rule that only a

company earning above RM500,000 will have to comply with the collection to GST,

they can always opt to shop in smaller shops to avoid GST. Consumers who

understand that the GST is merely replacing the sales and services tax will

understand that for certain services they are paying the same amount of tax and for

certain items they are indeed paying less than before when 10% tax was imposed.

The savings here is then transferred over to other items for which now they have to

pay a GST of 6%. . They should also be informed about the differences between

sales and service tax and GST.

Attitude towards GST is slightly positive and working towards improving this would

inevitably result in better acceptance of GST. To enhance this, the government has to

undertake the challenging task of creating a positive feeling towards GST amongst

consumers. The first and foremost action that must be undertaken by the government

is to improve their support system for the consumers. This transition time is the most

critical time and the government must employ as much manpower as possible to

answer consumer’s queries and complaints. Effort and care must be taken to ensure

the hotlines are available at all times and immediate, transparent actions are taken

137

against complaints. This sincerity in supporting consumers through this transition

will definitely result in the consumers feeling more positive about GST and the

authorities. To add on, it must be communicated to the consumers that all

neighboring countries have already implemented GST successfully and this will

prompt the consumers to feel more positively towards GST and its contributions to

the nation.

Perceived behavorial control is slightly low amongst the consumers indicating that

these consumers perceive internal and external constraints in them accepting GST.

Currently, the government has taken the well-applauded move to look into the

service charge that a number of outlets were charging. The government now has

made it mandatory for these outlets to clearly display their collective agreement with

its employees and the service they will be rendering to their customers in order for

the outlet to include the service charge. (Idris, 2015) The same modus operandi can

be implemented in other areas to make this new implementation easy for the

consumers. Another way would be for the government to enforce display of

information on whether the particular item is standard rated or zero rated. In the

current situation, consumers at many outlets are only aware of the GST implication

on the product when the payment is made and receipt is generated. If the authorities

make it mandatory for the display of the GST rate of the product, it will make it

easier on the consumer to be able to choose their products and manage their

consumption. Furthermore, it should also be made mandatory for outlets to clearly

display their GST compliance status; whether or not it is GST registered so that

consumers can make an informed decision to patronize their outlet. With these

138

measures in place, the constraints for consumers are reduced and they would find it

easier to accept GST. .

5.5 Limitations

There are a few noteworthy limitations to this study. Firstly, the unit of analysis of

this study has been confined to consumers who are currently pursuing their MBA in

the four selected universities. This may affect the generalizability of the study as

well as representation of all consumers’ intention to accept GST.

The second limitation was that convenience sampling was used due to lack of

support from the various graduate business school administrators. Convenience

sampling also may affect the generalizability of the findings.

Another limitation is that the timing of the research coincides with the transition of

GST implementation. Data was collected pre-GST but the findings were developed

post-GST, this might have an effect on the perceptions of the knowledgeable

consumers.

5.6 Suggestions for Future Research

First and foremost, the limitations discussed above should be addressed in future

research. A more wide spread population of respondents using random sampling

methods will provide highly representative findings.

139

Future research can opt to study the acceptance level amongst different categories

consumers. Amongst the categories they can look at are consumers with an

undergraduate degree or below and also consumers who are not educated. The latter

would be an interesting study as the level of tax knowledge amongst uneducated

consumers can also be examined.

Furthermore, the differences in mean within the four universities can be further

explored to understand the cause of the difference, be it geographical distribution,

academic exposure, work exposure, state government efforts or other reasons.

To add on, a similar study should be conducted after the actual implementation of

Goods & Services tax to study the responses of consumers after having experienced

GST for them. This research obtained responses from consumers on a perception-

basis whereas a follow-up research would produce more relevant responses.

Longitudinal studies can be conducted to study the differences in attitude, subjective

norms and perceived behavioral control across initial acceptance and continuance

stages.

For future research, it could be suggested that the perception of tax fairness be

evaluated using the Giligan & Richardson’s (2005) 21 item tax fairness perception

scale covering all fairness dimensions. This will enable the researcher to analyze in

detail the tax fairness perceptions of respondents and specific areas of improvement.

Finally, the research model in this study has managed to explain 73.2% of variation

in behavorial intention. However, 26.8% of the variation still remains unexplained.

140

Looking at other tax compliance studies, future research can look at other predictor

variables such as tax complexity (Saad, 2010) and law enforcement (Nordiana,

2012).

5.7 Conclusion

In view of the current turmoil amongst consumers with the implementation of Goods

& Services Tax, this research was undertaken to shed some light on the concerns of

consumers. Although, the consumers in this research were restricted to those

currently pursuing their MBA in either USM, UKM, UM or UPM, it acts as stepping

stone towards many more extensive research in this very new field of GST in

Malaysia. Previous researchers on an international level had studied compliances to

income tax submissions whilst Malaysian studies on GST had focused on the tax

collectors- companies’ compliance to GST.

In order to make sense of the various factors that influence these consumers in

intending to accept GST, the renowned Theory of Planned Behavior was employed.

Being an expansively tested theory, it once again proved to be a significant and

appropriate theory in this context as well. This was determined through statistical

measures. In addition to that, antecedent variables to the attitude and perceived

behavorial control components of the theory were identified. This is to enhance the

contribution of this research as practitioners would then be able to understand the

formation of attitude and perceived behavorial control. By understanding this, they

141

can then engage in practical methods of enhancing these components of attitude and

perceived behavorial control.

From the statistical analysis, significant factors were identified and recommendations

were given on improvement methods that can be undertaken. The theoretical and

practical implications of this study are significant as it is in relation to a current and

less-explored subject matter. This one research has provided new and interesting

insight to the actual concerns and issues of the consumers which if addressed

effectively by the authorities will lead to better acceptance of GST. Consumers now

have to focus on utilizing the knowledge they have on GST to manage their

consumption. Only by managing their consumption will they be able to reduce the

tax burden incurred.

142

REFERENCES

Abd-El-Fattah, S. M., &Fakhroo, H. A. (2012). The Relationship among

PaternalPsychological Control and Adolescents’ Perfectionism and Self-

Esteem: A Partial Least Squares Path Analysis. Psychology, 3(5), 428-439.

Ajzen, I. (1985) From intentions to actions: A theory of planned behavior. IN

KUHL, J. & BECKMANN, J. (Eds.) Springer series in social psychology.

Berlin, Springer.

Ajzen, I. (1988). Attitudes, personality, and behavior. Chicago: Dorsey Press.

Ajzen, I. (1991). The theory of planned behavior.Organizational Behavior and

Human Decision Processes. Vol 50 179-211 doi: 10.1016/0749-

5978(91)90020-T

Ajzen, I. (2001). Nature and operation of attitudes.Annual Review of Psychology.

Vol. 52, 27-58.

Ajzen, I. (2002). Perceived behavioral control, self-efficacy, locus of control, and

thetheory of planned behavior. Journal of Applied Social Psychology.

32, 665-683.

Ajzen, I. (2006). Constructing a TpB Questionnaire: Conceptual and Methodological

Considerations.Retrieved from http://www.people.umass.edu/.

143

Ajzen, I. (2013). Constructing a theory of planned behavior questionnaire.

[electronic resource]. Amherst, MA, USA: University of Massachusetts.

Retrieved from http://people.umass.edu/aizen/tpb.html

Ajzen, I. and Madden, T.J. (1986). Prediction of goal-directed behaviour: attitudes,

intentions, and perceived control. Journal of Experimental Social

Psychology. Vol. 22, 453-74

Bandura, A. (1986). Social foundations of thought and action: A Social Cognitive

Theory, Prentice Hall, Englewood Cliffs, NJ.

Bandura, A. (1997). Self-efficacy: toward a unifying theory of behavioral change.

Psychological Review, 84, 191-215 Retrieved from

http://www.uky.edu/~eushe2/Bandura/Bandura1977PR.pdf

Barclay, D., Thompson, R., & Higgins, C. (1995). The Partial Least Squares (PLS)

Approach to Causal Modeling: Personal Computer Adoption and Use an

Illustration. Technology Studies, 2(2), 285-309.

Beauregard, T.A. (2012) Perfectionism, self-efficacy and OCB: the moderating role

of gender. Personnel Review. 41(5), 590-608

http://www.uky.edu/~eushe2/Bandura/Bandura1977PR.pdf

144

Ben-Ami,M., Hornik,J., Eden,D. &Kaplan,O.(2014). Boosting consumers’

self-efficacy byrepositioning the self.European Journal of Marketing. 48

(11/12), 1914 – 1938 Doi: 10.1108/EJM-09-2010-0502

Bernama (2014, May 30), GST implementation to add up to RM27b to Malaysia's

income.NewStraitTimes. Retrieved from: http://www.nst.com.my/latest/gst-

implementation-to-add-up-to-rm27b-to-malaysia-s-income-1.280974

Bernama (2013, October 25). Budget 2014: Full text of Prime Minister's speech The

Star Online Retrieved from http://www.thestar.com.my/News/Nation/2013

/10/25/Budget-2014-PM-speech.aspx/

Bernama (2015, April 1) Malaysia joins the ranks of many in embracing the GST.

News Strait Times Online Retrieved from http://www.nst.com.my/node/78934

Bernama (2015, February 25). Ahmad Maslan: GST will help reduce black economy.

Malaymail Online Retrieved from fromhttp://www.themalaymailonline.com

/article /ahmad-maslan-gst-will-help-reduce-black-economy

Bernama(2014, February 10) ‘Price of 24 selected daily essentials at 881 premises

nationwide’ NewStraitTimes Retrieved from:

http://www.nst.com.my/latest/price-of-24-selected-daily-essentials-at-881-

premises-nationwide-1.481400

http://www.nst.com.my/latest/gst-implementation-to-add-up-to-rm27b-to-malaysia-s-income-1.280974

http://www.nst.com.my/latest/gst-implementation-to-add-up-to-rm27b-to-malaysia-s-income-1.280974

http://www.nst.com.my/latest/price-of-24-selected-daily-essentials-at-881-premises-nationwide-1.481400

http://www.nst.com.my/latest/price-of-24-selected-daily-essentials-at-881-premises-nationwide-1.481400

145

Bhattacherjee, A. (2000). Acceptance of e-commerce services: The case of electronic

brokerages. IEEE Transactions on Systems, Man, and Cybernetics: Part A.

Systems and Humans. 30(4), 411–420

Bidin,Z&Shamsudin,F.M. (2013). Using Theory of Reasoned Action to Explain

Taxpayer Intention to Comply with Goods and Services Tax (GST). Middle-

East Journal of Scientific Research 17 (3), 387-394. DOI:

10.5829/idosi.mejsr.2013.17.03.12167

Bobek D.D. and Hatfield R.C. (2003) An Investigation of the Theory of Planned

Behavior and the Role of Moral Obligation in Tax Compliance. Behavioral

Research in Accounting. Vol. 15, No. 1 13-38.

Bobek, D.D. (1997). Tax Fairness: How do individuals Judge Fairness and What

Effect Does It Have on Their Behaviour. Unpublished PHD thesis, University

of Florida

Boone, H.N. & Boone, D.A. (2012).Analyzing Likert Data. Journal of Extension.50(2)

Retrieved from http://www.joe.org/joe/2012april/pdf/JOE_v50_2tt2.pdf

Boonyarat,N., Sofian,S. &Wadeecharoen,W. (2014). The Antecedents of Taxpayers

Compliance Behavior and the Effectiveness of Thai Local Government Levied

Tax. IBT Journal of Business Studies. 10 (1), 23-39

146

Burgess, C.N.W. (1996) Identifying the Wilful Traffic Offender. Unpublished

MsCDissertation, University of Exeter.

Chandra, S., Theng, Y. L., O’Lwin, M., & Foo, S. (2011).Examining Trust for

Organizational Collaborations via the Virtual World.Journal of Computer

Mediated Communication.

Chang, M. K. (1998). Predicting Unethical Behavior: A Comparison of the Theory of

Reasoned Action and the Theory of Planned Behavior.Journal of Business

Ethics. 17 1825-1834.

Cheng,T.C.E., Lam,D.Y.C. &Yeung,A.C.L. (2006). Adoption of internet banking: An

empirical study in HongKong. Decision support systems.42(3) 1558-1572

Retrieved from:

http://www.sciencedirect.com/science/article/pii/S0167923606000066

Chiou, J.S. (1998). The Effects of Attitude, Subjective Norm, and Perceived

Behavioral Control on Consumers’ Purchase Intentions: The Moderating

Effects of Product Knowledge and Attention to Social Comparison

Information. Proceedings of the National Science Council, Republic of

China.9(2), 298-308 Retrieved from

http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.202.7526&rep=rep1

&type=pdf

http://www.sciencedirect.com/science/article/pii/S0167923606000066

147

Choo, T.S. (2014) ‘The battle for GST acceptance’ theSundaily Retrieved from:

http://www.thesundaily.my/node/251014

Christensen, A. L. &Weichrich S.G. (1996). Tax fairness: Different Roles, Different

Perspectives.Advances in Taxation. 8, 27-61.

Christensen, A. L., Weichrich, S.G. &Gerbing-Newman, M.D. (1994). Impact of

Education on Perceptions of Tax Fairness.Advances in Taxation. 6, 63-97.

Cohen, J (1988). Statistical Power Analysis for the Behavioural Sciences (2nd

ed.).

New Jersey: Lawrence Erlbaum Associates.

Cohen,N. &Arieli,T. (2011) Field research in conflict environments: Methodological

challenges and snowball sampling. Journal of Peace Research 48(4), 423-435

Doi: 10.1177/0022343311405698

Conner, M., and Sparks, P. (2005).The theory of planned behaviour and health

behaviours.In M. Conner & P. Norman (Eds.).Predicting Health Behaviour:

Research and Practice with Social Cognition Models. 2, 170–222

Culbertson, H.M. (1968). What is an attitude? Journal of cooperative extension. 79-84

Retrieved from http://www.joe.org/joe/1968summer/

http://www.thesundaily.my/node/251014

http://www.joe.org/joe/1968summer/

148

Customs Guide Book (2003). Sales Tax Procedures - Series No 3. Retrieved from

http://www.ctim.org.my/file/news/15/00539_Customs%20Guide%20Book%2

0Series%20No.%203%20-%20Sales%20Tax%20Procedures.pdf

Damayanti, T.W. (2012). Changes On Indonesia Tax Culture, Is There A Way?

Studies Through Theory Of Planned Behavior. Journal of Arts, Science &

Commerce.Vol 3(4) 8-15

Davis, F. D., Bagozzi, R.P. &Warshaw P.R. (1989) User Acceptance of Computer

Technology: A Comparison of Two Theoretical Models. Management Science.

35: 982-1003.

Davis,F.D. (1989). Perceived usefulness, perceived PEOU and user acceptance of

information technology.MIS Quarterly. 13(3), 319-40

Dawkins.C.E. and Frass J.W. (2005). Decision of union workers to participate in

employee involvement: an application of the theory of planned behavior.

Employee Relations. 27(5), 511-531

Department of Statistics Malaysia (2011, August 28) Household Expenditure Trend

1993/94 - 2009/10 Retrieved from

http://www.statistics.gov.my/index.php?r=column/cthemeByCat&cat

=120&bul_id=RVVlT0tWVG03TDdCcEtXV29CNkZndz09&menu_id=amVo

WU54UTl0a21NWmdhMjFMMWcyZz09

149

Devos,K. (2014). Factors Influencing Individual Taxpayer Compliance

Behaviour.Doi: 10.1007/978-94-007-7476-6

Droomers,M., Schrijvers, C.T.M., Mackenbach, J.P. (2004) Educational differences in

the intention to stop smoking:Explanations based on the Theory of Planned

Behaviour. European Journal of Public Health. 14, 194-198. Retrieved from

http://www.ncbi.nlm.nih.gov/pubmed /15230509

Economic Report 2013/2014 ‘Federal Government Finance’ Ministry of Finance

Malaysia Retrieved from:

http://www.treasury.gov.my/pdf/economy/er/1314/jp6_2.pdf

Economic Report 2013/2014 ‘Federal Government Revenue’ Ministry of Finance

Malaysia Retrieved from:http://www.treasury.gov.my/pdf/economy/er/

1314/jp6_3.pdf

Erich, K., A. Niemirowski&Wearing,A. (2006). Shared Subjective Views, Intent

to Cooperate and Tax Compliance: Similarities between Australian Taxpayers

and TaxOfficers. Journal of Economic Psychology. 27(4) 502-517

Filho, D.B.F., Silva,J.A. & Rocha, E. (2011) What is R2 all about? Leviathan –

Cadernos de PesquisaPolítica. 3, 60-68

http://www.treasury.gov.my/pdf/economy/er/1314/jp6_3.pdf

150

Fishbein, M., &Ajzen, I. B. (1975). Belief, attitude, intention and behavior: an

introduction to theory and research. Reading, MA: Addison-Wesley

Fornell, C., &Bookstein, F. L. (1982). Two structural equation models: LISREL and

PLS appliead to consumer exit-voice theory. Journal of Marketing

Research.19, 440–452.

Fornell, C., &Laracker, D. F. (1981).Evaluating structural equation models with

unobservable variables and measurement error.Journal of Marketing Research,

18(3), 39-50.

Frey,B.S. &Schneider,F.(2000) Informal and Underground Economy Encyclopedia of

Social and Behavioral Science. Amsterdam: Elsevier Science Publishing

Company.

Gerbing, M. D. (1988). An Empirical Study of Taxpayer Perceptions of Fairness.

Unpublished PhD thesis, University of Texas, Austin.

Gerpott, T.J. (1990). Intracompany job transfers: an exploratory two sample study of

the buffering effects of interpersonal support. Prevention in Human

Services.Vol. 8,139-58.

Giantari, G.A.K., Zain,D., Rahayu,M. &Solimun (2013). The role of perceived

behavioral control and trust as mediator of experience on online purchasing

intentions relationship a study on youths in denpasar city

151

(Indonesia).International Journal of Business and Management Invention.

2(1), 30-38 Retrieved from

http://www.ijbmi.org/papers/Vol%282%291/Version_3/D213038.pdf

Gilligan, G. and G. Richardson (2005). Perceptions of Tax Fairness and Tax

Compliance in Australia and Hong Kong: A Preliminary Study. Journal of

FinancialCrime.12(4) 331-343.

Groebner,D.F., Shannon,P.W. &Fry,P.C. (2014) Business Statistics. Edinburgh Gate,

Essex: Pearson Education Limited.

Guo, Q., C. A. Johnson, J. B. Unger, et al. (2007). Utility of the Theory of Reasoned

Action and Theory of Planned Behavior for Predicting Chinese Adolescent

Smoking.Addictive Behaviors. 32(5) 1066-1081.

Guthrie, J.P. &Schwoerer, C.E. (1994).Individual and contextual influences on self-

assessed training needs.Journal of Organizational Behavior. 15, 405-22

Hai, O.T. & See, L.M. (2011).Behavioral Intention of Tax Non-Compliance among

Sole-Proprietors in Malaysia.International Journal of Business and Social

Science.2(6), 142-152

Hair, J. F., Hult, G. T. M., Ringle, C. M., &Sarstedt, M. (2014).A Primer on Partial

Least Squares Structural Equation Modeling (PLS – SEM). Thousand Oaks:

Sage.

152

Hair, J., Babin, W., & Anderson, R. (2010).Multivariate data analysis (7th ed.). Upper

Saddle River, N.J.: Prentice Hall.

Hair, J.F., Black, W.C., Babin, B.J., Anderson, R.E., &Tatham, R.L. (2006).

Multivariate Data Analysis, 6th Ed., New Jersey: Pearson Prentice Hall

Hanno, M. D., &Violette, G. R. (1996).An Analysis of Moral and Social Compliance

on Taxpayer Behaviour.Behavioural Research in Accounting. 8

Haron,H., Khalid,S.N.A and Ganesan,Y. (2011) A handbook for Business Research

Methods. Petaling Jaya: Pearson Malaysia Sdn. Bhd.

Harris, T.D. (1989). The effect of type of tax knowledge on individuals’ perceptions

of fairness and compliance with the federal income tax system: An empirical

study.

PhD Thesis, University of South Carolina.

Hung,S.Y., Chang,C.M. and Yu,T.J. (2006). Determinants of user-acceptance of

the e-government services: The case of online tax filing and payment

system. GovernmentInformation Quarterly. Volume 23 (1) 97-122

Retrieved fromhttp://www.sciencedirect.com/science

?_ob=ArticleListURL&_method=list&_ArticleListID=782942896&_sort

=r&_st=13&view=c&md5=4e76b1e2bbb644386fb4b284cb3f59ac&searc

htype=a

153

Ian, L.S. (2014, May 15). You don’t understand GST, so don’t protest, says deputy

finance minister. The Malaysian Insider Retrieved from

http://www.themalaysianinsider.com/malaysia/article/you-dont-understand-

gst-so-dont-protest-says-deputy-finance-minister

Idris,A.A. (2015, April 6) Hotel, restaurants cannot impose service charges without

collective agreements. New Strait Times Online Retrieved from

http://www.nst.com.my/node/79570

Income Tax Act Malaysia (1967).

Jalil, H. (2015, March 16). Consumers lack knowledge on new tax regime.

theSundaily. Retrieved from http://www.thesundaily.my/news/1356200

Jeffrey,C. &Weatherbolt, N. (1996). Ethical development, professional commitment

and rule observance attitudes; a study of CPAs and corporate

accountants.Behavioral Research in Accounting. 8, 8-31

Jones, C.R. (2009) Understanding and improving use-tax compliance: A theory of

planned behavior approach. Graduate Theses and Dissertations.University of

South Florida Retrieved from http://scholarcommons.usf.edu/etd/2032

Jones, John F. (1970). The Theory of Attitude Formation and Change and Its

Application to Social Group Work.United College Journal, Vol. 8 21-28.

http://scholarcommons.usf.edu/etd/2032

154

Retrieved from

http://web.psych.utoronto.ca/psy320/Required%20readings_files/week2-2.pdf

Joseph, S. (2013, August 19). Malaysia’s household debt situation.KiniBiz. Retrieved

from http://www.kinibiz.com/story/issues/44119/malaysia%E2%80%99s-

household-debt-situation-2.html

Kasipillai, J. (2000). A Practical Guide to Malaysian Taxation-Current Year

Assessment. Kuala Lumpur: McGraw-Hill.

Kasipillai,J. (2005) ‘ A comprehensive guide to Malaysian Taxation, Malaysia :

McGraw Hill Education

Kasipillai, J. (2013) ‘Gearing GST towards public acceptance’ The Star Online

Retrieved from: http://www.thestar.com.my/Business/Business-

News/2013/10/19/Gearing-GST-towards-public-acceptance-A-more-efficient-

way-of-taxation.aspx/

Khasawneh,A., Obeidat, M.I. & Al-Momani,M.A. (2008). Income Tax Fairness and

the Taxpayers’ Compliance in Jordan.Journal of Economic and Administrative

Sciences. 24(1) 15-39 Doi: 10.1108/10264116200800002

Kraft,P., Rise,J., Sutton,S. &Roysamb,E. (2005). Perceived difficulty in the theory of

planned behaviour: Perceived behavioural control or affective attitude? British

Journal of Social Psychology.44, 479–496

http://web.psych.utoronto.ca/psy320/Required%20readings_files/week2-2.pdf

http://www.kinibiz.com/story/issues/44119/malaysia%E2%80%99s-household-debt-situation-2.html

http://www.kinibiz.com/story/issues/44119/malaysia%E2%80%99s-household-debt-situation-2.html

http://www.thestar.com.my/Business/Business-News/2013/10/19/Gearing-GST-towards-public-acceptance-A-more-efficient-way-of-taxation.aspx/



155

Kulviwat S., Bruner II G.C., Hall R. &Neelankavil J.P. (2014). Self-Efficacy as an

Antecedent of Cognition and Affect In Technology Acceptance. Journal of

Consumer Marketing. 31 (3), 190-199.Retrieved

fromhttp://www.emeraldinsight.com/doi/abs/10.1108/JCM-10-2013-0727

Lack of clarity on GST leads to much confusion. (2015, April 2). Borneo

PostOnline.Retrieved from http://www.theborneopost.com/2015/04/02/lack-of-

clarity-on-gst-leads-to-much-confusion/

Lai, D. (2014, October 28) Countdown to GST theSundaily Retrieved from

http://www.thesundaily.my/news/866844

Lancaster, G. (2005) Research Methods in Management. Retrieved from

http://books.google.com.my/books?id=lDZ9RlrvUk8C&printsec=frontcover&

source=gbs_ge_summary_r&cad=0#v=onepage&q&f=false

Likert, R. (1932). A Technique for the Measurement of Attitudes.Archives of

Psychology, 22(140), 1-55.

Loo,E.C., McKrechar, M. &Hansford,A. (2007) The Effect of Introduction of Self-

Assessment on Tax Knowledge of Individual Taxpayers in Malaysia: Survey

Findings. Paper presented at the Accounting Studies International Conference,

Kuala Lumpur, Malaysia.


156

Maydeu-Olivares, A., & Garcia-Forero, C. (2010).Goodness-of-Fit Testing.

International Encyclopedia of Education, 7, 190-196.

McKerchar, M. (1995). Understanding small business taxpayers: Their sources of

information and level of knowledge of taxation. Australian Tax Forum.12(1),

25-41.

McLeod,S. (2007). Simply Psychology. Retrieved from http://www.simplypsychology

.org/obedience.html

MerdekaCenter (2014, April).Majority of Malaysians do not understand or accept the

GST. Retrieved from http://www.merdeka.org/pages/02_research.html

Michael, L. G. (1978). Tax Avoidance, Horizontal Equity and Tax Reform: A

Proposed Synthesis.Southern Economic Journal. 44(4) 798-812.

Ministry of Education Malaysia (2012) Public Institutions OfHigher Education (Pihe).

Retrieved from http://www.moe.gov.my/v/ipta

Moomal,A. &Zakaria,Z. (2014). Awareness, Acceptance and Expectations of

Malaysian Consumers towards Goods and Services Tax (GST).Journal of

Contemporary Management Sciences.Vol 4 (2) 15-27 Retrieved from

http://jmsciences.com/Presentation/Client/detail.aspx?id=252

http://www.moe.gov.my/v/ipta

157

Nordiana, R. (2012) The perception of taxpayers toward goods and services tax (GST)

implementation. Unpublished thesis, University Utara Malaysia Retrieved

from http://etd.uum.edu.my/3247/3/NORDIANA_RAMLI.pdf

Nunnally, J. C., & Bernstein, I. H. (1994), Psychometric Theory (3rd

ed.) New York:

McGraw Hill.

Official Website Malaysia Goods and Services Tax (GST), Royal Malaysian Customs

Department Retrieved from: http://gst.customs.gov.my/en/Pages/default.aspx

Oxford Dictionaries (2015) English Language-Theory Retrieved from

http://www.oxforddictionaries.com/ms/definisi/bahasainggeris/theory?searchD

ictCode=all

Palil, M.R. & Ibrahim, M.A. (2011).The Impacts of Goods and Services Tax (GST) on

Middle Income Earnests in Malaysia.World Review of Business Research .Vol.

1.No. 3. July 2011. 192-206. Retrieved from http://www.wbiconpro.com/213-

Rizal.pdf

Palil, M.R. (2010). Tax knowledge and tax compliance determinants in self-

assessment system in Malaysia.Thesis submitted to University of Birmigham

Paris, H., and S. V. Broucke (2008).Measuring Cognitive Determinants of Speeding:

An Application of the Theory of Planned Behavior.Transportation Research

Part F: Traffic Psychology and Behavior. 11(3) 168-180.

http://etd.uum.edu.my/3247/3/NORDIANA_RAMLI.pdf

http://gst.customs.gov.my/en/Pages/default.aspx

http://www.wbiconpro.com/213-Rizal.pdf

http://www.wbiconpro.com/213-Rizal.pdf

158

Podsakoff, P. M., Mackenzie, S. B., Lee, J. Y., &Podsakoff, N. P. (2003). Common

method biases in behavioural research: a critical review of the Literature and

Recommended Remedies. Journal of Applied Psychology, 88(5), 879-903.

Raghuram, S., Wiesenfeld, B.M. and Garud, R. (2003). Technology enabled work: the

role of self-efficacy in determining telecommuter adjustment and structuring

behavior. Journal of Vocational Behavior.63, 180-98.

Ramalingam, L., Noor, N.F.M. &Azmi, A.A.C. (2014).The Unified Theory of

Acceptance and Use of Technology (UTAUT) and the Goods and Service Tax

(GST) Application System.Research Journal of Applied Sciences, Engineering

and Technology 8(17), 1911-1916. Retrieved from:

http://www.maxwellsci.com/print/rjaset/v8-1911-1916.pdf

Richardson, G. (2005a). A Preliminary Study of the Impact of Tax Fairness

Perception Dimensions on Tax Compliance Behaviour in Australia.Australian

TaxForum.20(3): 407-434.

Richardson, M., & Sawyer, A. (2001).A taxonomy of the tax compliance literature:

Further findings, problems and prospects. Australian Tax Forum. 16(2), 137-

320.

http://www.maxwellsci.com/print/rjaset/v8-1911-1916.pdf

159

Rosipal, R., & Kramer, N. (2006).Overview and Recent Advances in Partial Least

Squares. In. Craig Saunders C, Grobelnik M, Gunn J, Shawe-Taylor J,

eds.Subspace, Latent Structure and Feature Selection: Statistical and

Optimization Perspectives Workshop (SLSFS 2005). New York: Springer-

Verlag; 2006, 34-51.

Royal Malaysian Customs (2013) ‘Malaysia Goods and Services Tax – Guide on

Supply’ Retrieved from:http://gst.customs.gov.my/en/rg

/SiteAssets/industry_guides_pdf/GST%20GUIDE%20ON%20SUPPLY%20Fi

nal%20Draft%202013%20050613%20G5%20PARA%20Treatment.pdf

Royal Malaysian Customs Department- Malaysia Goods & Services Tax (GST).

(2013, February 15) GST Guide- General. Retrieved from

http://gst.customs.gov.my/en/rg/Pages /rg_gen.aspx

Saad, N. (2010). Fairness Perceptions and Compliance Behaviour: The Case of

Salaried Taxpayers in Malaysia after Implementation of the Self-Assessment

System eJournal of Tax Research Volume 8, Number 1 32-63 Retrieved from

http://www.austlii.edu.au/au/ journals/eJTR/2010/3.pdf

Saad, N. (2014) Tax Knowledge, Tax Complexity and Tax Compliance: Taxpayers’

View Procedia - Social and Behavioral Sciences, 109, 1069 – 1075 Retrieved

from: http://ac.els-cdn.com/S1877042813052294/1-s2.0-S1877042813052294

main.pdf?_tid=b496988a-f0e2-11e4-98ee00000aab0f6b&acdnat=1430581901

_f3d07e276f21d53cbd25007f4c507887

160

Saira, L., M.A., Zariyawati, M.Am& Yoke-May, L. (2010). An exploratory study of

goods and services tax awareness in Malaysia.Political Managements and

Policies in Malaysia.Universiti Utara Malaysia Sintok, 265-276. Retrieved

from http://core.ac.uk/download/pdf/12118568.pdf

Salancik, G. and Pfeffer, J. (1978).The External Control of Organisations. New York,

NY. Harper & Row.

Santos, J. R. A (1999). Cronbach’s Alpha: A Tool for Assessing the Reliability of

Scales. Journal of Extension.37 (2). Retrieved from

http://www.joe.org/joe/1999april/tt3.php

Sarina,I., Haron,H., Ismail,I. &Vinten,G. (2007) Factors Influencing Auditors

Acceptance of the Practice Review System. Asian Academy of Management

Journal.Vol. 12(2) 57–82

Schmidt,D. &Karsten,R.(2015). Self-efficacy and Tax Research

Performance. Advances in Taxation. 17, 121-140. Retrieved from

http://dx.doi.org/10.1016/S1058-7497(06)17005-2

Seidl,C. &Traub, S. (2001).Taxpayers’ attitudes, behavior, and perception of fairness.

Pacific Economic Review. 6(2), 255-267 DOI: 10.1111/1468-0106.00131

http://core.ac.uk/download/pdf/12118568.pdf

http://www.joe.org/joe/1999april/tt3.php

161

Sekaran, U. (2000). Research Methods for Business: A Skill -Building Approach

(3ed.). New York: John Wiley & Sons, Inc

Sekaran, U. (2003). Research methods for business (4th

ed.). Hoboken, NJ: John Wiley

& Sons.

Sekaran, U. and Bougie, R. (2010).Research Methods for Business; A Skill Building

Approach. 5th Ed. New York: John Wiley and Sons.

Shamsuddin, A., Ruslan, M.I.M., Halim, A.A., Zahari,N.F. &Fazi, N.F.M. (2014).

Educators’ Awareness And Acceptance Towards Goods And Services Tax

(Gst) Implementation In Malaysia: A Study In Bandar Muadzam Shah,

Pahang. International Journal of Business, Economics and Law. Vol 4 (1) 1-10

Retrieved from http://ijbel.com/wp-content/uploads/2014 /06/KLB4103-

Amanuddin-Muhammad-Ishfaq-Afifah-Nur-

FatinEDUCATORS%E2%80%99-AWARENESS-AND-ACCEPTANCE.pdf

Shimp,T. and Kavas,A. (1984). The Theory of Reasoned Action Applied to computer

usage. Journal of Consumer Research. 15, 325-343.

Shin, H., Collier, D.A., & Wilson, D.D. (2000). Supply management orientation and

supplier or buyer performance. Journal of operations management, 18, 317-

333.

162

Siahaan, F.O.P. (2012). The Influence of Tax Fairness and Communication on

Voluntary Compliance: Trust as an Intervening Variable. International Journal

of Business and Social Science. 3(21) 191-198 Retrieved from

http://ijbssnet.com/journals/Vol_3_No_21_November_2012/20.pdf

Singh V (2003). Malaysian Tax Administration. 6th ed. Kuala Lumpur: Longman

Singh, V. (1999).Malaysian Tax Administration. 4th ed. Kuala Lumpur: Longman.

Singh, V. (2013, October 27) Budget 2014: Bold and Yet Cautious? theSundaily

Retrieved from http://www.thesundaily.my/news/866620

Sommer,L. (2011). The Theory Of Planned Behaviour And The Impact Of Past

Behaviour. International Business & Economics Research Journal. Volume

10 (1) 91-110

Sparks,P., Gothrie,C.A. &Shepherd,R. (1997). The dimensional structure of the

perceived behavioral control construct.Journal of Applied Social Psychology.

27, 418-438

Tan, L. M., & Chin-Fatt, C. (1999).The Impact of Tax Knowledge on the Perceptions

of Tax Fairness and Attitudes towards Compliance.Massey University,

Department of Accountancy and Business Law.


163

Tan,B. (2014, May 16) Umno Youth to assist the government in explaining about

GST NewStraitTimes Retrieved from: http://www.nst.com.my/latest/umno-

youth-to-assist-the-government-in-explaining-about-gst-1.598731

Tan, E. (2009) What Malaysian consumers want TheEdge Retrieved from:

http://www.theedgemalaysia.com/media-a-advertising/146966-what-

malaysian-consumers-want.html

Tay, A. (2001). Management's perception of MBA graduates in Malaysia. Journal of

Management Development. 20(3), 258-274

Taylor, S., & Todd, P. (1995).Assessing it usage: the role of prior

experience.Management Information Systems Quarterly. 19(3), 561–570.

Tenenhaus, M., Vinzi, V. E., Chatelin, Y. M., Lauro, C. (2005).PLS path

modelling.Computational Statistics & Data Analysis, 48, 159-205.

Teo,T. & Lee, C.B. (2010). Explaining the intention to use technology among student

teachers.Campus-Wide Information Systems. 27(2), 60 – 67 Doi

10.1108/10650741011033035

Tetrick, L.E. and Larocco,J.M (1987). Understanding, prediction, and control as

moderators of the relationships between perceived stress, satisfaction, and

psychological well-being. Journal of Applied Psychology.72, 538-43.

http://www.theedgemalaysia.com/media-a-advertising/146966-what-malaysian-consumers-want.html

http://www.theedgemalaysia.com/media-a-advertising/146966-what-malaysian-consumers-want.html

164

The Malaysian Insider (2014, April 20) ‘Putrajaya to spend RM1.66 million in 2 months to

brief public on GST’ Retrieved from: http://www.themalaysianinsider.com/malaysia/article

/government-to-spend-rm1.66-million-in-2-months-to-brief-public-on-gst

The Star Online (2014, May 1) Thousands at May Day rally against GST’ Retrieved

from http://www.thestar.com.my/News/Nation/2014/05/01/May-Day-anti-

GST-rally-Kuala-Lumpur/

Thomas,C. (2012). Assessing Tax Fairness Dimensions in a Small Developing

Economy.Business and Economics Journal. 62.

Trevino, L. (1986). Ethical decision making in organizations: A person-situation

Interactionist model. Academy of Management Review, 601–617.

Trivedi, U.V., Shehata, M. &Mestelman, S.(2005) Attitudes, Incentives, and Tax

Compliance. Canadian tax journal. vol. 53, no 1

Turman, G. T. (1995). Perceptions of Vertical Equity and Noncompliant Income Tax

Behaviour: An Experimental Test of Inequity Theory. Unpublished PhD thesis.

Virginia Commonwealth University, Virginia.

Vijayasarathy L.R. (2003). Predicting consumer intentions to use on-line shopping:

the case for an augmented technology acceptance model.Information and

Management.

165

Voon, S. (2013) GST: Good or Bad?TheEdge Retrieved from

http://www.theedgemalaysia.com/personal-finance/257920-gst-good-or-

bad.html

Voorhis, W.V. & Morgan, C.R. (2007). Understanding Power and Rules of Thumb

for determining sample sizes. Tutorials in Quantitative Methods for

Psychology. 3(2), 43-50.

Wan,C., Cheung,R., Shen, G.Q. (2012). Recycling attitude and behaviour in university

campus: a case study in Hong Kong.Facilities. Vol. 30 (13/14) 630 - 646

Doi: 10.1108/02632771211270595

Why Do People Find It Hard To Accept GST If It Really Is Beneficial? (2015,

November 28) GST Malaysia Info. Retrieved from

http://gstmalaysiainfo.com/people-find-hard-accept-gst-really-beneficial/

Wold, H. (1980).Model construction and evaluation when theoretical knowledge is

scarce.In J. Kmenta, & J. B. Ramsey (Eds.).Evaluation of econometric

models.47–74.

Xavier,J.A. (2013,October 29) Will GST increase govt revenue? New Strait

TimesOnline Retrieved from http://www.nst.com.my/opinion/columnist/will-gst-

increase-govt-revenue-1.387243

http://www.theedgemalaysia.com/personal-finance/257920-gst-good-or-bad.html

http://www.theedgemalaysia.com/personal-finance/257920-gst-good-or-bad.html

http://www.nst.com.my/opinion/columnist/will-gst-increase-govt-revenue-1.387243

http://www.nst.com.my/opinion/columnist/will-gst-increase-govt-revenue-1.387243

166

Zamhari,A. (2014, April 10) Think tank: Middle-class biggest losers with GST.

MalaysiaKini. Retrieved from http://www.malaysiakini.com/news/259644

Zikmund, W. G. (2003). Business Research Method (8th ed.). Cincinnati, Ohio:

Thomson/South-Western.

167

APPENDIX A

ECONOMIC REPORT 2010-2014 - FEDERAL GOVERNMENT REVENUE

168

APPENDIX B

ECONOMIC REPORT 2010-2014 - FEDERAL GOVERNMENT FINANCE

169

APPENDIX C

RESULTS OF MERDEKA CENTRE SURVEY ON GST

176

APPENDIX D – QUESTIONNAIRE

31 January 2015

Dear Sir/Madam,

FACTORS INFLUENCING CONSUMERS’ LEVEL OF ACCEPTANCE WITH

GOODS & SERVICES TAX (GST)

I am conducting a research on the above topic as part of the requirement for my Masters in

Business Administration programme with the Graduate School of Business, UniversitiSains

Malaysia. The main objective of the research is to gather your perceptions on the Goods &

Services Tax (GST) which will be implemented in Malaysia from 1st April 2015. Even

though our government has made a decision to implement GST, we would like to gather your

level of acceptance or agreement with it.

The questionnaire would take only about 15 minutes of your time and it would contribute

significantly to the success of this research. Strict confidentiality is assured.

For any queries on the research, you can email me, Shalene at [email protected] or call

at 016-4178005.

If you are interested in the findings of this research, kindly note down your email address at

the bottom of the filled questionnaire. The summary of the findings will be e-mailed to you.

I would like to thank you in advance for your kind support and participation.

mailto:[email protected]

177

Yours Sincerely,

--------------------------------- ------------------------------- -------------------------------

(ShaleneKalyanasundaram) (Professor HasnahHaron) (DrYulitaHanumIskandar)

MBA Student Supervisor Co-Supervisor

178

Dear Respondents,

There are FIVE sections in the questionnaire, Section A to Section E. Kindly answer all

sections.

SECTION A: Demographic Profile

Please tick on the box which fits your profile.

1. Age

18-25 41-60

26-40 61 and above

2. Gender

Male Female

3. Nationality

Malaysian Others

4. Current Level of Education

Diploma Others

Pre- University Please specify:

__________________

Undergraduate

Postgraduate

179

5. Please indicate the programme that you are enrolled in

MBA Others

Please specify:

________________________

DBA

PhD

6. Please indicate your mode of study

Full Time

Part Time

7. If your response to No. 5 is MBA, please indicate the following:

Institute

8. If your response to Question 6 is part time, please indicate whether you are (You can

tick more than one):

Self-employed Working in an organization

Owner of Business Unemployed

180

9. If your response to No. 8 was “Working in an Organization”, please indicate your

position level:

Top Management (CEO, President, General Manager)

Finance/ Accounts Manager

Marketing/Sales Manager

Technical/Operations Manager

Others

Please Specify: _________________________________

10. Income Level

RM 1000 – RM 5000 RM 10001 – RM 15000

RM 5001 – RM 10000 > RM 15001

181

SECTION B

Goods and Services Tax (GST) will be implemented in Malaysia on 1st April 2015. GST is a

consumption tax based on value-added concept. It is a multi-stage tax and is imposed on all

goods and services except for those exempted.

Please read each statement carefully and indicate your degree of agreement with it by

circling the appropriate number on the 5- point scale given below. Please give your

honest opinion.

1 2 3 4 5

Strongly Disagree Neutral Agree Strongly

Disagree (D) (N) (A) Agree

(SD) (SA)

(i) Rule Observance Behaviour

Rule observance behavior relates to obedience to authority.

Statements SD D N A SA

1 I always stick to the letter of

rules

1 2 3 4 5

2 I always check to see that I am

following rules

1 2 3 4 5

3 I expect others to follow rules 1 2 3 4 5

4 I have to follow strict rules at all

times

1 2 3 4 5

5 I belief rules lead to

improvement

1 2 3 4 5

182

(ii) Perception of tax fairness

Fairness perception is the judgement of individuals that something or someone is free of

biasness and injustice


1 I believe the government utilizes a reasonable amount of

tax revenue to achieve social goals, such as the provision

of benefits for low-income families.

1 2 3 4 5

2 I believe everyone pays their fair share of tax under the

proposed GST system

1 2 3 4 5

3 I think the government spends sufficient tax revenue

onnecessary welfare assistance.

1 2 3 4 5

4 I will receive fair value from the government in return

for my GST paid

1 2 3 4 5

5 It is fair that low-income earners receive more benefits

from the government compared to high-income earners.

1 2 3 4 5

6 The GST that I have to pay isfair considering the

benefits I receive from the government. 1 2 3 4 5

(iii)Attitude

Attitude is an individual’s positive and negative feelings about performing a target behavior.

The feelings are brought about after evaluations conducted based on beliefs.


1. GST is a good implementation for the improvement of

the country

1 2 3 4 5

2. GST is a wise decision for the country and its Citizens 1 2 3 4 5

3. I am willing to pay taxes as it would help my country. 1 2 3 4 5

4. I would feel guilty if I do not accept the implementation

of GST

1 2 3 4 5

5. I am comfortable with the implementation of GST 1 2 3 4 5

(iv) Subjective Norm

Subjective Norm is the individual’s perception that most people who are important to

him/her think he/she should not perform the behavior in question.


1. My peers would think that I should accept GST 1 2 3 4 5

2. My peers would accept GST 1 2 3 4 5

3. My family members would think that I should accept

GST

1 2 3 4 5

4. My family members would accept GST 1 2 3 4 5

183

5. I am very particular of what my peers and family think

of me

1 2 3 4 5

(v) Self-Efficacy

Self-efficacy is defined as people’s judgments of their capabilities to organize and execute

courses of action required to attain designated types of performances.


1. I would be able to manage my finances even with the

implementation of GST

1 2 3 4 5

2. I will be able to manage consumption in order to avoid

my GST burden

1 2 3 4 5

3. I believe in my ability to spend within my means even

with implementation of GST.

1 2 3 4 5

4. Managing my expenditure is not a problem for me 1 2 3 4 5

(vi) Perceived Behavioral Control

Perceived behavioral control is the perceived internal and external constraints on behavior

or intention to behave


1. It is easy for me to accept GST 1 2 3 4 5

2. There are no barriers that would prevent me from

accepting GST

1 2 3 4 5

3 I have no problems in accepting GST 1 2 3 4 5

(vii) Level of Acceptance

Level of Acceptance refers to the level of intention to accept

Please note that acceptance of GST refers to your level of agreement that GST should

be implementedand not your act of paying GST.


1. I intend to accept GST 1 2 3 4 5

2. I intend to help others accept GST 1 2 3 4 5

184

SECTION C(with marks stated – Correct Answer : 1 ; Wrong Answer : 0)

Please read each statement carefully and indicate your degree of agreement with it by

circling either

1 (YES) or 2 (NO).

(viii) Level of Tax Knowledge

No Questions YES NO

1 The GST is a legitimate way for the government to collect revenue to

manage an economy

1 0

2 The GST is tax based on consumption and not income 1 0

3 The GST is not a new tax, it is to replace the Sales & Service Tax

(SST)

1 0

4 Companies earning RM 500,000 per annum/year is subjected to the

collection of GST

1 0

5 GST will be implemented from 1st January 2016 0 1

6 A standard rated product will be imposed with 5% GST. 0 1

7 The three categories of products is Zero-rated, Standard rated and

Exempt-rated

1 0

8 I will be paying the same amount of tax under GST for professional

services because I initially had to pay 6% service tax for the service

1 0

9 I do not have to pay GST for zero rated and exempted products 1 0

10 There is no difference between zero-rated and exempt rated products 0 1

SECTION D(with marks stated – Correct Answer : 1 ; Wrong Answer : 0)

This section provides you with 5 short scenarios.

Pleasereadthescenarioscarefullyandtickyourchoiceofanswer in Part A (oneansweronly)

withregardstotheGoods/Servicescategoryandtickyourchoiceofanswer in Part B

(oneansweronly), withregardstothetaxratefortheGoods/Servicescategorythat you

haveticked in Part A.

Part A : Goods/ Servicessupplycategoryindicatingwhethertheproduct/service in

thescenarioissubjectedtostandard-rated GST, zero-rated GST or isexemptrated.

Part B : Thetaxratethattheproducts/serviceswill be subjectedto ; None, 0% or 6%

185

SECTION E

Kindly provide us with your views and comments on the statements and questions posed

below.

1. Your views on other government initiatives that can be taken to lessen the

burden of GST. Amongst the current initiatives are the distribution of Bantuan

Rakyat 1 Malaysian (BRIM) and reduction of income tax percentage for

individuals.

_____________________________________________________________________

_____________________________________________________________________

_____________________________________________________________________

_____________________________________________________________________

____________________________________________________________________

2. ‘Government initiatives should encompass all groups of people and not only to

low income group’. Your views on the above statement.

Scenarios Part A Part B

Goods/Servicessupplycategory Taxrate

Standard

rated

Zero rated Exempt

rated

6% 0% None

En Mat had an accidentand had

toundergosurgery in a localprivate hospital.

Theserviceprovidedbythe hospital tohimis:

0 0 1 0 0 1

AuntyLimwenttothebookstoretobuyherthre

edaughtersrevisionbooksfortheirschoolsubj

ects.

Thebooksboughtareunderthecategoryof :

0 1 0 0 1 0

MrLingambrought his

familyoutfordinnertohavethefamous

‘Penang FriedChicken’.

Thefriedchickenwill be:

1 0 0 1 0 0

Swee An wenttothemarkettobuy 2

kilogramsofchicken. Thechickenwill be: 0 1 0 0 1 0

Markbrought his

familyforChristmasshoppingat a

clothingstore. Theclothesare:

1 0 0 1 0 0

186

_____________________________________________________________________

_____________________________________________________________________

_____________________________________________________________________

_____________________________________________________________________

____________________________________________________________________

3. In your opinion, who will be affected most by the implementation of GST in

Malaysia. Why?

_____________________________________________________________________

_____________________________________________________________________

_____________________________________________________________________

_____________________________________________________________________

____________________________________________________________________

_____________________________________________________________________

4. In your opinion, will there be an increase in price with the implementation of

GST? Why?

_____________________________________________________________________

_____________________________________________________________________

_____________________________________________________________________

_____________________________________________________________________

____________________________________________________________________

_____________________________________________________________________

5. Other Comments

_____________________________________________________________________

_____________________________________________________________________

_____________________________________________________________________

_____________________________________________________________________

____________________________________________________________________

_____________________________________________________________________

THANK YOU FOR YOUR PARTICPATION

187

APPENDIX E –PLS ALGORITHM REPORT (PILOT STUDY)

Cronbachs Alpha

Model 1

Model 2

Model 3

Cronbachs Alpha

ATT 0.898

PTF 0.920

ROB 0.848

Cronbachs Alpha

LTK 1.000

PBC 0.878

SE 0.833

Cronbachs Alpha

ATT 0.898

BI 0.868

LTK 1.000

PBC 0.878

PTF 0.920

ROB 0.848

SE 0.833

SN 0.885

188

APPENDIX F –PLS ALGORITHM REPORT

MODEL 1

Path Coefficients

ATT PTF ROB

ATT

PTF 0.550

ROB 0.256

Outer Loadings

ATT PTF ROB

ATT1 0.932

ATT2 0.925

ATT3 0.752

ATT4 0.819

ATT5 0.886

PTF1 0.897

PTF2 0.837

PTF3 0.501

PTF4 0.918

PTF5 0.689

PTF6 0.592

ROB1 0.897

ROB2 0.905

ROB3 0.825

ROB4 0.808

ROB5 0.809

189

R Square

Average Variance Extracted (AVE)

AVE

ATT 0.749

PTF 0.571

ROB 0.722

R Square

ATT 0.499 R Square Adjusted

R Square

ATT 0.494

190

Composite Reliability

Cronbachs Alpha

Cronbachs Alpha

ATT 0.915

PTF 0.839

ROB 0.903

Discriminant Validity

Fornell-Larcker Criterion

ATT PTF ROB

ATT 0.866

PTF 0.668 0.756

ROB 0.510 0.462 0.850


ATT 0.937

PTF 0.884

ROB 0.928

191

MODEL 2

Path Coefficients LTK PBC SE

LTK 0.129

PBC

SE 0.509

Outer Loadings LTK PBC SE

LTK 1.000

PBC1 0.950

PBC2 0.870

PBC3 0.931

SE1 0.867

SE2 0.824

SE3 0.854

SE4 0.799

R Square

R Square

PBC 0.281

R Square Adjusted

192

R Square

PBC 0.275

Average Variance Extracted (AVE) AVE

LTK 1.000

PBC 0.842

SE 0.699



LTK 1.000

PBC 0.941

SE 0.903

193

Cronbachs Alpha Cronbachs Alpha

LTK 1.000

PBC 0.907

SE 0.858

Discriminant Validity Fornell-Larcker Criterion

LTK PBC SE

LTK 1.000

PBC 0.151 0.917

SE 0.043 0.514 0.836

MODEL 3

Outer Loadings (Before removal of SN5)

ATT LA LTK PBC PTF ROB SE SN

ATT1 0.931

ATT2 0.923

ATT3 0.748

ATT4 0.825

ATT5 0.889

LA1 0.961

LA2 0.948

LTK 1.000

PBC1 0.950

PBC2 0.873

PBC3 0.928

PTF1 0.897

PTF2 0.837

194

PTF3 0.501

PTF4 0.918

PTF5 0.689

PTF6 0.592

ROB1 0.897

ROB2 0.904

ROB3 0.825

ROB4 0.808

ROB5 0.809

SE1 0.867

SE2 0.824

SE3 0.854

SE4 0.799

SN1 0.855

SN2 0.648

SN3 0.642

SN4 0.892

SN5 0.489

Path Coefficients (After removal of Sn5) ATT LA LTK PBC PTF ROB SE SN

ATT 0.570

LA

LTK 0.128

PBC 0.243

PTF 0.548

ROB 0.257

SE 0.508

SN 0.100

195

Outer Loadings (After removal of SN5) ATT LA LTK PBC PTF ROB SE SN

ATT1 0.931

ATT2 0.923

ATT3 0.748

ATT4 0.825

ATT5 0.889

LA1 0.961

LA2 0.948

LTK 1.000

PBC1 0.950

PBC2 0.873

PBC3 0.929

PTF1 0.897

PTF2 0.837

PTF3 0.501

PTF4 0.918

PTF5 0.689

PTF6 0.592

ROB1 0.897

ROB2 0.904

ROB3 0.825

ROB4 0.808

ROB5 0.809

SE1 0.867

SE2 0.824

SE3 0.854

SE4 0.799

SN1 0.850

SN2 0.676

SN3 0.669

SN4 0.884

R Square R Square

ATT 0.496

LA 0.732

PBC 0.280

196

R Square Adjusted

R Square

ATT 0.491

LA 0.728

PBC 0.273

Average Variance Extracted (AVE) AVE

ATT 0.750

LA 0.911

LTK 1.000

PBC 0.842

PTF 0.571

ROB 0.722

SE 0.699

SN 0.602

197



ATT 0.937

LA 0.954

LTK 1.000

PBC 0.941

PTF 0.884

ROB 0.928

SE 0.903

SN 0.856

Cronbachs Alpha Cronbachs Alpha

ATT 0.915

LA 0.903

LTK 1.000

PBC 0.907

PTF 0.839

ROB 0.903

SE 0.858

SN 0.781

198

Discriminant Validity

Fornell-Larcker Criterion

ATT LA LTK PBC PTF ROB SE SN

ATT 0.866

LA 0.833 0.955

LTK 0.311 0.372 1.000

PBC 0.793 0.772 0.150 0.918

PTF 0.666 0.666 0.216 0.607 0.756

ROB 0.509 0.533 0.192 0.452 0.461 0.850

SE 0.486 0.390 0.043 0.513 0.319 0.249 0.836

SN 0.698 0.683 0.127 0.757 0.619 0.478 0.304 0.776

199

APPENDIX G – BOOTSTRAPPING PLS REPORT

MODEL 1

Path Coefficients

Mean, STDEV, T-Values, P-Values

Original Sample (O)

Sample Mean (M)

Standard Error (STERR)

T Statistics (|O/STERR|)

P Values

PTF -> ATT 0.550 0.551 0.054 10.280 0.000

ROB -> ATT 0.256 0.256 0.067 3.853 0.000

Outer Loadings Mean, STDEV, T-Values, P-Values

Original Sample (O)

Sample Mean (M)



P Values

ATT1 <- ATT 0.932 0.932 0.009 101.798 0.000

ATT2 <- ATT 0.925 0.926 0.013 70.915 0.000

ATT3 <- ATT 0.752 0.753 0.046 16.387 0.000

ATT4 <- ATT 0.819 0.819 0.033 24.716 0.000

ATT5 <- ATT 0.886 0.887 0.020 45.373 0.000

PTF1 <- PTF 0.897 0.897 0.015 59.627 0.000

PTF2 <- PTF 0.837 0.836 0.023 36.395 0.000

PTF3 <- PTF 0.501 0.503 0.079 6.324 0.000

PTF4 <- PTF 0.918 0.918 0.012 75.332 0.000

PTF5 <- PTF 0.689 0.688 0.044 15.735 0.000

PTF6 <- PTF 0.592 0.590 0.080 7.391 0.000

ROB1 <- ROB 0.897 0.896 0.023 38.466 0.000

ROB2 <- ROB 0.905 0.903 0.018 49.252 0.000

ROB3 <- ROB 0.825 0.821 0.033 25.138 0.000

200

ROB4 <- ROB 0.808 0.809 0.025 32.900 0.000

ROB5 <- ROB 0.809 0.804 0.042 19.460 0.000

MODEL 2

Path Coefficients Mean, STDEV, T-Values, P-Values

Original Sample (O)

Sample Mean (M)



P Values

LTK -> PBC 0.129 0.132 0.057 2.272 0.024

SE -> PBC 0.509 0.515 0.050 10.117 0.000


Original Sample (O)

Sample Mean (M)



P Values

LTK <- LTK 1.000 1.000 0.000

PBC1 <- PBC 0.950 0.950 0.010 90.954 0.000

PBC2 <- PBC 0.870 0.868 0.038 22.635 0.000

PBC3 <- PBC 0.931 0.932 0.011 87.608 0.000

SE1 <- SE 0.867 0.867 0.020 43.119 0.000

SE2 <- SE 0.824 0.823 0.030 27.397 0.000

SE3 <- SE 0.854 0.852 0.033 26.156 0.000

SE4 <- SE 0.799 0.798 0.041 19.274 0.000

201

MODEL 3

Path Coefficients Mean, STDEV, T-Values, P-Values

Original Sample (O)

Sample Mean (M)



P Values

ATT -> LA 0.570 0.574 0.061 9.379 0.000

LTK -> PBC 0.128 0.128 0.056 2.304 0.022

PBC -> LA 0.243 0.236 0.076 3.207 0.001

PTF -> ATT 0.548 0.555 0.053 10.422 0.000

ROB -> ATT 0.257 0.252 0.059 4.361 0.000

SE -> PBC 0.508 0.509 0.052 9.792 0.000

SN -> LA 0.100 0.105 0.062 1.625 0.105


Original Sample (O)

Sample Mean (M)



P Values

ATT1 <- ATT 0.931 0.931 0.010 93.594 0.000

ATT2 <- ATT 0.923 0.923 0.014 66.727 0.000

ATT3 <- ATT 0.748 0.745 0.047 15.881 0.000

ATT4 <- ATT 0.825 0.823 0.033 24.663 0.000

ATT5 <- ATT 0.889 0.889 0.018 48.547 0.000

LA1 <- LA 0.961 0.962 0.005 181.620 0.000

LA2 <- LA 0.948 0.947 0.012 78.141 0.000

LTK <- LTK 1.000 1.000 0.000

PBC1 <- PBC 0.950 0.950 0.010 91.050 0.000

PBC2 <- PBC 0.873 0.873 0.033 26.446 0.000

PBC3 <- PBC 0.929 0.929 0.011 85.535 0.000

PTF1 <- PTF 0.897 0.898 0.015 59.512 0.000

PTF2 <- PTF 0.837 0.836 0.024 34.524 0.000

PTF3 <- 0.501 0.495 0.082 6.149 0.000

202

PTF

PTF4 <- PTF 0.918 0.919 0.012 78.190 0.000

PTF5 <- PTF 0.689 0.688 0.045 15.375 0.000

PTF6 <- PTF 0.592 0.587 0.080 7.404 0.000

ROB1 <- ROB 0.897 0.895 0.023 39.664 0.000

ROB2 <- ROB 0.904 0.903 0.017 53.435 0.000

ROB3 <- ROB 0.825 0.822 0.031 26.963 0.000

ROB4 <- ROB 0.808 0.808 0.025 31.928 0.000

ROB5 <- ROB 0.809 0.805 0.039 20.636 0.000

SE1 <- SE 0.867 0.866 0.018 48.809 0.000

SE2 <- SE 0.824 0.822 0.032 26.130 0.000

SE3 <- SE 0.854 0.850 0.033 25.923 0.000

SE4 <- SE 0.799 0.797 0.042 19.247 0.000

SN1 <- SN 0.850 0.854 0.025 34.663 0.000

SN2 <- SN 0.676 0.669 0.077 8.819 0.000

SN3 <- SN 0.669 0.664 0.078 8.580 0.000

SN4 <- SN 0.884 0.886 0.018 48.018 0.000

203

APPENDIX H – IBM SPSS STATISTICS REPORT

APPENDIX H -1:Frequencies of Demographics

Statistics

Age Gender Nationality Education Programme Mode

N Valid 228 228 228 228 228 228

Missing 0 0 0 0 0 0

Institution Employment Position Income

N Valid 228 228 228 228

Missing 0 0 0 0

Frequency Table

Age

Frequency Percent Valid Percent

Cumulative

Percent

Valid 18-25 9 3.9 3.9 3.9

26-10 210 92.1 92.1 96.1

41-60 9 3.9 3.9 100.0

Total 228 100.0 100.0

204

Gender


Cumulative

Percent

Valid Male 106 46.5 46.5 46.5

Female 122 53.5 53.5 100.0

Total 228 100.0 100.0

Nationality


Cumulative

Percent

Valid Malaysian 228 100.0 100.0 100.0

Education


Cumulative

Percent

Valid Postgraduate 228 100.0 100.0 100.0

Programme


Cumulative

Percent

Valid MBA 228 100.0 100.0 100.0

205

Mode


Cumulative

Percent

Valid Full Time 32 14.0 14.0 14.0

Part Time 196 86.0 86.0 100.0

Total 228 100.0 100.0

Institution


Cumulative

Percent

Valid UniversitiSains Malaysia 86 37.7 37.7 37.7

UniversitiKebangsaan

Malaysia

48 21.1 21.1 58.8

Universiti Malaya 43 18.9 18.9 77.6

Universiti Putra Malaysia 51 22.4 22.4 100.0

Total 228 100.0 100.0

Employment


Cumulative

Percent

Valid Not Applicable 32 14.0 14.0 14.0

Self-Employed 8 3.5 3.5 17.5

Owner of Business 5 2.2 2.2 19.7

Working in an Organization 181 79.4 79.4 99.1

206

Unemployed 2 .9 .9 100.0

Total 228 100.0 100.0

Position


Cumulative

Percent


Top Management 3 1.3 1.3 14.9

Finance/ Accounts Manager 8 3.5 3.5 18.4

Marketing/ Sales Manager 24 10.5 10.5 28.9

Technical/ Operations

Manager

22 9.6 9.6 38.6

Others 140 61.4 61.4 100.0

Total 228 100.0 100.0

Income


Cumulative

Percent


RM 1000 - RM 5000 126 55.3 55.3 68.9

RM 5001- RM 10000 35 15.4 15.4 84.2

RM 10001 - RM 15000 33 14.5 14.5 98.7

> RM 15000 3 1.3 1.3 100.0

207

Income


Cumulative

Percent


RM 1000 - RM 5000 126 55.3 55.3 68.9

RM 5001- RM 10000 35 15.4 15.4 84.2

RM 10001 - RM 15000 33 14.5 14.5 98.7

> RM 15000 3 1.3 1.3 100.0

Total 228 100.0 100.0

208

APPENDIX H- 2 Total Variance Explained

Compone

nt

Initial Eigenvalues

Extraction Sums of

Squared Loadings

Rotation Sums of

Squared Loadings

Total

% of

Varian

ce

Cumulati

ve % Total

% of

Varian

ce

Cumulati

ve %

Tota

l

% of

Varian

ce

Cumulati

ve %

1 12.94

1

41.746 41.746

12.94

1

41.746 41.746

7.44

7

24.022 24.022

2

2.671 8.616 50.362 2.671 8.616 50.362

4.10

8

13.251 37.273

3

2.342 7.554 57.916 2.342 7.554 57.916

3.77

6

12.182 49.455

4

1.871 6.037 63.953 1.871 6.037 63.953

3.06

1

9.875 59.330

5

1.491 4.809 68.762 1.491 4.809 68.762

1.99

6

6.439 65.769

6

1.222 3.941 72.703 1.222 3.941 72.703

1.91

1

6.164 71.932

7

1.056 3.407 76.110 1.056 3.407 76.110

1.29

5

4.177 76.110

8 .780 2.515 78.625

9 .756 2.440 81.065

10 .579 1.867 82.932

11 .554 1.788 84.720

12 .495 1.597 86.317

209

13 .442 1.427 87.744

14 .416 1.342 89.086

15 .394 1.272 90.357

16 .361 1.164 91.522

17 .328 1.059 92.581

18 .319 1.028 93.609

19 .276 .891 94.500

20 .249 .804 95.304

21 .217 .699 96.003

22 .203 .653 96.657

23 .188 .606 97.263

24 .167 .540 97.803

25 .150 .482 98.286

26 .115 .370 98.656

27 .104 .337 98.993

28 .093 .299 99.292

29 .081 .262 99.554

30 .078 .252 99.807

31 .060 .193 100.000

Extraction Method: Principal Component Analysis.

210

APPENDIX H-3: Independent t- tests

GENDER

Group Statistics

Gender N Mean Std. Deviation Std. Error Mean

AveLA MALE 105 3.0762 1.09145 .10651

FEMALE 122 2.8279 1.05973 .09594

Independent Samples Test

Levene's Test

for Equality of

Variances t-test for Equality of Means

F Sig. t df

Sig.

(2-

tailed)

Mean

Difference

Std. Error

Difference

95% Confidence

Interval of the

Difference

Lower Upper

AveLA Equal

variances

assumed

.207 .650 1.736 225 .084 .24832 .14304 -.03354 .53018

Equal

variances not

assumed

1.732 217.922 .085 .24832 .14335 -.03422 .53086

MODE

Group Statistics

Mode N Mean Std. Deviation Std. Error Mean

AveLA FULL TIME 32 2.7031 1.08404 .19163

PART TIME 196 2.9847 1.07406 .07672

Independent Samples Test

211

Levene's Test for

Equality of

Variances t-test for Equality of Means

F Sig. t df

Sig. (2-

tailed)

Mean

Difference

Std. Error

Difference

95% Confidence

Interval of the

Difference

Lower Upper

AveLA Equal

variances

assumed

1.237 .267 -

1.373 226 .171 -.28157 .20505 -.68561 .12248

Equal

variances not

assumed

-

1.364 41.563 .180 -.28157 .20642 -.69827 .13513

APPENDIX H-4: One-Way ANOVA

INSTITUTION

Descriptives

AveLA

N Mean

Std.

Deviation Std. Error

95% Confidence Interval for

Mean

Minimum Maximum Lower Bound Upper Bound

USM 86 2.6977 .99493 .10729 2.4844 2.9110 1.00 5.00

UKM 48 2.8542 .96733 .13962 2.5733 3.1350 1.00 5.00

UM 43 2.8256 1.09590 .16712 2.4883 3.1628 1.00 5.00

UPM 51 3.5490 1.09661 .15356 3.2406 3.8574 1.00 5.00

Total 228 2.9452 1.07753 .07136 2.8046 3.0858 1.00 5.00

ANOVA

AveLA

Sum of Squares df Mean Square F Sig.

Between Groups 24.877 3 8.292 7.782 .000

Within Groups 238.688 224 1.066

Total 263.565 227

212

INCOME

Descriptives

AveLA

N Mean

Std.

Deviation Std. Error

95% Confidence Interval for

Mean

Minimum Maximum Lower Bound Upper Bound

.00 34 2.7353 1.06767 .18310 2.3628 3.1078 1.00 4.50

1.00 126 2.9405 1.16894 .10414 2.7344 3.1466 1.00 5.00

2.00 35 3.1143 .68691 .11611 2.8783 3.3502 2.00 4.50

3.00 33 3.0000 1.06800 .18592 2.6213 3.3787 1.50 5.00

Total 228 2.9452 1.07753 .07136 2.8046 3.0858 1.00 5.00

ANOVA

AveLA

Sum of Squares df Mean Square F Sig.

Between Groups 2.601 3 .867 .744 .527

Within Groups 260.964 224 1.165

Total 263.565 227

factors influencing knowledgeable consumers' level of - USM

Documents

Transcript of factors influencing knowledgeable consumers' level of - USM