consumer satisfaction and repurchase intention from cross

CONSUMER SATISFACTION AND REPURCHASE INTENTION FROM

CROSS-BORDER E-COMMERCE: A TRUST-RISK-BASED STUDY

BY

MR. NATTHAKORN KHAYAIYAM

A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS

FOR THE DEGREE OF MASTER OF SCIENCE PROGRAM

(MANAGEMENT INFORMATION SYSTEMS)

MANAGEMENT INFORMATION SYSTEMS

FACULTY OF COMMERCE AND ACCOUNTANCY

THAMMASAT UNIVERSITY

ACADEMIC YEAR 2018

COPYRIGHT OF THAMMASAT UNIVERSITY

Ref. code: 25615802037480DCY

(1)

Thesis Title CONSUMER SATISFACTION AND REPURCHASE

INTENTION FROM CROSS-BORDER

E-COMMERCE: A TRUST-RISK-BASED STUDY

Author Mr. Natthakorn Khayaiyam

Degree Master of Science Program

(Management Information Systems)

Major Field/Faculty/University Management Information Systems

Commerce and Accountancy

Thammasat University

Thesis Advisor Professor Siriluck Rotchanakitumnuai, Ph.D.

Academic Years 2018

ABSTRACT

The waves of cross-border e-commerce (CBEC) is growing very fast resulting

big opportunities as well as threats to firms. Understanding consumer perception

towards cross-border e-commerce is crucial as it would result in consumer loyalty and

sustainable profit. This study is formulated based on a question that how trust and

perceived risk of the consumer influence repurchase from CBEC in both pre and post

purchase amongst Thai e-shoppers. The data was collected via a designated online

questionnaire distributed via social media and sharable link. Structural Equation Model

(SEM) and Confirmatory Factor Analysis was conducted to analyse the data together

with Paired T-test for analysing differences of trust and perceived risk in pre-to-post

purchase and domestic against cross-border e-commerce. The SEM result suggests that

post-purchase trust and perceived risk are highly influence to consumer satisfaction

and repurchase intention. Surprisingly, consumer tend to expose themselves to risk

before shopping; however, higher trust and lower risk are shaped firmly after they

experienced shopping from CBEC. Moreover, consumer do not perceived e-seller trust

and risk much difference between CBEC and domestic e-commerce. This study extends

the current Expectation-disconfirmation Theory and knowledge of trust and perceived

risk towards cross-border e-commerce. Furthermore, the implication of this study helps

Ref. code: 25615802037480DCY

(2)

firms to emphasize on the trust and perceived risk in CBEC challenge within Thailand

and for firm who would like to expand to cross-border business.

Keywords: e-commerce, cross-border e-commerce, trust, perceived risk, repurchase

intention, Expectation-disconfirmation theory.

Ref. code: 25615802037480DCY

(3)

ACKNOWLEDGEMENTS

I would like to express my deep gratitude to Professor Siriluck

Rotchanakitumnuai my research supervisors, for her patient guidance, enthusiastic

encouragement, and useful critiques of this work. I would like to thank Assistant

Professor Mathuspayas Thongmak, the chairman of my research, for her advice and

assistance in reviewing my research. Moreover, I would like to thank you Assistant

Professor Chatpong Tangmanee from Department of Statistics, Chulalongkorn

University, for his valuable guidance on statistical analysis and Structural Equation

Modelling used in this research. My grateful thanks are also extended to officers in

Master of Science Program in Management Information Systems, Thammasat Business

School, who always help and offer me the advices needed for research. I would like

to thank to Library of Thammasat University who provides necessary tools for thesis

development and

Finally, I wish to thank my parents and colleagues for their support and

encouragement throughout my study.

Mr. Natthakorn Khayaiyam

Ref. code: 25615802037480DCY

(4)

TABLE OF CONTENTS

Page

ABSTRACT (1)

ACKNOWLEDGEMENTS (3)

LIST OF TABLES (7)

LIST OF FIGURES (8)

LIST OF ABBREVIATIONS (9)

CHAPTER 1 INTRODUCTION 1

1.1 Background of e-commerce 1

1.1.1 Electronics Commerce and Cross-border Electronic 1

Commerce

1.1.2 Global E-commerce Economy 2

1.1.3 Global Cross-border E-commerce Economy 3

1.1.4 Southeast Asia and Thailand Cross-border E-commerce 3

Economy

1.2 Rational of the study 5

1.3 Research questions 6

1.3 Objectives 6

CHAPTER 2 REVIEW OF LITERATURE 8

2.1 Theory 8

2.1.1 Trust-based Consumer Decision Making Model 8

2.1.2 Expectation-Confirmation Theory 9

Ref. code: 25615802037480DCY

(5)

2.2 Related research 10

2.2.1 Trust 10

2.2.2 Perceived Risk 13

2.2.3 Interaction between Trust and Perceived Risk 16

2.2.4 Incorporating Trust in Expectation-disconfirmation theory 16

CHAPTER 3 RESEARCH METHODOLOGY 19

3.1 Conceptual model 19

3.2 Variable definition 20

3.3 Research hypotheses 21

3.4 Population and samples 25

3.5 Research Instruments and Data Collection 26

3.6 Data analysis Process 34

CHAPTER 4 RESULTS AND DISCUSSION 38

4.1 Descriptive statistics 38

4.1.1 Variables Descriptive Statistics and Normality of Residuals 38

4.1.2 Measurement Model Validity and Reliability 42

4.1.3 Samples demographics 45

4.2 Paired t-test for Mean Difference 49

4.3 Confirmatory Factor Analysis 52

4.4 Structural Equation Model 54

4.5 Hypothesis testing summary 57

CHAPTER 5 CONCLUSIONS AND RECOMMENDATIONS 61

5.1 Conclusion 61

5.2 Benefit 62

5.2.1 Theoretical benefits 63

Ref. code: 25615802037480DCY

(6)

5.2.2 Practical benefits 63

5.3 Research limitation 64

5.4 Recommendation for future research 64

REFERENCES 65

APPENDICES

APPENDIX A QUESTIONNAIRE 72

APPENDIX B CONFIRMATORY FACTOR ANALYSIS DETAIL 79

APPENDIX C MEASUREMENT MODEL RESIDUAL VARIANCE 82

APPENDIX D MEASUREMENT MODEL CORRELATION 84

BIOGRAPHY 86

Ref. code: 25615802037480DCY

(7)

LIST OF TABLES

Tables Page

3.1 Measurement items in pre-purchase and post-purchase stage 28

3.2 Model Validity and Reliability Analysis Criteria 35

3.3 SEM Recommended Fit Index and Criteria 36

3.4 SEM Fit Index Combination Criteria. 37

4.1 Factor Descriptive Statistics 38

4.2 Assessment of Multivariate Normality 41

4.3 Construct Summary 42

4.4 Cross-border E-commerce Variable Correlations for Discriminant

Validity

44

4.5 Measurement Model Validity and Reliability Analysis Result 45

4.6 Gender 46

4.7 Monthly Income 47

4.8 Highest Education Obtained 47

4.9 Occupation 48

4.10 Pair Summary for Paired T-test of Mean Difference 49

4.11 Paired Samples Test of Mean between Pre-purchase and

Post-purchase from Cross-border E-commerce and Domestic

E-commerce.

51

4.12 Confirmatory Factor Analysis Coefficients and Squared Multiple

Coefficients.

52

4.13 Structural Model Goodness of Fit Evaluation Summary. 54

4.14 Standardized Coefficients Effect in Structural Model. 56

4.15 Paired T-test Hypothesis Testing Result. 57

4.16 SEM Hypothesis Testing Result. 58

B.1 Confirmatory Factor Analysis Result. 79

C.1 Residuals Variance Summary. 82

D.1 Measurement Model Correlation Matrix. 84

Ref. code: 25615802037480DCY

(8)

LIST OF FIGURES

Figures Page

1.1 Online Retail Market Size Year-over-year. 2

1.2 Global E-Commerce Market Share. 3

1.3 Top 5 Shopping Application in Southeast Asia in 2017. 4

2.1 Trust-based Consumer Decision Making Model. 9

2.2 Expectation-Confirmation Decision Theory. 10

2.3 McKnight and Chervany ‘s Trust dimensions. 11

2.4 Featherman and Pavlou’s Risk Facets. 14

2.5 Lanktan et al.’s integrating technology Trust in EDT. 17

3.1 Proposed research model. 19

4.1 Histogram of sample age distribution. 46

4.2 Frequency of participants’ purchase via cross-border e-commerce. 48

4.3 Structural Model Standardised Coefficients. 55

B.1 Structural Model and CFA 79

Ref. code: 25615802037480DCY

(9)

LIST OF ABBREVIATIONS

Symbols/Abbreviations Terms

a

Cronbach’s alpha (alpha)

b Standardised regression coefficient

B Unstandardised regression coefficient df Degree of freedom c2 Chi’s Squared

R2 R-Sqaured SMC Squared Multiple Correlation (Multi-

variate R-sqaured)

Ref. code: 25615802037480DCY

1

CHAPTER 1

INTRODUCTION

1.1 Background of E-Commerce

1.1.1 Electronics Commerce and Cross-border Electronic Commerce

Electronic commerce (EC) is a form of exchange communication

through network technology. It includes data management, system and data security,

and system communication. This exchange is a form of paperless transaction for data,

product, or service exchange between parties. The term electronic commerce was

referring to electronic communication sending purchase order or business document

back in 1970s; however, the term was then developed into a form of commercial

exchange through websites since the development of world wide web as a new

business model.

Based on the relationship of the business to the parties, there are

four types of e-commerce principally – B2B, B2C, C2B, and C2C. Firstly, B2B or business-

to-business e-commerce is an e-commerce that both parties are businesses that

exchange data, services, or product between each other. It can be considered as a

form of a firm and its suppliers. Secondly, B2C or business-to-consumers e-commerce

refers to a form of electronic commerce that a business provides services or products

to end-user consumers per order. It is widely adopted and can been seen everywhere

nowadays. Thirdly, C2B consumer-to-business e-commerce is an exchange of data,

product, or services that created by the end-users offer to a firm. Lastly, C2C consumer-

to-consumer is an electronic business that the consumers exchange data, products, or

services which they produce them by themselves. The platform from maybe

constructed by a vendor to facilitate the transaction and it may be called a market

place.

The cross-border e-commerce (CBEC) is not a big different to the

aforementioned term of e-commerce. It expands the term to cover an aspect of

borderless commerce, thanks to the growing and expanding of internet access and

communication technologies in the last decades. The advanced technology in

Ref. code: 25615802037480DCY

2

communication and information exchange provide a solid foundation to the growth of

e-commerce firms such as Amazon, Alibaba, etc. to gain more investment power that

they can penetrate other market outside their home countries alone.

1.1.2 Global E-commerce Economy

Electronic commerce, or E-Commerce, is inarguably growing and

expanding throughout the world. Market value of E-Commerce triplicates from about

one million US dollars in 2012 to 3 million USD in 2017 globally. Electronic commerce

offers customer various benefit on shopping such as variety of product, fast searching,

decreasing asymmetric of product information, etc. (Amaral, 2017).

Figure 1.1 Online Retail Market Size Year-over-year (Euromonitor International, 2018)

There are various of players in global E-Commerce but major firms

which are Amazon.com, Alibaba Group, JD.com, and eBay take about a half of the

market share. (Euromonitor International, 2018) These companies have been

penetrated their domestic market and cross-border market.

Ref. code: 25615802037480DCY

3

Figure 1.2 Global E-Commerce Market Share (DHL Express, 2016)

1.1.3 Global Cross-border E-commerce Economy

International electronic trade or cross-border E-Commerce has been

enriched thank to technology advancement and connected world. It is today’s growth

rocker in E-Commerce with USD 300 million market size in 2017 and average growth

of 20 percent per year (DHL Express, 2016). Approximately 56 percent of cross- border

E-Commerce is sold via Alibaba, Amazon, and eBay whereas 34 percent of order made

in China (International Post Corporation, 2018). China apparently becomes a big market

for E-Commerce and Chinese firms are expanding their market to nearby region

especially South East Asia where they consider it as a potential market for E-

Commerce. The expansion can be notice obviously in recent years that major Chinese

firms bought quite a number of shares from current and new e-commerce platform

providers such as Lazada, Shopee, or even expanding their services to the region such

as Alibaba that starts a unified service for shopping experience from the purchasing to

payment gateway service.

1.1.4 Southeast Asia and Thailand Cross-border E-commerce

ASEAN, Association of South East Asian Nations, has announced that

they will boost digital economy throughout the region which will benefit cross-border

E-Commerce. The ministry of trade and industry of Singapore said that they would

Ref. code: 25615802037480DCY

4

support advanced payment and transaction to help small and medium firms to expand

their E-Commerce; however, there are challenges for less developed counties such as

Myanmar, Cambodia, and the Philippines on logistic and internet penetration

(Iwamoto, 2018). However, between the support of local government, E-Commerce

giants from around the world are already on the market such as e-Bay, Amazon, and

Alibaba Group; especially Chinese E-Commerce firms that are penetrating in South East

Asian such as Lazada, and Shopee (SEA Limited, 2017).

Figure 1.3 Top 5 Shopping Application in Southeast Asia in 2017 (SEA Limited, 2017)

Lazada is apparently the biggest online shopping platform that

operates in South East Asia providing over 2,500 brands and 100,000 sellers to serve

560 million customers in the Thailand, Vietnam, Indonesia, Malaysia, the Philippines,

and Singapore. In 2017, Alibaba acquired approximately 81 percent of its stake making

Alibaba to be the major stakeholder and key player in this region (SEA Limited, 2017).

ASEAN is seen as a potential market for many investors due to it low market saturation,

low competitors, and high adoption of information and communication technology.

Cross-border E-Commerce in Thailand has been growing in

accordance to the region trend. In Thailand, E-Commerce has been adopted for many

years with the linear increasing sales from 2,865 million USD in 2015 to 4,239 million

Ref. code: 25615802037480DCY

5

USD in 2017 (Priest, 2017). There are major players in Thailand market which are

Lazada, weloveshopping, Tarad, Zalora. Acquired by Alibaba, Lazada is closely

supported by its major stakeholder to advance its strategy towards “Thailand 4.0” and

remain its position as market leader in Thailand (SEA Limited, 2017). Although E-

Commerce competition in Thailand is not aggressive but there are opportunities and

high potential to grow. In 2017, Shopee, a part of SEA Group (former Garena), has

started their E-Commerce business in the region including Thailand which warming

competition in Thailand market (SEA Limited, 2017) while JD.com, Alibaba’s fierce

competitor, will join the market very soon after they have prepared logistics and

branding in the region (Priest, 2017).

In the customer perspective, the E-Commerce benefits their

shopping. Consumers across markets are motivated to shop cross-border for

fundamental reasons - product availability, a more attractive offering (including price),

and trust. Fashion and electronics are long-known cross-border top sellers, but

consumers now crave more on other product categories such as beauty and cosmetics,

pet care, food and beverage, and sporting goods (DHL Express, 2016).

1.2 Rational of the study

Cross-border e-commerce becomes a new economy of global trade due

to supporting factors on information technology and communication technology. The

advancement of the technologies provides a huge opportunity for large capital firms

to expand their business to go cross border to make more competitiveness in the

trading business; however, cross-border e-commerce has received relatively less

research attention than in domestic e-commerce context (Xiao, Wang, & Liu, 2018)

while the business has been changing very fast. Exploring the theory and factors that

are applicable to cross-border e-commerce context is necessary. Nevertheless, the

research on the e-commerce settings has been widely study and well settled where

trust and risk play an important role in the adoption of e-service. There are many

theories and factors influencing adoption of e-commerce on both pre-purchase and

post purchase stage in which this study aims to explore them under cross-border

Ref. code: 25615802037480DCY

6

e-commerce context to expand current knowledge on e-commerce exposing adoption

factor to better business decision on their investment. Studying on Thailand’s

population is also an interesting aspect due to a high growth in both digital commerce

adopters and Thailand is part of South East Asia region which is a fast-growing region

in e-commerce market (Priest, 2017). Thais are also a well digital adopters with

supportive infrastructure in term of internet speed, devices, and purchasing power.

1.3 Research questions

1.3.1 Do consumer trust expectation and risk expectation influence

consumer disconfirmation and later trust and risk perception towards cross-border e-

commerce or not?

1.3.2 Do disconfirmation of consumer trust and risk expectation, consumer

satisfaction, post-purchase trust and risk perception influence repurchase intention

from cross-border e-commerce or not?

1.3.3 Does consumer trust affects perceived risk at the pre-purchase stage

as well as post-purchase stage or not?

1.3.4 Do consumers expect trust and risk from cross-border e-commerce

sellers and domestic e-commerce sellers differently across the pre-purchase and post-

purchase stage or not?

1.4 Objective

1.4.1 To study consumer trust expectation and risk expectation that

influence consumer disconfirmation and later trust and risk perception towards cross-

border e-commerce.

1.4.2 To study disconfirmation of consumer trust and risk expectation,

consumer satisfaction, post-purchase trust and risk perception that influence

repurchase intention from cross-border e-commerce.

1.4.3 To study consumer trust affecting perceived risk at the pre-purchase

stage as well as post-purchase stage.

Ref. code: 25615802037480DCY

7

1.4.4 To study consumers difference perception of trust and risk from

cross-border e-commerce sellers and domestic e-commerce sellers across the pre-

purchase and post-purchase stage.

Ref. code: 25615802037480DCY

8

CHAPTER 2

REVIEW OF LITERATURE

2.1 Theory

In this study, we explore previous literatures regarding the consumer

decision making and intention to purchase product. There are theory and model that

are interesting in the context of cross-border e-commerce which are Trust-based

Consumer Decision Making Model (TCDM) and the Expectation-Confirmation Theory

(ECT). The TCDM model features trust and risk as the main construct that influence

intention to buy or adopt services while the ECT evaluates repurchase intention of the

consumers in collaboration between confirmation of pre-purchase expectation and

satisfaction after purchase.

2.1.1 Trust-based Consumer Decision Making Model

The Trust-based consumer decision making theory propose three

main factors that influence purchase intention which are Trust, Perceived Risk, and

Perceived Benefits. The model was developed from the previous proposed Valence

Framework resulting in the three aforementioned independent variables (Kim, Ferrin,

& Rao, 2008; Naovarat, 2015). The model diagram showing independent variables and

dependent variables is shown in the Figure 2.1.

Trust in the model refers to the consumer trust on the E-Commerce

platform they are using with the risk and uncertainty for making purchase. If the

consumer has more trust in the platform, they will feel more confidence. Risk refers

to the customer perception that online purchasing has some risks or uncertainties that

they need to accept when making a purchase. There are risks of monetary loss,

shipment loss, or unable to deliver in the field of e-commerce that are significant

factors for customer’s purchase decision making. Lastly, Benefit or perceived benefit

refers to consumers perception towards E-Commerce benefiting them in form of time

efficacy e.g. fast navigation comparing to physical store shopping and the convenience

in the shopping experience. The benefits may be preceded by a cognitive stage leading

Ref. code: 25615802037480DCY

9

an overall perception of benefit in form of monetary savvy, time savvy, and effort

savvy. These three constructs have effect over customer intention to making a

purchase on online scenario.

Figure 2.1 Trust-based Consumer Decision Making Model (Kim et al., 2008)

2.1.2 Expectation-Confirmation Theory

The expectation-confirmation theory has been widely adopted and

many fields of study, especially marketing field. The theory extensively studies the

post-purchase experience involving customer satisfaction and continuance of using

services or repurchase based on their satisfaction of prior purchase, usage of services,

or adoption. The theory proposes that the expectation which is a pre-purchase stage

affects the confirmation or disconfirmation of the belief which can be resulted in actual

purchase or adoption (Oliver, 1980). Satisfaction in the theory refers to the customer

overall satisfaction after adoption, purchase, or usage of product or services.

Although ECT has been widely adopted in marketing field; however,

many researchers discuss that the theory ignores viability of the post-experience

expectation i.e. the customer changes their perception after period of time after

making purchase from negative to positive or vice versa due to intermediate effect of

environment possibly social influence and marketing campaign (Bhattacherjee, 2001).

Ref. code: 25615802037480DCY

10

Also, the pre- purchase stage influenced by the external factors and the post-purchase

influenced by the consumers perception themselves are also discussed.

Figure 2.2 Expectation-Confirmation Decision Theory (Oliver, 1980).

2.2 Related research

We have reviewed the previous literatures in the context of e-commerce

and cross-border e-commerce adoption. There are constructs that are related to the

ECT and TCDM model; however, the literatures expand more knowledge in each

variable in both ECT and TCDM model such as the multi-dimension of trust and risk

which is intensively studied.

2.2.1 Trust

The definition of trust is complicate due to its abstract and complex

factor nature. Trust has been addressed in many previous researches with vast

definition and stages (Bonsón Ponte, Carvajal-Trujillo, & Escobar-Rodríguez, 2015).

Several theories regarding trust are diverse into many stages of the interaction between

trustee and trustor (Stouthuysen, Teunis, Reusen, & Slabbinck, 2018). Trust has been

studied in many fields, yet the diverse conceptualisation of trust is ununified. Fisher

and Zoe refers to McKnight et al. previous research that an important consideration in

exchange relationships is determining the parties with whom an individual is willing to

interact. Trust is of central importance in this decision (Fisher & Zoe Chu, 2009). Trust

reflects the willingness of a party to be vulnerable to the actions of another party

Ref. code: 25615802037480DCY

11

based on positive expectations regarding the other party’s motivation and/or

behaviour so that in the field of electronic commerce, trust formulate a belief of the

consumer on the seller to the extent of positive belief. Due to the nature of the

electronic commerce, consumers experience uncertainty of the online purchase;

therefore, trust plays an important role as a solution to uncertainty/risk (Kim et al.,

2008). Similarly, the model of dyadic trust in organizational relationships proposed by

Mayer et al. states that if the trustor perceives a trustee’s ability, benevolence, and

integrity to be sufficient, the trustor will develop trust (an intention to accept

vulnerability) toward the trustee; besides, trust can alleviate the effect of risk creating

a favourable condition for the trustee to accept the vulnerabilities.

Figure 2.3 McKnight and Chervany’s Trust dimensions. (McKnight, Choudhury, & Kacmar,

2002)

Nevertheless, in the context of B2C e-commerce, a consumer and

an online vendor would only be fully familiar after the consumer transacts with the

vendor and analyses the results. On McKnight and Chervany’s previous research in the

Ref. code: 25615802037480DCY

12

area of electronic commerce proposes a decomposition and conceptualisation of trust

and the antecedents of trust that trust constructors can be differentiated into two

dimensions which are Institutional trust and dispositional trust (Harrison McKnight &

Chervany, 2001).

2.2.1.1 Dispositional Trust

Disposition to trust means the extent to which one displays

a consistent tendency to be willing to depend on others in general across a broad

spectrum of situations and persons (McKnight et al., 2002). There are two sub-

constructs in the dispositional trust which are Faith in humanity and Trusting Stance.

Faith in humanity means that one assumes others are usually competent, benevolent,

honest/ethical, and predictable. Trusting stance means that, regardless of what one

assumes about other people generally, one assumes that one will achieve better

outcomes by dealing with people as though they were well-meaning and reliable.

Therefore, trusting stance is like a personal choice or strategy to trust others. Because

it involves a choice that is presumably based on a subjective calculation of the odds

of success in a venture, trusting stance derives from the calculative, economics-based

trust research stream.

2.2.1.2 Institutional-based Trust

Institutional trust refers to an individual’s beliefs about the

structural safety or favourability of the conditions. Institutional trust ‘‘generalizes

beyond a given transaction and beyond specific sets of exchange partners”.

“Favourable conditions” refers to the legal, regulatory, business, and technical

environment perceived to support success. Structural assurance means that one

believes that protective structures - guarantees, contracts, regulations, promises, legal

recourse, processes, or procedures are in place that are conducive to situational

success. These two characteristic groups of trust formulate the trusting belief and

trusting intention which is the belief act on their benefits. The trusting belief according

to (Harrison McKnight & Chervany, 2001), can be classified into four categories -

Competence, Benevolence, Integrity, and Predictability belief.

Ref. code: 25615802037480DCY

13

(1) Competence

Competence is connected to objective perception (e.g.

reliability, technical capabilities, know-how, and skills) of the seller/vender in the

context of e-commerce. Thus, customer review or feedback from the past experience

informs first-time online shopper the reliability and ability of the online vendor.

(2) Benevolence

Benevolence refers to the belief that the opposite side of the

communication cares about and act in one’s interest. Benevolence reflects the specific

relationship between trustor and trustee. It is not the overall trustee kindness in

general, but it is a subjective belief towards individual relationship between trustor

and trustee.

(3) Integrity

Integrity means that one believes that the other party makes

good-faith agreements, tells the truth, acts ethically, and fulfils promises. This would

reflect the belief that the Internet vendor will come through on its promises and

ethical obligations, such as to deliver goods or services or to keep private information

secure. Thus, integrity is more about the character of the trustee in the same direction

towards the trustor than about the trustor-trustee relationship.

(4) Predictability

Predictability means that one believes the other party’s

actions (good or bad) are consistent enough that one can forecast them in a given

situation. Those with high trusting belief-predictability would believe that they can

predict the Internet vendor’s future behaviour in a given situation.

2.2.2 Perceived Risk

Risk is one of the most important factors to study in cross-border e-

commerce due to its uncertainty in transaction and experience. Cross-border online

shopping is an unfamiliar and uncertain activity for consumers, more so than domestic

electronic commerce (Lesma & Okada). Risk is a situation that that may be resulted in

uncertainty or negative consequence (Naovarat, 2015). Many researches in the past

have studied the relationship between Trust and Perceived Risk. Perceived Risk has a

negative impact on customer trust in online transaction. It is an important factor in

Ref. code: 25615802037480DCY

14

online purchasing scenario. The consumer’s will reluctantly be making decision to

purchase comparing to traditional channel of purchase due to lack of touch and feel

or product trial. Risk may be classified into various type such as Jacoby and Kaplan’s

seven type of risk which are financial, performance, physical, psychological, time,

social, and opportunity cost risk (Kim et al., 2008). Likewise, risk can be grouped into

System-dependent uncertainty and Transaction-specific uncertainty according to study

from (Rouibah, Lowry, & Hwang, 2016).

Figure 2.4 Featherman and Pavlou’s Risk Facets (Featherman & Pavlou, 2003)

There are many previous research areas that intensively study on the

perceived risk as well as consumer trust in different context e.g. information system,

marketing, and psychology. In the previous research from Featherman and Pavlou on

the consumer risk on an aspect of consumer behaviour, they intensively study the

various aspect of risk and classify them into 7 facets of risk which are performance risk,

financial risk, time risk, psychological risk, social risk, privacy risk, and overall risk

(Featherman & Pavlou, 2003).

Ref. code: 25615802037480DCY

15

2.2.2.1 Performance Risk

Performance Risk is the possibility of the product

malfunctioning and not performing as it was designed and advertised and therefore

failing to deliver the desired benefits. In the area of e-commerce, this performance risk

is quite important for the representation of the product is on the website only leading

lack of touch and feel of the product compare to traditional channel of purchase.

2.2.2.2 Financial Risk

Financial risk is the potential monetary outlay associated with

the initial purchase price as well as the subsequent maintenance cost of the product.

In the context of e-commerce, it is the risk that the transaction payment may fail

leading to losses of money. This scenario is different to traditional channel transaction

that the consumer pays directly to the vendor and receive the product without delay.

2.2.2.3 Time Risk

Time risk refers to the consumers perception that they are

wasting time for doing the shopping e.g. searching for website information or products

information and product search before making a purchase. It may occur at the pre-

purchase navigation and at the purchase stage when making a wrong decision or

transaction on the website. The consumers may feel that it is wasting of their time

refunding, returning, or petitioning to the vender to change or claim if the product or

service will not fit to their need. This is the projection of uncertainties on the after-

sales services.

2.2.2.4 Psychological Risk

Psychological risk refers to the self-reflection uncertainties

which introduce projected loss of self-esteem. This facet of risk cases the consumers

to potentially perceive that the e-service or product does not reflect their self-images

or identity and/or not achieving their purchase goal from e-service is frustrating and

deteriorate their self-respect or appreciation towards themselves.

2.2.2.5 Social Risk

Social risk is an uncertainty that the consumer perceived from

external agents that using e-service will lead to social perception changes towards

them such as negative feeling from the friends or family, getting lost of trend, lower

Ref. code: 25615802037480DCY

16

respect from the others, or potential perception as a foolish from others adopting or

use the e-service.

2.2.2.6 Privacy Risk

Privacy risk is a concern of potential losses of personal

information and transaction information without notice and permission granted

occurring during the usage of e-service. The riskiest potential loss from the privacy risk

is the fraudulent transaction by digital thief inside the same e-service platform and

outside the platform.

2.2.2.7 Overall Risk

Overall risk is a perception of overall aspect of risk in general

that the customer perceived from the e-service usage. The risk assesses as the overall

feeling of the consumers regardless of specific aspect of risk.

2.2.3 Trust and Perceived Risk Interaction

Incorporating trust and risk in e-commerce context has been

observed in many previous papers; however, the direction of effect is quite

controversial whether trust affects perceived risk, perceived risk affects trust, or they

both affect each other at the same time.

Pavlou’s study on Trust and Risk interaction towards Intention

suggest that Trust affects Risk (P. A. Pavlou, 2003) based on TAM (Technology

Acceptance Model) in accordance to the study by Kim et. al. which conduct

longitudinal study based on EDT suggests the same effect direction (Kim, Ferrin, & Rao,

2009) as well as the work done by Zhu et. al. in later year (Zhu, Neal, Lee, & Chen,

2009). These studies are significant statistically affirming the researcher’s belief that

Trust affects Risk. However, there are recent studies that suggest that Perceived Risk

affect Consumer Trust. These studies are conducted in marketing perspective towards

online purchase which the researchers consider Risk as an independent variable that

influencing consumer trust (Pappas, 2016; Rouibah et al., 2016).

2.2.4 Incorporating Trust and Perceived Risk in Expectation-

Disconfirmation Theory

In previous study on the usage of EDT with technology adoption,

there are several models proposed by the researchers to incorporate the trust into

Ref. code: 25615802037480DCY

17

EDT model. Trust theory in technology and e-commerce setting has been intensive

studied by McKnight. In their paper regard the trust and technology adoption and

continuance intention, they proposed the model in Figure 2.5 showing relationships of

trust expectation, trust disconfirmation, satisfaction, perceived performance, and

trusting intention in the aftersales stage to the technology continuance as the outcome

(N. Lankton, McKnight, & Thatcher, 2014).

Figure 2.5 Lanktan et al.’s integrating technology Trust in EDT (N. Lankton et al., 2014)

Based on Lanktan’s study on trusting belief as an expectation in EDT

model, it is interesting that the proposed Technology Trusting Expectation does not

affect the Technology Trusting Disconfirmation, but the construct positively affects the

Technology Trusting Performance which is a post-evaluation belief after they

experience the technology.

Another study was conducted to observe the influence on Trust and

Risk in post-purchase experience by Mou et. al. incorporating the aforementioned

constructs in Expectation Disconfirmation Theory. Although Trust and Risk has various

antecedent to examine, post-adoption/behavioural action’s perception of Trust and

Risk are not very influential to customer intention to repurchase or continuance

intention even though the previous study shows strong relationship of Trust to

intention to use in TAM model (Gefen, Karahanna, & Straub, 2003). This conflict of

relationship is due to strong and dominant effect of customer satisfaction (Mou, Shin,

Ref. code: 25615802037480DCY

18

& Cohen, 2015) in the Expectation Disconfirmation Theory while TAM does not

incorporating Satisfaction into the Theory.

Ref. code: 25615802037480DCY

19

CHAPTER 3

RESEARCH METHODOLOGY

3.1 Conceptual Model

Based on Trust-based Consumer’s Decision-making Model (TCDM) and the

Expectation-confirmation Theory (ECT), we adapt the ECT and Trust-based Consumer’s

Decision-making Model and propose the following conceptual model as shown in the

Figure 3.1 to study on the cross-border e-commerce pre and post purchase stage

leading to consumer repurchase intention.

Figure 3.1 Proposed research model

The model incorporates Trust and Risk as consumer expectation (Trusting

Expectation and Risk Expectation). The trust and risk are evaluated again at the post-

purchase stage as a combination with satisfaction and disconfirmation to have direct

effect on consumer repurchase intention.

Ref. code: 25615802037480DCY

20

3.2 Variable Definition

3.2.1 Trusting Expectation and Trusting Performance

Trust in this study refers to the customer belief towards cross-border

e-commerce and domestic e-commerce. The measurements aim to evaluate

trustworthiness of e-sellers based on consumer’s trusting belief in previous studies on

trust towards information systems (Gefen et al., 2003; McKnight, Carter, Thatcher, &

Clay, 2011; McKnight & Chervany, 2014) and trust in electronic commerce (Bonsón

Ponte et al., 2015; Kim, 2014; Kim et al., 2008; McKnight et al., 2002; Mukherjee, Arnott,

& Nath, 2007; Sfenrianto, Wijaya, & Wang, 2018; Stouthuysen et al., 2018) where

measurements are evaluating interpersonal trusting belief on overall trustworthiness

of the e-sellers, act on consumer benefit (Integrity), willingness to support

(Benevolence), and information quality (Competence) provided by e-sellers.

3.2.2 Perceived Risk Expectation and Perceived Risk Performance

In this study, Perceived Risk in 4 different facets which are financial

risk, delivery time risk, privacy risk, and overall risk are measured in accordance to

previous studies on risk dimensions towards information system and e-commerce area

(Featherman & Pavlou, 2003; Mou, Cohen, Dou, & Zhang, 2017; Mou et al., 2015;

Pappas, 2016; Rouibah et al., 2016). The financial risk is the risk of possibility to lose

money during transaction while the delivery risk concerns timeliness of product deliver,

and the privacy risk is the risk that their personal information will be exposed due to

using the website/application. These risks show high importance in previous studies,

but they were measured in e-commerce context and cross-border e-commerce in e-

sellers’ perspective (Guo, Bao, Stuart, & Le-Nguyen, 2018). This study will comparatively

measure the four risks in cross-border and domestic context as well as pre-purchase

as an expectation (Perceived Risk Expectation) and post-purchase setting (Perceived

Risk Performance) as a performance perception according to Expectation-

disconfirmation Theory.

3.2.3 Expectation Disconfirmation

Expectation Disconfirmation in this study refers to the confirmation

of overall expectation of past experience (Oliver, 1980), confirmation of trusting belief

Ref. code: 25615802037480DCY

21

(N. Lankton et al., 2014), and this study propose of disconfirmation of perceived risk

expectation from purchasing via cross-border e-commerce. The measurements are

based on Expectation Disconfirmation Theory within the context of e-commerce (Kim

et al., 2008, 2009), but they are adapted to cross-border e-commerce in this study.

3.2.4 Satisfaction

Satisfaction has been studied for many years. In the previous

literatures in marketing and psychology researchers defines the satisfaction differently.

In this study, we define consumer’s satisfaction based on the Expectation-Confirmation

Theory (Bhattacherjee, 2001; Fang, George, Shao, & Wen, 2016; Kim et al., 2008; Mou

et al., 2017) that the term refers to the customer feeling of contented or pleased after

experienced product and services from cross-border e-commerce.

3.2.5 Repurchase Intention

The Repurchase Intention in this study is a tendency that the

consumer is likely to repurchase from cross-border e-commerce again. This construct

is found in various studies on e-service or information system continuance and loyalty

(Chong, 2015; Mou et al., 2017) as well as repurchase intention in accordance to the

Expectation-Confirmation Theory (Ambalov, 2018; Bhattacherjee, 2001; Kim et al., 2008;

Shang & Wu, 2017).

3.3 Research hypothesis

3.3.1 Relationship of Trusting Expectation and Perceived Risk

Expectation

Based on previous study on trust and risk interaction, it is very

controversial on examination of their relationship; however, the meta-analysis on this

relationship suggest that Trust negatively affecting Risk has a stronger effect than the

other paths (Mou et al., 2015). We derived this ground analysis to propose our

hypothesis as the following.

H1 Trusting Expectation from cross-border e-commerce has negative

effect on Perceived Risk Expectation from cross-border e-commerce.

Ref. code: 25615802037480DCY

22

3.3.2 Relationship of Trusting Expectation, Perceived Risk Expectation,

Expectation Disconfirmation.

Trust Expectation and Perceived Risk Expectation are considered

different perspective of consumer expectation that influenced by disconfirmation of

the consumer belief based on EDT model and the previous study on mobile

commerce (Chong, 2015). They both affect the consumer intention to purchase in the

context of Theory of Planned Behaviour as well (Park, Lee, & Ahn, 2014) Based on

previous study of Trust and Risk in the EDT model (Chong, 2015; Mou et al., 2015), the

path of the performance stage is not significantly influencing the consumer intention

due to stronger influence of customer satisfaction, but the study on Trusting Belief

towards Expectation-Disconfirmation in EDT shows a significant relationship. Therefore,

we propose the hypothesis of their relationship in accordance to Expectation-

disconfirmation theory as the following.

H2a Trusting Expectation from cross-border e-commerce has positive

effect on Expectation Disconfirmation from cross-border e-commerce.

H2b Perceived Risk Expectation from cross-border e-commerce has

negative effect on Expectation Confirmation from cross-border e-commerce.

H2c Trusting Expectation from cross-border e-commerce is higher

than domestic e-commerce on average.

H2d Perceived Risk Expectation from cross-border e-commerce is

less than domestic e-commerce on average.

3.3.3 Relationship of Trusting Performance, Perceived Risk

Performance, and Expectation Disconfirmation.

The performance stage of the consumers’ perception of Risk and

Trust are formed at the post-purchase stage. Based on the study on EDT and Trust,

the path of Trusting Performance influences the Disconfirmation of Expectation in the

post-purchase stage (Ambalov, 2018; Kim et al., 2009; N. Lankton et al., 2014). This

study includes the Perceived Risk Performance into the EDT model as well in order to

observe both Trust and Risk in the post-purchase stage. Therefore, the hypothesis

regarding this cross-border e-commerce constructs’ interaction are proposed as the

following.

Ref. code: 25615802037480DCY

23

H3a Trusting Performance from cross-border e-commerce has


H3b Perceived Risk Performance from cross-border e-commerce has


H3c Trusting Performance from cross-border e-commerce is higher

than domestic e-commerce on average.

H3d Perceived Risk Performance from cross-border e-commerce is

less than domestic e-commerce on average.

3.3.4 Relationship of Trusting Performance and Perceived Risk

Performance from cross-border e-commerce.

Based on the meta-analysis on this relationship, this study follow the

concept that Trust negatively affecting Risk which has a stronger effect (Mou et al.,

2015). The Trusting Performance and Perceived Risk Performance are the post-

purchase perception which share the same properties to the pre-purchase perception

except that this is an updated belief after the consumer experienced e-service in

accordance to Expectation Disconfirmation context (N. Lankton et al., 2014; P. A.

Pavlou, 2003). Therefore, this study proposes the hypothesis over their relationship as

the following.

H6 Trusting Performance has negative effect on Perceived Risk

Performance

3.3.5 Relationship of Expectation Disconfirmation and Satisfaction

from cross-border e-commerce.

The Expectation-disconfirmation Theory suggest that the Expectation

Disconfirmation/Confirmation has positive effect on consumer satisfaction (Oliver,

1980) after they have made a purchase. Previous study on Expectation-disconfirmation

Theory and consumer satisfaction also confirm that the relationship between these

latent variables is affirmative (Ambalov, 2018; Bhattacherjee, 2001; N. Lankton et al.,

2014; N. K. Lankton, McKnight, Wright, & Thatcher, 2016).

H7 Expectation Disconfirmation has positive effect on Satisfaction.

Ref. code: 25615802037480DCY

24

3.3.6 Relationship between Pre-purchase and Post-Purchase of

Consumer Trust and Perceived Risk from cross-border e-commerce.

In this study, we believe that the perception of pre-purchase risk and

pot-purchase trust are different. The customers have an updated belief of trust

towards the seller after they experienced the purchase via cross-border e-commerce

in accordance to previous study of Trust in EDT context (Ambalov, 2018; Bhattacherjee,

2001; N. Lankton et al., 2014; N. K. Lankton et al., 2016; Shang & Wu, 2017). The

proposed Perceived Risk Expectation and Perceived Risk Performance share the same

concept of Trust that the study can observe consumer perception of risk on pre-to-

post purchase as well. Therefore, this study proposes the following hypothesis.

H4a Trusting Expectation (pre-purchase) has positive influence

towards Trusting Performance (post-purchase).

H4b Trusting Performance (post-purchase) is increased from Trusting

Expectation (pre-purchase).

H5a Perceived Risk Expectation (pre-purchase) has positive influence

towards Perceived Risk Performance (post-purchase).

H5b Perceived Risk Performance (post-purchase) is decreased from

Perceived Risk Expectation (pre-purchase).

3.3.7 Satisfaction and Repurchase Intention from cross-border e-

commerce.

Customer satisfaction influence Intention to repurchase or

continuance usage of e-services in various study on their relationships (Lee, Park, &

Han, 2011; Marinkovic & Kalinic, 2017; Pham & Ahammad, 2017; Sfenrianto et al., 2018;

Shang & Wu, 2017). In this study, the observation of their relationships is based

principally on the Expectation-Disconfirmation Theory under e-service context (Bonsón

Ponte et al., 2015; Kim et al., 2008, 2009; Zhu et al., 2009) which define satisfaction as

a consumer overall satisfactorily experience after usage of e-service and propose the

following hypothesis.

H8 Satisfaction has positive effect on Repurchase Intention.

Ref. code: 25615802037480DCY

25

3.4 Population and samples

3.4.1 Population

Based on Statista, 12.1 million Thailand’s online shopper is estimated

in 2017 and the number of digital shoppers is estimated to 13.9 million in 2021. These

12.1 million e-commerce shoppers can be estimated as a population of total online

shoppers including both online shopper who experience and not experience cross-

border e-commerce; however, no evidence from any sources that classify number

cross-border shoppers is addressed; therefore, the population for our study is unknown

within the 12.1 million users.

3.4.2 Samples

Sample size for analysis is calculated based on Cohen’s guideline for

power analysis (Cohen, 1988). This study is a new context of e-commerce therefore

the effect size is assumed to be trivial (f2 = 0.05 i.e. at least 5% variance explained in

the multiple linear regression model). Moreover, this study sets up 0.95 confidence

interval with 95 power of test (1 – Type II error probability = 0.95) based on Cohen’s

formula and guideline.

! =$(1 − ())

()

Using R’s pwr package, the package calculates the result with

aforementioned parameters to retrieve 416 minimum sample size for the regression

analysis (u = 6, f 2 = 0.05, Power Test (b) = 0.95, Significant Level (a) =0.05).

In term of Student’s T-test, Cohen’s guideline for small effect size is

used for two-tailed and two-sided hypothesis testing (Effect Size(d) = 0.2, Power Test

(b) = 0.95, Significant Level (a) = 0.05, Type = Two-tailed) in which the result in 326

minimum sample size which is the same to power table provided from Cohen’s

guideline.

Due to the two-methods analysis used in this study – regression and

t-test analysis, sample size to be used for the full model will be based on minimum

sample size for regression analysis which is 416 samples.

Ref. code: 25615802037480DCY

26

3.5 Research Instruments and Data Collection

This study is a quantitative exploratory research aiming to quantify factors

influencing repurchase intention and difference perception of Trust and Risk between

pre-purchase and post-purchase across cross-border and domestic purchase;

therefore, the questionnaire is used with Likert-scale questions to ask each respondent

on the main continuance model and the attitude change model on Trust and Risk.

The pre-purchase constructs consist of Perceived Risk Expectation and

Trusting Expectation. These two variables are measured two times, domestic context

and cross-border context, on the same measurement item. The post-purchase Trusting

Performance and Risk Performance are measured on the same setting, then, they are

compared on consumer Trust and Risk perception between domestic e-commerce and

cross-border e-commerce. Cross-border pre-purchase and post-purchase perception of

Trust and Risk are evaluated as well in order to observed how the consumer

perception changes; however, the study of Expectation on pre-purchase stage is based

on the recall of previous experience.

3.5.1 Research Instrument Development

The questionnaire is developed to collect the sample’s response

regarding their perception on cross-border e-commerce purchase in electronic form

via SurveyMonkey. The questionnaire consists of 3 parts as described below.

Part 1 provides introductory information survey purpose, e-

commerce and cross-border e-commerce purchase, and direction for the respondents

to understand the subject and scope of the survey they are answering. This part also

asks the respondent whether they have experienced cross-border e-commerce

purchase or not in order to screening out ineligible sample from the analysis.

Part 2 consists of interval measurement items in Likert-scale to

observe sample attitudes towards pre-purchase and post-purchase experience based

on Trust, Risk, and Expectation-disconfirmation Theory. Each item requires the

respondent to provide opinion in scale of 1 to 5 which 1 is Strongly Disagree, 2 is Agree,

Ref. code: 25615802037480DCY

27

3 is Neutral, 4 is Agree, and 5 is Strongly Agree. The measurement items can be referred

on Table 3.1.

Part 3 contains items to collect respondent demographic information

which are gender, highest education level, monthly income, and frequency of online

purchase.

This study conducts a pre-test with 30 sample surveys once before

the actual data collection. The pre-test samples are analysed to ensure reliability of

the research tools using Cronbach’s Alpha as measurement (a > 0.8); factor analysis

is conducted (factor loading > 0.7) as well to ensure factor correlation and grouping

within each construct are acceptable statistically.

3.5.2 Data Collection

The questionnaire was transformed into online form on

SurveyMonkey.com, and the questionnaire was distributed online to participants. The

participant is voluntarily to answer the questions without any interference from the

researcher.

Ref. code: 25615802037480DCY

Table 3.1

Measurement items in pre-purchase and post-purchase stage. Construct Factor Abbr. Original Measurement Item Adapted Measurement Item Adapted from

Trusting Expectation/

Trusting Performance

(Cross-border and

Domestic e-commerce)

CBTRE1, CBTRP1

DMTRE1, DMTRP1

This site is trustworthy. This seller is trust worthy. (Kim et al., 2008)

(Gefen et al., 2003)

(P. A. Pavlou, 2003)

This Web retailer is trustworthy.

CBTRE2, CBTRP2

DMTRE2, DMTRP2

If I required help, LegalAdvice.com

would do its best to help me.

The seller would be willing to

help when there is question or

problem.

(McKnight et al., 2002)

CBTRE3, CBTRP3

DMTRE3, DMTRP3

This Web retailer is known as one

that keeps promises and

commitments.

The seller would keep promises

and commitments for their

services.


(Pavlou & Fygenson, 2006)

CBTRE4, CBTRP4

DMTRE4, DMTRP4

This Web vendor would be honest

in providing accurate information

about this product.

would provide product and

services information correctly.


(Pavlou & Fygenson, 2006)

28

Ref. code: 25615802037480DCY

Table 3.1

Measurement items in pre-purchase and post-purchase stage. (Cont’d) Construct Factor Abbr. Original Measurement Item Adapted Measurement Item Adapted from

Perceived Risk Expectation/

Perceived Risk Performance

(Cross-border and


CBPRE1, CBPRP1

DMPRE1, DMPRP1

Purchasing from this Website

would involve more product risk

(i.e. not working, defective

product) when compared with

more traditional ways of shopping.

Purchasing from this Website

would involve more product risk

(i.e. not working, defective

product).

(Kim et al., 2008)

CBPRE2, CBPRP2

DMPRE2, DMPRP2

I would be concerned that I may

suffer from monetary loss due to

the seller’s fraudulent acts.

Purchasing from this seller

would involve more financial risk

(i.e. fraud, hard to return).

(Hong, 2015)

(Kim et al., 2008)

(Featherman & Pavlou,

2003) Purchasing from this Website

would involve more financial risk

(i.e. fraud, hard to return) when

compared with more traditional

ways of shopping.

29

Ref. code: 25615802037480DCY

Table 3.1


Perceived Risk Expectation/

Perceived Risk Performance

(Cross-border and


CBPRE3, CBPRP3

DMPRE3, DMPRP3

Purchasing from this seller

would involve uncertainty on

product delivery.

New item

CBPRE4, CBPRP4

DMPRE4, DMPRP4

How would you rate your overall

perception of risk from the seller?

Overall, purchasing from this

seller would be risky.

(Kim et al., 2008)


2003) On the whole, considering all sorts

of factors combined, about how

risky would you say it would be to

sign up for and use XXXX?

30

Ref. code: 25615802037480DCY

Table 3.1


Expectation

Disconfirmation

(Cross-border e-commerce)

DCE1 Overall, most of my expectations

from using this website were

confirmed.

Overall, most of my

expectations were confirmed.

(Mou et al., 2017)

DCE2 My experience with using this

website was better than what I had

expected.

My experience with purchasing

from this seller was better than

what I had expected.

(Mou et al., 2017)

DCE3 Overall, purchasing from this

seller was not risky as expected.

New item

DCE4 My experience with using this

website was better than what I had

expected.

Overall, purchasing from this

seller is trust worthy as

expected.

(Kim et al., 2008)

(Pham & Ahammad, 2017)

31

Ref. code: 25615802037480DCY

Table 3.1


Satisfaction

(Cross-border

e-commerce)

SAT1 I am satisfied with the purchase experience from

this website (e.g., ordering, payment procedure).

I am satisfied from purchasing

from this seller.


(Sfenrianto et al., 2018)

Overall, I am quite satisfied with my experience

dealing with e-sellers.

SAT2 My experience with using this website was better

than what I had expected.

Overall, I am quite satisfied

with my experience dealing

with e-sellers.

(Kim et al., 2008)

(Featherman & Pavlou, 2003)

SAT3 I have good impression with the service provided by

e- sellers.

I have good impression with

the service provided by e-

sellers.


(Kim et al., 2009)

(Featherman & Pavlou, 2003) How do you feel about your overall experience of

the purchase through this Website?

I am satisfied with my overall experiences of online

shopping at this website.

32

Ref. code: 25615802037480DCY

Table 3.1


Satisfaction


SAT4 My experience with using this website

was better than what I had expected.

Overall, how would you rate

your experience purchasing from

this seller?

(Kim et al., 2008)


2003)

Repurchase Intention


CON1 I intend to continue using Microsoft

Access to create databases.

I intended to continue

purchasing from cross-border

seller again in the future.


2003)

CON2 I plan to continue using Microsoft

Access after this class.

I plan to purchase from cross-

border seller again in the future.

(N. Lankton et al., 2014)

CON3 I intend to continue using [Microsoft

Access/MySNW.com].

I intend to repurchase from

cross-border seller in near

future.


CON4 In the near future, I intend to continue

using Microsoft Access.

I intend to continue purchasing

from cross-border seller.


33

Ref. code: 25615802037480DCY

34

3.6 Data Analysis Process

IBM SPSS, a software package, is used to conduct descriptive statistics,

reliability test, normality test, correlation analysis, student’s t-test, and IBM AMOS is

used to conduct CFA and SEM on the data collected. The process of data analysis is

based on two aspects which are descriptive statistics, and inferential statistics.

3.6.1 Descriptive Statistics

The questions regarding consumer’s profile (e.g. age, gender, income,

purchase frequency) are analysed using descriptive statistics consisting of valid

percentage, frequency, mean, and standard deviation.

3.6.1.1 Variable Distribution and Normality Test

The analysis purpose is to observe the data distribution to

ensure that the data is normally distributed aligned to regression assumption. Due to

large sample size (between 30 – 1,000 samples in the test) (Hair, Black, Babin, &

Anderson, 2014), the Kolmogorov-Smirnov nonparametric test of normality is

incorporated in to the analysis with acceptable Lilliefors Significance Correction (Sig. P

< 0.05) (Hair et al., 2014; Nau, 2018).

3.6.1.2 Instrument Validity and Reliability Analysis

The research instrument reliability is analysed using

descriptive statistic tool, Cronbach’s alpha. The pilot group responses are analysed to

ensured that the factors will have internal integrity where Cronbach’s alpha for

extraversion subscale should be higher than 0.8 and each items alpha should be

sufficiently high which alpha is or higher than 0.7 (α >= 0.7) is considered acceptable

while alpha is or higher than 0.8 (α >= .8) is considered excellent (Cerny & Kaiser, 1977;

Kaiser, 1974).

3.6.1.3 Discriminant Validity Analysis

Measurement Model in SEM required good covariation between

factors in each construct. The validity and reliability analysis of measurement model

is conducted. The validity of measurement model consists of convergence validity

which can be evaluated using Average Variance Extracted, Construct Validity which can

Ref. code: 25615802037480DCY

35

be evaluated by achieving fitness indexes (refer to Fit Index summary in Table 3.3 and

Table 3.4), and discriminant validity which can be observed through correlation matrix.

The reliability of the instrument can be evaluated via Internal Reliability, Construct

Reliability, and Average Variance Extracted (Ahmad, Zulkurnain, & Khairushalimi, 2016).

Table 3.2

Model Validity and Reliability Analysis Result.

Test Result

Validity

Convergence Validity The value of AVE should be greater or equal

to 0.5 in order to achieve this validity.

Construct Validity The construct validity is achieved when the

Fit Indexes achieve the level of acceptance.

Discriminant Validity The correlation between each pair of latent

exogenous construct should be less than

0.85.

Reliability

Internal Reliability The Cronbach’s Alpha value is 0.6 (α >= 0.6)

or higher. Alpha of 0.8 or higher (α >= .8) is

considered excellent (Cerny & Kaiser, 1977;

Kaiser, 1974).

Construct Reliability CR ≥ 0.6 is required.

Average Variance Extracted AVE ≥ 0.5 for each construct.

Note. From (Ahmad et al., 2016; Bagozzi & Yi, 1988; Cerny & Kaiser, 1977; Hooper, Coughlan, & R. Mullen, 2007;

Kenny, 2015)

3.6.2 Inferential Statistics

In term of inferring the population properties, inferential statistics is

used to analyse the attitude measurement data in Likert-scale with defined 0.05

significant level for hypothesis test within 95% confidence interval.

3.6.2.1 Student t-test

The test is used to quantify difference in mean between pre

and post-purchase perception on Trust and Risk in which the function Paired Student’s

Ref. code: 25615802037480DCY

36

T-test is selected. The paired sample t-test with lower p-value (p < 0.05) within 95%

confident interval indicates significant difference in mean (two-tailed)(Student, 1908).

The expectation from student t-test is that there should be no difference between

pre-purchase and post-purchase construct as well as the domestic against cross-border

perception of trust and risk.

3.6.2.2 Structural Equation Model (SEM) and Confirmatory Factor

Analysis (CFA)

Confirmatory Factor Analysis (CFA) and Structural Equation

Model (SEM) is used to determine model structure based and factor covariation to

make a casual conclusion from the model. The goodness of fit of the model can be

determined by various fit index which can be classified in to 3 major groups which are

Absolute Fit Index, Comparative Fit Index, and Other Indices; however, it is

recommended to use Comparative Fit Index or combination of indices to evaluate the

goodness of fit rather than using Absolute Fit Index due to the AFI is highly sensitive

to sample size meaning that the index will performs better if the sample size increases

(Hooper et al., 2007; Hu & Bentler, 1999; Kenny, 2015; Schreiber, Nora, Stage, Barlow,

& King, 2006). The recommended indices to be used in this study are Standardized

Root Mean Square Residuals (SRMR), Root Mean Square Error of Approximation

(RMSEA), Bentler-Bonett Index or Normed-fit Index (NFI), Tucker-Lewis’s Non-normed

Fit Index (TLI or NNFI), and Comparative Fit Index (CFI). The criteria for each fit index

are shown on Table 3.3 below.

Ref. code: 25615802037480DCY

37

Table 3.3

SEM Recommended Fit Index and Criteria.

Fit Index Acceptable Threshold Level

Comparative Fit Index

NFI NFI > 0.95

Table 3.3

SEM Recommended Fit Index and Criteria. (Cont’d)

TLI TLI > 0.95

CFI CFI > 0.95

Other Fit Index

SRMR SRMR < 0.08

RMSEA RMSEA < 0.07 which RMSEA < 0.03 indicates excellent fit. Note. From (Schreiber et al., 2006)

Apart from using individual fit index to determine goodness

of fit of Structural Equation model, a combination of fit index is also suggested in

determining the goodness of fit (Hu & Bentler, 1999) which are combinations of NNFI

(TLI) with SRMR, RMSEA with SRMR, and CFI with SRMR together. The criteria used in

combination of fit index is provided on Table 3.4 below.

Table 3.4

SEM Fit Index Combination Criteria.

Combination Criteria Acceptable Threshold Level

NNFI (TLI) and SRMR NNFI of 0.96 or higher and an SRMR of .09 or lower

RMSEA and SRMR RMSEA of 0.06 or lower and a SRMR of 0.09 or lower

CFI and SRMR CFI of .96 or higher and a SRMR of 0.09 or lower Note. From (Hu & Bentler, 1999; Schreiber et al., 2006)

Ref. code: 25615802037480DCY

38

CHAPTER 4

RESULTS AND DISCUSSION

This study is an exploratory research on cross-border repurchase intention

based on Trust and Perceived Risk as main factors in both pre-purchase and pose-

purchase experience. We develop a questionnaire and conduct a survey to collect 462

samples. There are 440 valid respondents for they had experience on purchasing

tangible products from cross-border e-sellers which is considered as a good

representation of cross-border e-commerce population in Thailand.

4.1 Descriptive Statistics

Regression analysis involves critical assumptions that analyst has to ensure.

In this section, the observation and test result of normality of residuals,

multicollinearity of latent variables, reliability of measurement items, and factor

analysis are explained.

4.1.1 Variables Descriptive Statistics and Normality of Residuals

The descriptive statistics of measurement items for both SEM

analysis and Paired T-test Analysis is shown on Table 4.1.

Table 4.1

Factor Descriptive Statistics.

Latent Variable Factor N Minimum Maximum Mean Std. Deviation

Cross-border

Perceived Risk

Expectation

(CBPRE)

CBPRE1 440 1 5 3.41 0.921

CBPRE2 440 1 5 3.36 0.976

CBPRE3 440 1 5 3.6 0.799

CBPRE4 440 1 5 3.73 0.763

Ref. code: 25615802037480DCY

39

Table 4.1

Factor Descriptive Statistics. (Cont’d)


Cross-border

Perceived Risk

Performance

(CBPRP)

CBPRP1 440 1 5 3.25 0.76

CBPRP2 440 1 5 3.13 0.81

CBPRP3 440 1 5 3.5 0.8

CBPRP4 440 1 5 3.5 0.68

Cross-border

Trusting

Expectation

(CBTRE)

CBTRE1 440 1 5 3.93 0.48

CBTRE2 440 1 5 3.88 0.62

CBTRE3 440 1 5 3.78 0.68

CBTRE4 440 1 5 3.76 0.66

Cross-border

Trusting

Performance

(CBTRP)

CBTRP1 440 2 5 3.9 0.49

CBTRP2 440 2 5 3.95 0.59

CBTRP3 440 2 5 3.74 0.69

CBTRP4 440 2 5 3.83 0.6

Domestic

Perceived Risk

Expectation

(DMPRE)

DMPRE1 440 1 5 3.37 0.9

DMPRE2 440 1 5 3.4 0.93

DMPRE3 440 1 5 3.88 0.82

DMPRE4 440 1 5 4 0.71

Domestic

Perceived Risk

Performance

(DMPRP)

DMPRP1 440 1 5 3.18 1.11

DMPRP2 440 1 5 3.1 1.24

DMPRP3 440 1 5 3.28 1.14

DMPRP4 440 1 5 3.78 0.85

Domestic

Trusting

Expectation

(DMTRE)

DMTRE1 440 2 5 4.28 0.7

DMTRE2 440 1 5 4.04 0.66

DMTRE3 440 1 5 4.06 0.71

DMTRE4 440 1 5 4 0.73

Ref. code: 25615802037480DCY

40

Table 4.1

Factor Descriptive Statistics. (Cont’d)


Domestic

Trusting

Performance

(DMTRP)

DMTRP1 440 1 5 4.21 0.8

DMTRP2 440 1 5 4.08 0.69

DMTRP3 440 1 5 3.84 0.88

DMTRP4 440 1 5 4.06 0.82

Expectation

Disconfirmation

(DCE)

DCE1 440 1 5 3.77 0.58

DCE2 440 1 5 3.81 0.64

DCE3 440 1 5 3.75 0.69

DCE4 440 1 5 3.71 0.67

Satisfaction

(SAT)

SAT1 440 1 5 3.92 0.74

SAT2 440 1 5 3.92 0.74

SAT3 440 1 5 3.9 0.72

SAT4 440 1 5 4.06 0.67

Repurchase

Intention

(CON)

CON1 440 1 5 4.01 0.69

CON2 440 1 5 4.02 0.71

CON3 440 1 5 4.03 0.68

CON4 440 1 5 4.01 0.68

In the samples, perception of cross-border trusting performance

(CBTRP 1-4) has minimum Likert-scale value of 2 as well as domestic trusting

expectation item 1 (DMTRE1). The maximum Likert-scale value of all factors from the

samples is 5. The average of the scales in all factor are between 3 – 4. Standard

Deviation of the factor DMPRP 1 – 3 are quite stand out from the other factors (Std.

Deviation > 1) which reflect a wide spread of the distribution.

Apart from descriptive statistic for these factors, the cross-border

factors are observed their normality to comply with multivariate regression assumption

Ref. code: 25615802037480DCY

41

that the data should be normally distributed. The assessment detail produced from

IBM AMOS for SEM analysis is provided on Table 4.2.

Table 4.2

Assessment of Multivariate Normality.

Factor Min Max Skewness Skewness

Critical Ratio Kurtosis

Kurtosis

Critical Ratio

CBTRE1 1 5 -1.711 -14.649 8.623 36.920

CBTRE2 1 5 -0.926 -7.932 3.453 14.786

CBTRE3 1 5 -0.623 -5.337 1.727 7.395

CBTRE4 1 5 -0.672 -5.755 1.601 6.856

SAT1 1 5 -1.168 -10.006 3.094 13.249

SAT2 1 5 -0.965 -8.26 2.273 9.730

SAT3 1 5 -0.756 -6.477 1.552 6.646

SAT4 1 5 -0.973 -8.328 3.046 13.043

DCE1 1 5 -1.562 -13.372 3.569 15.28

DCE2 1 5 -1.411 -12.079 3.718 15.919

DCE3 1 5 -1.223 -10.475 2.151 9.211

DCE4 1 5 -1.227 -10.507 2.354 10.078

CON4 1 5 -0.872 -7.469 2.724 11.663

CON3 1 5 -0.647 -5.541 1.592 6.817

CON2 1 5 -0.928 -7.949 2.429 10.402

CON1 1 5 -0.983 -8.419 2.979 12.755

CBPRP4 1 5 -0.98 -8.393 0.296 1.267

CBPRP3 1 5 -0.644 -5.517 -0.05 -0.213

CBPRP2 1 5 -0.208 -1.781 -0.981 -4.198

CBPRP1 1 5 -0.48 -4.112 -0.496 -2.124

CBTRP4 2 5 -0.246 -2.11 0.349 1.495

CBTRP3 2 5 -0.077 -0.657 -0.218 -0.934

CBTRP2 2 5 -0.122 -1.046 0.265 1.137

CBTRP1 2 5 -0.678 -5.807 2.281 9.768

CBPRE1 1 5 0.109 0.929 -0.284 -1.217

Ref. code: 25615802037480DCY

42

Table 4.2

Assessment of Multivariate Normality. (Cont’d)

Factor Min Max Skewness Skewness

Critical Ratio Kurtosis

Kurtosis

Critical Ratio

CBPRE2 1 5 0.062 0.527 -0.418 -1.790

CBPRE3 1 5 -0.22 -1.881 0.019 0.081

CBPRE4 1 5 -0.292 -2.498 0.31 1.325

Multivariate Normality 43.122 11.034

It is very important to point that this study samples are not normally

distributed as the normality can be observed via the skewness statistics and skewness

critical ratio which higher than 1.96. This study does not remove any outlier from the

valid samples because this study believes that the responses from the samples are

skewed and not normal by nature. This lead the selection of the algorithm to use in

Structural Equation Model to handle non-normal data which maximum likelihood

estimator should be selected (Gao, Mokhtarian, & Johnston, 2008).

4.1.2 Measurement Model Validity and Reliability

The validity and reliability of the measurement model is analysed

based on the criteria on Chapter 3. Table 4.3 provides a summary of construct including

factor loading, Cronbach’s alpha (a), Average Variance Extracted (AVE), and Construct

Reliability (CR).

Table 4.3

Construct Summary

Construct Measurement Item Factor Loading a AVE CR

Cross-border Perceived Risk

Expectation

(CBPRE)

CBPRE4 0.674 0.839 0.730 0.862

CBPRE3 0.701

CBPRE2 0.781

CBPRE1 0.765

Ref. code: 25615802037480DCY

43

Table 4.3

Construct Summary. (Cont’d)

Construct Measurement Item Factor Loading a AVE CR

Cross-border Trusting

Performance

(CBTRP)

CBTRP1 0.747 0.778 0.691 0.910

CBTRP2 0.678

CBTRP3 0.635

CBTRP4 0.702

Cross-border Perceived Risk

Performance

(CBPRP)

CBPRP1 0.654 0.807 0.714 0.881

CBPRP2 0.593

CBPRP3 0.919

CBPRP4 0.688

Repurchase Intention

(CON)

CON1 0.715 0.804 0.712 0.896

CON2 0.712

CON3 0.674

CON4 0.748

Expectation Disconfirmation

(DCE)

DCE4 0.773 0.868 0.785 0.938

DCE3 0.768

DCE2 0.780

DCE1 0.820

Satisfaction

(SAT)

SAT4 0.751 0.813 0.699 0.881

SAT3 0.698

SAT2 0.618

SAT1 0.730

Cross-border Trusting

Expectation

(CBTRE)

CBTRE4 0.669 0.721 0.630 0.876

CBTRE3 0.628

CBTRE2 0.652

CBTRE1 0.571

Note. a = Cronbach’s Alpha; AVE = Average Variance Extracted; CR = Construct Validity

Referring to discriminant validity analysis, the correlation matrix is

observed showing that there is no too highly correlated construct (correlation

coefficients > 0.85) in the model as shown on Table 4.4. Therefore, there is no

multicollinearity evident on the model.

Ref. code: 25615802037480DCY

44

Table 4.4

Cross-border E-commerce Variable Correlations for Discriminant Validity

CBTRE CBPRE CBTRP CBPRP DIS SAT CON

CBTRE 1

CBPRE .512** 1

CBTRP 0.083 -.191** 1

CBPRP .189** -0.090 .431** 1

DIS -.115* -.389** .282** .475** 1

SAT -0.006 -.256** .162** .439** .636** 1

CON -.120* -.168** .123** .192** .335** .222** 1

Note. CBTRE = Trusting Expectation; CBPRE = Perceived Risk Performance; CBTRP = Trusting Performance; CBPRP

= Perceived Risk Performance; DIS = Disconfirmation; SAT = Satisfaction; CON = Repurchase Intention

In term of CFA, discriminant validity in each factor is observed to

ensure non-multicollinearity issue on the model. A correlation of 0.85 and higher (or -

0.85 and lower) indicates poor discriminant validity which leads to multicollinearity

issue. The factor correlation on Appendix B’s Table D.1 indicates that all factors in the

model have lower correlation and they are acceptable for CFA and SEM.

Summary of Validity and Reliability of Measurement Model is

provided on the Table 4.5 passing all criteria required which indicates that the model

is valid and reliable to make prediction of the population.

Ref. code: 25615802037480DCY

45

Table 4.5

Measurement Model Validity and Reliability Analysis Result

Test Result

Validity

Convergence Validity All items in a measurement model are statistically

significant. (Refer to Table B.1) Other than that, the

value of AVE for all construct is greater than 0.50.

The Convergent Validity was achieved the

required level. (Refer to Table 4.3)

Construct Validity The construct validity was achieved the required

level. (Refer to Table 4.3)

Discriminant Validity The correlation between all constructs are not

lower than -0.85 or higher than 0.85. (Refer to

Table 4.4)

Reliability

Internal Reliability The value of Cronbach Alpha is greater than 0.60.

The internal reliability was achieved the required

level. (Refer Table 4.4)

Construct Reliability The value of CR for all constructs are greater than

0.60. The composite reliability was achieved the

required level.

Average Variance Extracted The value of AVE for all constructs are greater than

0.50. The required level was achieved. (Refer

Table 8)

4.1.2 Samples demographic

Based on 440 samples, descriptive statistics is used to explain

properties of the samples collected. There are six different demographics observed

which are age, gender, occupation, highest education, monthly income, and frequency

of purchase.

Ref. code: 25615802037480DCY

46

The histogram on Figure 4.1 below shows the respondent age

distribution mostly fall between late 20s to mid 30s as detail provided which are a

major consumer of e-commerce in Thailand. This statistic is aligned to penetration of

digital adopters in Thailand based on National Statistical Office of Thailand (National

Statistical Office of Thailand, 2019).

Figure 4.1 Histogram of sample gender.

53 percent of total respondents is female, and another 47 percent is male.

This proportion of gender is slightly different to Thailand population which is

approximately symmetric in gender (50 to 50 percent of male and female).

Table 4.6

Gender

Frequency Percent

Female 233 53.0

Male 207 47.0

61 percent of total respondents, a major group, earns less than 35

thousand Thai Baht a month. The second largest group earns around 25 to 50 thousand

Ref. code: 25615802037480DCY

47

a month contributed to 33 percent of total respondents. The smallest groups of are

50 to 75 thousand, 75 to 100 thousand, and over 100 thousand monthly income which

contribute 3 percent, 1 percent, and 0.7 percent respectively.

Table 4.7

Monthly Income

Frequency Percent

25K - 50K 149 33.9

50K - 75K 14 3.2

75K - 100K 5 1.1

Less than 25K 269 61.1

Over 100K 3 0.7

74 percent of the respondents obtained bachelor’s degree which is the

major respondents of this study. 10 and 13 percent of the respondents obtain master’s

degree and diploma/certificate respectively. Only 2 percent of the respondents obtain

only high school level. These statistics also represent the distribution of Thai

population regarding the National Statistical Office of Thailand(National Statistical

Office of Thailand, 2019).

Table 4.8

Highest Education Obtained

Frequency Percent

Bachelor’s degree 326 74.1

Diploma/Certificates 58 13.2

High School 9 2.0

Master’s degree 47 10.7

69 percent of the respondents works in private company while a 10 and

15 percent are students and government officers respectively. The smallest group of

Ref. code: 25615802037480DCY

48

respondents reported at 5.2 percent is a business owner or a freelancer group as

showed on Table 4.9 below.

Table 4.9

Occupation

Frequency Percent

Business owner/ Freelancer 23 5.2

Employee 307 69.8

Government Officers 66 15.0

Student 44 10.0

In term of frequency of purchase via cross-border e-commerce, the

participants purchase 3 times per month on average; however, there are some

respondents make purchase more than 10 times a month. Nearly 50 respondents make

purchase via cross-border e-commerce only once a month. The distribution of

purchase frequency is showed on Figure 4.2 below.

Figure 4.2 Frequency participants’ purchase via cross-border e-commerce (monthly)

Ref. code: 25615802037480DCY

49

4.2 Paired T-test for Mean Difference

In this study, paired t-test are conducted to measure statistical difference

of pre-purchase trusting expectation, trusting performance, and perceived risk

expectation and perceived risk performance between cross-border e-commerce and

domestic e-commerce. Additionally, the paired t-test is used to compare mean

difference between pre-purchase trusting belief and perceived risk (Trusting

Expectation and Perceived Risk Performance) and post-purchase trusting belief and

perceived risk (Trusting Performance and Perceived Risk Performance). The pair

summary of each paired t-test conducted in this study is provided on Table 4.10.

Table 4.10

Pair Summary for Paired T-test of Mean Difference.

Pair Variables

Pair 1 Cross-border Trusting Expectation (CBTRE)

Domestic Trusting Expectation (DMTRE)

Pair 2 Cross-border Perceived Risk Expectation (CBPRE)

Domestic Perceived Risk Expectation (DMPRE)

Pair 3 Cross-border Trusting Performance (CBTRP)

Domestic Trusting Performance (DMTRP)

Pair 4 Cross-border Perceived Risk Performance (CBPRP)

Domestic Perceived Risk Performance (DMPRP)

Pair 5 Cross-border Trusting Expectation (CBTRE)

Cross-border Trusting Performance (CBTRP)

Pair 6 Cross-border Perceived Risk Expectation (CBPRE)

Cross-border Perceived Risk Performance (CBPRP)

A paired t-test on Pair 1 showed statistically insignificant difference

between pre-purchase trusting belief (Trusting Expectation) and post-purchase trusting

belief (Trusting Performance) from cross-border e-commerce (Mean difference = -.018,

SD = .488, p > 0.05) with a 95% confidence interval ranging from -0.63 to .027.

Ref. code: 25615802037480DCY

50

A paired t-test on Pair 2 showed statistically significant increase of post-

purchase perceived risk (Perceived Risk Performance) over pre-purchase perceived risk

(Perceived Risk Expectation) from cross-border e-commerce (Mean difference = -.018,

SD = .658, p < 0.05) with a 95% confidence interval ranging from .116 to .240.

A paired t-test on Pair 3 showed statistically significant lower mean of

cross-border trusting belief (Trusting Expectation) over domestic e-commerce trusting

belief (Trusting Expectation) commerce on pre-purchase stage (Mean difference = -

.260, SD = .613, p < 0.05) with a 95% confidence interval ranging from -.317 to -.202.


cross-border perceived risk (Perceived Risk Expectation) over domestic perceived risk

(Perceived Risk Expectation) from pre-purchase stage (Mean difference = -.138, SD =

.765, p < 0.05) with a 95% confidence interval ranging from -.209 to -.066.


cross-border trusting belief (Trusting Performance) over domestic trusting belief

(Trusting Performance) from post-purchase stage (Mean difference = -.193, SD = .649,

p < 0.05) with a 95% confidence interval ranging from -.254 to -.132.

A paired t-test on Pair 6 showed statistically insignificant difference

between cross-border perceived risk (Perceived Risk Performance) and domestic

perceived risk (Perceived Risk Performance) from post-purchase stage (Mean difference

= .007, SD = .970, p > 0.05) with a 95% confidence interval ranging from -0.82 to .098.

The summary detail is provided on Table 4.11

Ref. code: 25615802037480DCY

Table 4.11 Paired Samples Test of Mean between Pre-purchase and Post-purchase from Cross-border E-commerce and Domestic E-commerce.

Paired Differences

t df Sig. (2-tailed) Mean Std. Deviation Std. Error Mean

95% Confidence Interval of the Difference

Lower Upper Pair 1 CBTRE - DMTRE -.260 .613 .029 -.317 -.202 -8.898 439 .000** Pair 2 CBPRE - DMPRE -.138 .765 .036 -.209 -.066 -3.784 439 .000** Pair 3 CBTRP - DMTRP -.193 .649 .030 -.254 -.132 -6.262 439 .000** Pair 4 CBPRP - DMPRP .007 .970 .046 -.082 .098 .172 439 .864 Pair 5 CBTRE - CBTRP -.018 .488 .023 -.063 .027 -.780 439 .436 Pair 6 CBPRE - CBPRP .178 .658 .031 .116 .240 5.681 439 .000** Note. CBTRE = Cross-border Trusting Expectation, CBPRE = Cross-border Perceived Risk Expectation, CBTRP = Cross-border Trusting Performance, CBPRP = Cross-border Perceived Risk Performance, DMTRE = Domestic Trusting Expectation, DMPRE = Domestic Perceived Risk Expectation, DMTRP = Domestic Trusting Performance, DMPRP = Domestic Perceived Risk Performance * p < .01. **p < .001.

51

Ref. code: 25615802037480DCY

52

4.3 Confirmatory Factor Analysis

In Confirmatory Factor Analysis, each factor in the model is observed their

critical ratio (t-value) in which C.R. higher than 2.56 indicate significant error variance at

0.01 level. In this study, all factors in the model have very high C.R. indicating that they

are significant in term of CFA. Square Multiple Correlation (SMC) on each latent variable

are positive indicating significant relationship between each measurement items and

its latent variable.

Table 4.12

Confirmatory Factor Analysis Coefficients and Squared Multiple Coefficients.

Observed

Variable

Latent

Construct b B S.E. C.R. SMC

P-

value

CBPRE4

CBPRE

0.674 1.000 0.447

CBPRE3 0.701 1.088 0.082 13.282 0.394 ***

CBPRE2 0.781 1.481 0.139 10.656 0.425 ***

CBPRE1 0.765 1.369 0.130 10.516 0.326 ***

CBTRP1

CBTRP

0.747 1.000 0.558

CBTRP2 0.678 1.084 0.085 12.780 0.459 ***

CBTRP3 0.635 1.182 0.098 12.030 0.404 ***

CBTRP4 0.702 1.135 0.086 13.198 0.493 ***

CBPRP1

CBPRP

0.654 1.000 0.414

CBPRP2 0.593 0.965 0.067 14.417 0.356 ***

CBPRP3 0.919 1.489 0.120 12.450 0.847 ***

CBPRP4 0.688 0.950 0.094 10.152 0.445 ***

CON1

CON

0.715 1.000 0.474

CON2 0.712 1.033 0.082 12.582 0.505 ***

CON3 0.674 0.930 0.077 12.049 0.439 ***

CON4 0.748 1.038 0.080 13.001 0.524 ***

Ref. code: 25615802037480DCY

53

Table 4.12

Confirmatory Factor Analysis Coefficients and Squared Multiple Coefficients.

(Cont’d)

Observed

Variable

Latent

Construct b B S.E. C.R. SMC

P-

value

DCE4 DCE 0.773 1.000 0.598

DCE3 0.768 1.020 0.063 16.218 0.589 ***

DCE2 0.780 0.968 0.059 16.537 0.609 ***

DCE1 0.820 0.927 0.052 17.659 0.672 ***

SAT4 SAT 0.751 1.000 0.564

SAT3 0.698 0.998 0.076 13.090 0.487 ***

SAT2 0.618 0.901 0.081 11.182 0.382 ***

SAT1 0.730 1.070 0.079 13.596 0.533 ***

CBTRE4 CBTRE 0.669 1.000 0.447

CBTRE3 0.628 0.957 0.097 9.894 0.488 ***

CBTRE2 0.652 0.918 0.091 10.125 0.617 ***

CBTRE1 0.571 0.616 0.067 9.264 0.586 ***

Note. b = Standardized Coefficient Estimate; B = Unstandardized Coefficients; S.E. = Standard Error; P = p-value;

C.R. = Critical Ratio (t-value); SMS = Squared Multiple Correlation

The Goodness of Fit (GOF) of the model for baseline model (c2 =

518.700, DF = 328) is provided from Confirmatory Factor Analysis. The GOF is evaluated

in accordance to recommendation criteria on Chapter 3 which are now summaries in

Table 4.13 below. The model passes TLI, CFI, and SRMR criteria but does not achieve

the NFI criteria. In term of combination criteria, RMSEA & SRMR and CFI & SRMR are

achieved; however, only the combination of NNFI (TLI) & SRMR is not achieved. In

conclusion, most of the criteria for goodness of fit is achieved to justify that this model

is valid.

Ref. code: 25615802037480DCY

54

Table 4.13 Structural Model Goodness of Fit Evaluation Summary.

Fit Index Acceptable Threshold Level Model Result

Model Fit Acceptance

Comparative Fit Index NFI NFI > 0.95 0.901 Not

Accepted TLI TLI > 0.95 0.955 Accepted CFI CFI > 0.95 0.960 Accepted

Other Fit Index SRMR SRMR < 0.08 0.055 Accepted RMSEA RMSEA < 0.07

RMSEA < 0.03 indicates excellent fit.

0.037 Accepted

Combination Criteria NNFI (TLI) and SRMR NNFI (TLI) ³ 0.96 and SRMR £

0.09.

Not

Accepted RMSEA and SRMR RMSEA £ and SRMR £ 0.09.

Accepted

CFI and SRMR CFI ³ 0.96 and SRMR £ 0.09 Accepted

4.4 Structural Equation Model

Result from SEM estimation, this model converges properly with Maximum

Likelihood Estimator providing overall model summary and CFA result as discussed on

the previous section. The model also identifies direct effect, indirect effect, and total

effect as shown on Table 4.14 below. The discussion of direct and indirect effect

within the model are discussed on 4.4.1 and 4.1.2.

Ref. code: 25615802037480DCY

55

4.4.1 Direct Effect

This study hypothesises that each construct formulates a direct

relationship to the dependent variable; therefore, the direct effect is observed. Trusting

Expectation (CBTRE) (Beta = 0.11, p > 0.05) and Perceived Risk Expectation (CBPRE)

(Beta = 0.04, p > 0.05) are not significantly related to Expectation Disconfirmation (DCE)

and they have very little effect size towards the Expectation Disconfirmation (DCE).

Trusting Expectation (CBTRE) has positive effect on Trusting Performance (CBTRP) on

the post-purchase stage (Beta = 0.58, p < 0.05). Perceived Risk Expectation (CBPRE) has

positive effect on Perceived Risk Performance (CBPRP) on the post-purchase stage (Beta

= 0.62, p < 0.05). Trusting Performance is predictive of Expectation Disconfirmation

(DCE) (Beta = 0.53, p <0.05) as well as the Perceived Risk Performance (CBPRP) (Beta =

-0.43, p < 0.05). The relationship of Expectation Disconfirmation (DCE), Satisfaction

(SAT), and Repurchase Intention (CON) is significant as well where Expectation

Disconfirmation (DCE) is predictive of Satisfaction (SAT) (Beta = 0.80, p < 0.05) and

Satisfaction (SAT) is predictive of Repurchase Intention (CON) (Beta = 0.51, p < 0.05).

Figure 4.3 Structural Model Standardised Coefficients Note. c2 = 518.700; DF = 328; Standardised Root Mean Square Residuals (SRMR) = 0.030; Root Mean Square Error

of Approximation (RMSEA) = 0.037; Normed Fit Index (NFI) = 0.901; Non-normed Fit Index (TLI) = 0.955;

Comparative Fit Index (CFI) = 0.960.

Ref. code: 25615802037480DCY

56

4.4.2 Indirect Effect

We hypothesized that the relationship between the belief that

knowledge as isolated facts and course achievement was mediated has an indirect

effect on course achievement, by deep processing. The result (standardized indirect

coefficient = –.06, p > .05) was not statistically significant.

Table 4.14

Standardized Coefficients Effect in Structural Model.

Model Effect Standardized Coefficients (b)

R2 CBTRE CBTRP CBPRE CBPRP DCE SAT CON

CBTRP 0.589 0.000

CBPRE 0.117 0.347

CBPRP -0.300 0.625 0.014

DCE -0.850 0.585 0.490 -0.436 0.459

SAT 0.820 0.589

CON 0.510 0.642

CBTRP

CBPRE

CBPRP -0.140

DCE 0.395 0.131 -0.272

SAT 0.249 0.574 -0.179 -0.349

CON 0.127 0.293 -0.910 -0.178 0.490

CBTRP 0.589

CBPRE 0.117

CBPRP -0.140 -0.300 0.625

DCE 0.310 0.715 -0.224 -0.436

SAT 0.249 0.574 -0.179 -0.349 0.820

CON 0.127 0.293 -0.910 -0.178 0.490 0.510

Ref. code: 25615802037480DCY

57

4.5 Hypothesis Testing Summary

Based on SEM, each hypothesis proposed in this study has been evaluated

and summarized as shown on the Table 4.15 for paired T-test and 4.16 for SEM model

hypothesis.

4.5.1 Hypothesis Testing Result

In term of Paired t-test Hypothesis testing, the result shows that the

H2c and H3c are supported as referred to Table 4.18 (p < .05) within 95% confidence

interval and the mean difference is negative in accordance to the research hypothesis.

Table 4.15

Paired T-test Hypothesis Testing Result

Hypothesis Result

H2c Trusting Expectation from cross-border e-commerce is less than

domestic e-commerce on average. Supported

H2d Perceived Risk Expectation from cross-border e-commerce is higher

than domestic e-commerce on average. Not Supported

H3c Trusting Performance from cross-border e-commerce is less than

domestic e-commerce on average. Supported

H3d Perceived Risk Performance from cross-border e-commerce is higher

than domestic e-commerce on average. Not Supported

H4b Trusting Performance (post-purchase) is increased from Trusting

Expectation (pre-purchase). Not Supported

H5b Perceived Risk Performance (post-purchase) is decreased from

Perceived Risk Expectation (pre-purchase). Not Supported

The Structural Equation Model conducted on the cross-border

variables provides a hypothesis testing result as provided on Table 4.16 below. The

hypotheses H1, H2a and H2b are not supported in this study. The discussion on the

hypothesis result is discussed on 4.5.1 to 4.5.5.

Ref. code: 25615802037480DCY

58

Table 4.16

SEM Hypothesis Testing Result

Hypothesis Result

H1 Trusting Expectation from cross-border e-commerce has negative

effect on Perceived Risk Expectation from cross-border e-commerce.

Not Supported

H2a Trusting Expectation from cross-border e-commerce has positive effect

on Expectation Disconfirmation from cross-border e-commerce.

Not Supported

H2b Perceived Risk Expectation from cross-border e-commerce has

negative effect on Expectation Confirmation from cross-border e-

commerce.

Not Supported

H3a Trusting Performance from cross-border e-commerce has positive

effect on Expectation Confirmation from cross-border e-commerce.

Supported

H3b Perceived Risk Performance from cross-border e-commerce has

negative effect on Expectation Confirmation from cross-border e-

commerce.

Supported

H4a Trusting Expectation has positive effect on Trusting Performance. Supported

H5a Perceived Risk Expectation has negative effect on Perceived Risk

Performance.

Supported

H6 Trusting Performance has negative effect on Perceived Risk

Performance

Supported

H7 Expectation Disconfirmation has positive effect on Satisfaction. Supported

H8 Satisfaction has positive effect on Repurchase Intention. Supported

4.5.2 Relationship of Trusting Expectation and Perceived Risk

Expectation

The result from hypothesis test shows that consumer’s expectation

of Trust is not influential to Perceived Risk Expectation (H1) which is against the

previous study on trust and risk interaction (Kim et al., 2008, 2009; Mou et al., 2015;

Zhu et al., 2009); however, this relationship is may not relevant in the context of cross-

border e-commerce amongst Thai consumers.

Ref. code: 25615802037480DCY

59

4.5.3 Relationship Trusting Expectation, Perceived Risk Expectation,

and Expectation Disconfirmation.

The regression analysis result suggests that the Trusting Expectation

and Perceived Risk Expectation are significantly affect consumer’s Expectation-

disconfirmation of purchasing via cross-border e-commerce. The relationship of

expectation variables in this study reject the result from previous study on the

existence of the expectation construct in the EDT (Ambalov, 2018; Bhattacherjee, 2001;

N. Lankton et al., 2014; N. K. Lankton et al., 2016; Zhang, Lu, Gupta, & Gao, 2015), and

the extension of expectation to Trusting Expectation and Perceived Risk Expectation.

The interesting finding from this expectation variables is that they do not expect the

e-sellers much in term of trust and risk or they have quite vary perception in the

shopper’s mind.

Trusting Performance (H2a) and Perceived Risk Performance (H2b)

which extend the original EDT model (Ambalov, 2018; Bhattacherjee, 2001; Kim et al.,

2009; N. Lankton et al., 2014; N. K. Lankton et al., 2016) replacing the post-purchase

stage’s Performance in this study are significantly affect Expectation-disconfirmation

in the EDT model as well.

4.5.4 Relationship of Trusting Performance, Perceived Risk

Performance, and Expectation Disconfirmation.

Trusting Performance and Perceived Risk Performance is a post-

purchase evaluation of Trust and Risk at the pre-purchase stage. The hypothesis test

result suggests that the Trusting Performance has positive effect over customer

disconfirmation (H3a) while the Perceived Risk Performance negatively influences the

disconfirmation (H5). This result is aligned with the EDT in which the Performance

variables have influence Expectation Disconfirmation (Ambalov, 2018; Bhattacherjee,

2001; N. K. Lankton et al., 2016; Shang & Wu, 2017). Trusting Performance (H3a) and

Perceived Risk Performance (H3b) which extend the original EDT model replacing the

post-purchase stage’s Perceived Performance in this study (N. Lankton et al., 2014) are

significantly affect Expectation-disconfirmation in the EDT model as well.

Ref. code: 25615802037480DCY

60

4.5.5 Relationship of Trusting Performance and Perceived Risk

Performance

The result from hypothesis test shows that consumer Trust does not

significant affect Perceived Risk Expectation (H6) in the accordance to the Trusting

Expectation and Perceived Risk Expectation (H1). This is an alignment that may suggest

that trusting belief and perceived risk interaction may not exist in cross-border e-

commerce context; however, in the other social context, these variables may be

related.

4.5.6 Relationship of Disconfirmation, Satisfaction and Repurchase

Intention

Based on the regression analysis and hypothesis testing, the result

supports the hypothesis that the disconfirmation has positive influence on consumer

satisfaction (H7), and the consumer satisfaction has positive effect on continuance

intention (H8). These relationships affirm the EDT variables relationships in previous

study (Ambalov, 2018; Bhattacherjee, 2001; N. Lankton et al., 2014; N. K. Lankton et

al., 2016; Zhang et al., 2015), but the result in this study is in a context of cross-border

e-commerce.

Ref. code: 25615802037480DCY

61

CHAPTER 5

CONCLUSIONS AND RECOMMENDATIONS

In this chapter, conclusion of the study has provided as well as benefit of

the study, limitation, and recommendation for future study.

5.1 Conclusion

E-commerce has become a new world economy due to advanced

technology nowadays. The connecting world has provided a vast opportunity in

commerce for everybody and the e-commerce has been expanding from domestic

shopping into cross-border shopping; however, there are many barriers that can slow

down the cross-border e-commerce especially risks and trust that the consumer needs

to be evaluate. The domestic sellers have to adapt themselves to the new wave of

commerce transformation and challenges. Understanding the key factor that the

consumer adopt cross-border e-commerce is a key to their competitiveness in both

retaining domestic customer and expanding current business to become cross-border

e-commerce.

This study has conducted a measurement of consumer trusting belief and

perceived risk in pre and post purchase within an Expectation-disconfirmation Theory

in context of cross-border e-commerce. The result from this study shows that trust

and risk in cross-border e-commerce context are significantly important in post-

purchase stage which is the same result in the previous study. This study also confirm

that the expectation of these constructs not is settled well within Expectation-

disconfirmation Theory, and relationships between trust and risk in this study shows

insignificance.

In term of difference perception of pre-to-post purchase trust, the paired

t-test shows that the consumer expects that domestic e-sellers would be more

Ref. code: 25615802037480DCY

62

trustworthy than cross-border e-sellers, and their belief is confirmed as showed in the

post-purchase trusting belief which is still lower than domestic e-commerce.

Regarding the difference perception of pre-to-post purchase perceived risk,

the paired t-test shows that the consumer expects that cross-border e-commerce

would be less risky than domestic e-commerce. Interestingly, the consumer perceived

that purchasing from cross-border e-commerce may be risker than expected but the

risk in not different to purchasing via domestic e-commerce. This evaluation of risks is

different to trust for trust is determining on human-like trust while the perceived risks

involves process and systems. This phenomenon imply that cross-border e-commerce

may have more or equally risk over systems and process to domestic e-commerce

while e-seller in cross-border shops are less trustworthy.

The relationship of cross-border variables in the analysis shows that the

effect of trust and risk towards consumer disconfirmation of their initial belief is strong

and stronger in the post-purchase stage. It is very important to firms who would like

to gain higher competitive advantages over the others to pay their attention to building

trust over e-sellers and reducing risks over systems and processes. A proper

management on trust and risk in cross-border e-commerce or domestic e-commerce

will be resulted in higher satisfaction and loyalty of the consumers.

In the context of domestic e-commerce providers, trust and risk are crucial

factors to compete with this new era of colonisation, cross-border e-commerce

expansion, as reflected on t-test that the consumers still believe that domestic e-

sellers are more reliable, trust worthy, and less risky. If the belief is confirmed and

intensified positively, the domestic e-commerce sellers will position themselves

strongly against a high influential cross-border e-commerce firm.

5.2 Benefit

This study provides two aspects of benefits which are theoretical benefit

and the practical benefit for business and practitioners in the area the e-commerce.

Ref. code: 25615802037480DCY

63

5.2.1 Theoretical benefit

This study derived the Expectation Disconfirmation Theory to apply

on a context of cross-border e-commerce. The previous studies have been formulating

the EDT model with various constructs; however, in the cross-border e-commerce

setting, the knowledge have not been explored. This study has expanded the

Expectation-disconfirmation Theory to the cross-border e-commerce extent. Moreover,

the previous study on Trust and EDT (Bhattacherjee, 2001; N. Lankton et al., 2014) on

e-service has been reformulated with Perceived Risk in area of information system

(Featherman & Pavlou, 2003; P. A. Pavlou, 2003) into the EDT model providing insight

in how cross-border shoppers perceived towards e-sellers an service providers.

Additionally, the Expectation-disconfirmation Theory with pre-purchase and post-

purchase evaluation driving with Trusting Belief and Perceived Risks in the context

cross-border e-commerce is well settled harmoniously; however, the interaction of

Trust and Risk in previous research (Kim et al., 2008; Mou et al., 2015; P. A. Pavlou,

2003; Visschers & Siegrist, 2008; Zhu et al., 2009) is not confirmed in this study.

5.2.2 Practical benefit

Entrepreneurs and business owner who are running e-commerce and

is suffering from challenges from big e-commerce companies can adopt this study to

improve their business to be able to compete or mitigate the impact of cross-border

expansion. On domestic firms’ perspective, the many big firms running cross-border e-

commerce are challenging domestic e-commerce provider with their vast amount of

investment; however, the result of Trust and Risk in this study suggests the domestic

e-commerce firms to be proactive to raising trust and reducing risks as they are highly

influential to consumer disconfirmation of their prior belief as well as the post-

purchase trust and risk perception. Vice versa, for entrepreneurs who is looking for

expanding their cross-border e-commerce, especially to Thailand, they need to

carefully develop strategies to capture consumer trust and reducing risks especially

risk of deliver and logistic service in Thailand. The product and financial risk are more

concerning factors in post-purchase via cross-border e-commerce i.e. the product

might be defective or bogus, and the process of transaction both payment and refund

Ref. code: 25615802037480DCY

64

involve more risk concerns as well which may due to distance effect of sellers and

consumers.

5.3 Limitation

This study was conducted on Thai population only. The result may not be

able to be applied to people in different countries or different social and culture

context. Additionally, the cross-border e-commerce in Thailand is still at a growing

state and not matured as global e-commerce; therefore, the consumer may not

compare them distinctively. Moreover, risks in this study are evaluated as a general

perception in combination of financial risk, delivery risk, and overall risk. This study

measures the pre-purchase expectation as consumer recall of their past-experience

rather than measuring before the actual purchase; therefore, the result of expectation

construct may be distorted by time of answering the questions.

5.4 Recommendation for future study

The result from this study shows that the incorporating trust and risks into

The EDT model is significant; however, the interaction on the trust and risks does not

support the hypothesis. The future research on cross-border e-commerce may observe

their relationships again in a different culture and social setting which may have

different outcome. The perceived risk in this study may be separated and emphasize

independently in the future research to observe them differently such as financial risk,

performance risk (product defection or bogus), and overall risk. There are other factors

that might affect the perception of risk amongst the consumer as well, but they are

not included in this study such as psychological risk and social risk. Moreover, the

research design in the future may conduct twice in a style of longitudinal study

separating the pre-purchase and post-purchase measurement conducting on the same

samples.

Ref. code: 25615802037480DCY

65

REFERENCE

Journal Article

Ambalov, I. A. (2018). A meta-analysis of IT continuance: An evaluation of the

expectation-confirmation model. Telematics and Informatics, 35(6), 1561-

1571. doi:10.1016/j.tele.2018.03.016

Bagozzi, R. P., & Yi, Y. J. J. o. t. A. o. M. S. (1988). On the evaluation of structural

equation models. 16(1), 74-94. doi:10.1007/bf02723327

Bhattacherjee, A. (2001). Understanding Information Systems Continuance: An

Expectation-Confirmation Model. MIS Quarterly, 25(3), 351-370.

doi:10.2307/3250921

Bonsón Ponte, E., Carvajal-Trujillo, E., & Escobar-Rodríguez, T. (2015). Influence of

trust and perceived value on the intention to purchase travel online:

Integrating the effects of assurance on trust antecedents. Tourism

Management, 47, 286-302. doi:10.1016/j.tourman.2014.10.009

Cerny, B. A., & Kaiser, H. F. (1977). A Study Of A Measure Of Sampling Adequacy For

Factor-Analytic Correlation Matrices. Multivariate Behavioral Research, 12(1),

43-47. doi:10.1207/s15327906mbr1201_3

Chong, A. Y.-L. (2015). Understanding Mobile Commerce Continuance Intentions: An

Empirical Analysis of Chinese Consumers. Journal of Computer Information

Systems, 53(4), 22-30. doi:10.1080/08874417.2013.11645647

Fang, J., George, B., Shao, Y., & Wen, C. (2016). Affective and cognitive factors

influencing repeat buying in e-commerce. Electronic Commerce Research and

Applications, 19, 44-55. doi:10.1016/j.elerap.2016.08.001

Featherman, M. S., & Pavlou, P. A. (2003). Predicting e-services adoption: a perceived

risk facets perspective. International Journal of Human-Computer Studies,

59(4), 451-474. doi:10.1016/S1071-5819(03)00111-3

Ref. code: 25615802037480DCY

66

Fisher, R., & Zoe Chu, S. (2009). Initial online trust formation: the role of company

location and web assurance. Managerial Auditing Journal, 24(6), 542-563.

doi:10.1108/02686900910966521

Gao, S., Mokhtarian, P. L., & Johnston, R. A. (2008). Nonnormality of Data in Structural

Equation Models. 2082(1), 116-124. doi:10.3141/2082-14

Gefen, D., Karahanna, E., & Straub, D. W. (2003). Trust and TAM in Online Shopping:

An Integrated Model. MIS Quarterly, 27(1), 51-90. doi:10.2307/30036519

Guo, Y., Bao, Y., Stuart, B. J., & Le-Nguyen, K. (2018). To sell or not to sell: Exploring

sellers' trust and risk of chargeback fraud in cross-border electronic

commerce. Information Systems Journal, 28(2), 359-383. doi:10.1111/isj.12144

Hong, I. B. (2015). Understanding the consumer's online merchant selection process:

The roles of product involvement, perceived risk, and trust expectation.

International Journal of Information Management, 35(3), 322-336.

doi:10.1016/j.ijinfomgt.2015.01.003

Hu, L. t., & Bentler, P. M. (1999). Cutoff criteria for fit indexes in covariance structure

analysis: Conventional criteria versus new alternatives. Structural Equation

Modeling: A Multidisciplinary Journal, 6(1), 1-55.

doi:10.1080/10705519909540118

Kaiser, H. F. (1974). An index of factorial simplicity. Psychometrika, 39(1), 31-36.

doi:10.1007/BF02291575

Kim, D. J. (2014). A Study of the Multilevel and Dynamic Nature of Trust in E-

Commerce from a Cross-Stage Perspective. International Journal of Electronic

Commerce, 19(1), 11-64. doi:10.2753/jec1086-4415190101

Kim, D. J., Ferrin, D. L., & Rao, H. R. (2008). A trust-based consumer decision-making

model in electronic commerce: The role of trust, perceived risk, and their

antecedents. Decision Support Systems, 44(2), 544-564.

doi:10.1016/j.dss.2007.07.001

Kim, D. J., Ferrin, D. L., & Rao, H. R. (2009). Trust and Satisfaction, Two Stepping

Stones for Successful E-Commerce Relationships: A Longitudinal Exploration.

Information Systems Research, 20(2), 237-257. doi:10.1287/isre.1080.0188

Ref. code: 25615802037480DCY

67

Lankton, N., McKnight, D. H., & Thatcher, J. B. (2014). Incorporating trust-in-technology

into Expectation Disconfirmation Theory. The Journal of Strategic Information

Systems, 23(2), 128-145. doi:10.1016/j.jsis.2013.09.001

Lankton, N. K., McKnight, D. H., Wright, R. T., & Thatcher, J. B. (2016). Research Note—

Using Expectation Disconfirmation Theory and Polynomial Modeling to

Understand Trust in Technology. Information Systems Research, 27(1), 197-

213. doi:10.1287/isre.2015.0611

Lee, J., Park, D. H., & Han, I. (2011). The different effects of online consumer reviews

on consumers' purchase intentions depending on trust in online shopping

malls. Internet Research, 21(2), 187-206. doi:10.1108/10662241111123766

Marinkovic, V., & Kalinic, Z. (2017). Antecedents of customer satisfaction in mobile

commerce. Online Information Review, 41(2), 138-154. doi:10.1108/oir-11-

2015-0364

McKnight, D. H., Carter, M., Thatcher, J. B., & Clay, P. F. (2011). Trust in a specific

technology. ACM Transactions on Management Information Systems, 2(2), 1-

25. doi:10.1145/1985347.1985353

McKnight, D. H., & Chervany, N. L. (2014). What Trust Means in E-Commerce Customer

Relationships: An Interdisciplinary Conceptual Typology. International Journal

of Electronic Commerce, 6(2), 35-59. doi:10.1080/10864415.2001.11044235

McKnight, D. H., Choudhury, V., & Kacmar, C. (2002). Developing and Validating Trust

Measures for e-Commerce: An Integrative Typology. Information Systems

Research, 13(3), 334-359. doi:10.1287/isre.13.3.334.81

Mou, J., Shin, D.-H., & Cohen, J. F. (2015). Trust and risk in consumer acceptance of e-

services. Electronic Commerce Research, 17(2), 255-288. doi:10.1007/s10660-

015-9205-4

Mukherjee, A., Arnott, D. C., & Nath, P. (2007). Role of electronic trust in online

retailing. European Journal of Marketing, 41(9/10), 1173-1202.

doi:10.1108/03090560710773390

Naovarat, S. (2015). Factors that affect intention to purchase of customer and the

role of trust as a moderator in E-Marketplace. Thammasat University,

Ref. code: 25615802037480DCY

68

Oliver, R. L. (1980). A Cognitive Model of the Antecedents and Consequences of

Satisfaction Decisions. Journal of Marketing Research, 17(4), 460-469.

doi:10.2307/3150499

Pappas, N. (2016). Marketing strategies, perceived risks, and consumer trust in online

buying behaviour. Journal of Retailing and Consumer Services, 29, 92-103.

doi:10.1016/j.jretconser.2015.11.007

Park, J., Lee, D., & Ahn, J. (2014). Risk-Focused E-Commerce Adoption Model: A Cross-

Country Study. Journal of Global Information Technology Management, 7(2),

6-30. doi:10.1080/1097198x.2004.10856370

Pavlou, & Fygenson. (2006). Understanding and Predicting Electronic Commerce

Adoption: An Extension of the Theory of Planned Behavior. MIS Quarterly,

30(1), 115-115. doi:10.2307/25148720

Pavlou, P. A. (2003). Consumer Acceptance of Electronic Commerce: Integrating Trust

and Risk with the Technology Acceptance Model. International Journal of

Electronic Commerce, 7(3), 101-134.

Pham, T. S. H., & Ahammad, M. F. (2017). Antecedents and consequences of online

customer satisfaction: A holistic process perspective. Technological

Forecasting and Social Change, 124, 332-342.

doi:10.1016/j.techfore.2017.04.003

Rouibah, K., Lowry, P. B., & Hwang, Y. (2016). The effects of perceived enjoyment and

perceived risks on trust formation and intentions to use online payment

systems: New perspectives from an Arab country. Electronic Commerce

Research and Applications, 19, 33-43. doi:10.1016/j.elerap.2016.07.001

Schreiber, J. B., Nora, A., Stage, F. K., Barlow, E. A., & King, J. (2006). Reporting

Structural Equation Modeling and Confirmatory Factor Analysis Results: A

Review. The Journal of Educational Research, 99(6), 323-338.

doi:10.3200/JOER.99.6.323-338

Sfenrianto, S., Wijaya, T., & Wang, G. (2018). Assessing the Buyer Trust and Satisfaction

Factors in the E-Marketplace. Journal of theoretical and applied electronic

commerce research, 13(2), 43-57. doi:10.4067/s0718-18762018000200105

Ref. code: 25615802037480DCY

69

Shang, D., & Wu, W. (2017). Understanding mobile shopping consumers’ continuance

intention. Industrial Management & Data Systems, 117(1), 213-227.

doi:10.1108/imds-02-2016-0052

Stouthuysen, K., Teunis, I., Reusen, E., & Slabbinck, H. (2018). Initial trust and

intentions to buy: The effect of vendor-specific guarantees, customer reviews

and the role of online shopping experience ☆. Electronic Commerce

Research and Applications, 27, 23-38. doi:10.1016/j.elerap.2017.11.002

Student. (1908). The Probable Error of Mean. Biometrika, 6(1), 1-25.

doi:10.1093/biomet/6.1.1

Visschers, V. H. M., & Siegrist, M. (2008). Exploring the Triangular Relationship between

Trust, Affect, and Risk Perception: A Review of the Literature. Risk

Management, 10(3), 156-167.

Xiao, Z., Wang, J. J., & Liu, Q. (2018). The impacts of final delivery solutions on e-

shopping usage behaviour. International Journal of Retail & Distribution

Management, 46(1), 2-20. doi:10.1108/ijrdm-03-2016-0036

Zhang, H., Lu, Y., Gupta, S., & Gao, P. (2015). Understanding group-buying websites

continuance. Internet Research, 25(5), 767-793. doi:10.1108/IntR-05-2014-0127

Conference Preceedings

Harrison McKnight, D., & Chervany, N. L. (2001, 2001//). Trust and Distrust Definitions:

One Bite at a Time. Paper presented at the Trust in Cyber-societies, Berlin,

Heidelberg.

Lesma, V. R. B., & Okada, H. (2012/10//). Feedback and trust-related factors of

consumer behavior in cross-border electronic commerce.

Mou, J., Cohen, J., Dou, Y., & Zhang, B. (2017). Predicting Buyers’ Repurchase

Intentions in Cross-Border E-Commerce: a Valence Framework Perspective,

Guimarães, Portugal.

Ref. code: 25615802037480DCY

70

Zhu, D., Neal, G. S. O., Lee, Z., & Chen, Y. (2009, 29-31 Aug. 2009). The Effect of Trust

and Perceived Risk on Consumers' Online Purchase Intention. Paper

presented at the 2009 International Conference on Computational Science

and Engineering.

Electronic Media

Amaral, T. P. M., Teresa Garín; Iñigo, García Herguera; Unda, Angel Valarezo. (2017).

Drivers and Barriers to cross-border e-commerce: evidence from Spanish

household.

DHL Express. (2016). THE 21ST CENTURY SPICE TRADEA GUIDE TO THE CROSS-BORDER

E-COMMERCE OPPORTUNITY. Retrieved from Press Release website:

http://www.dhl.com/content/dam/downloads/g0/press/publication/g0_dhl_e

xpress_cross_border_ecommerce_21st_century_spice_trade.pdf

International Post Corporation. (2018). IPC CROSS-BORDER E-COMMERCE SHOPPER

SURVEY. (January, 2018), 24. https://www.ipc.be/services/market-

research/cross-border-shopper-survey

Iwamoto, K. (2018). ASEAN to promote cross-border e-commerce with new

framework. NIKKEI ASIAN REVIEW.

Kenny, D. A. (2015). Measuring Model Fit. Retrieved from

http://www.davidakenny.net/cm/fit.htm

National Statistical Office of Thailand. (2019). Demography Population and Housing

Branch. Retrieved 10/01/2019

http://statbbi.nso.go.th/staticreport/page/sector/en/01.aspx

Nau, R. (2018, 1/06/2018). Regression diagnostics: testing the assumptions of linear

regression. Retrieved from http://people.duke.edu/~rnau/testing.htm

Priest, C. (2017). Commerce 2018 — Internet Trends Report. Retrieved from Medium

website: https://medium.com/@cameronpriest/commerce-2018-internet-

trends-report-cd5ee157bb0d

Ref. code: 25615802037480DCY

71

SEA Limited. (2017). FORM F-1 REGISTRATION STATEMENT SEA LIMITED. Washington

D.C., United States: Securities and Exchange Commission Retrieved from

https://www.sec.gov/Archives/edgar/data/1703399/000119312517291352/d36

3501df1.htm

Book

Ahmad, S., Zulkurnain, N., & Khairushalimi, F. (2016). Assessing the Validity and

Reliability of a Measurement Model in Structural Equation Modeling (SEM)

(Vol. 15).

Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.).

Hillsdale, N.J.: L. Erlbaum Associates.

Hair, J. F., Black, W. C., Babin, B. J., & Anderson, R. E. (2014). Multivariate data

analysis (Seventh edition, Pearson new international edition ed.): Harlow:

Pearson Education Limited.

Hooper, D., Coughlan, J., & R. Mullen, M. (2007). Structural Equation Modeling:

Guidelines for Determining Model Fit (Vol. 6).

Ref. code: 25615802037480DCY

APPENDICES

Ref. code: 25615802037480DCY

72

APPENDIX A

QUESTIONNAIRE

แบบสอบถามฉบบนเปนสวนหนงของการศกษาหลกสตรปรญญาโท สาขาวชาการบรหาร

ระบบสารสนเทศ (MIS) คณะพาณชยศาสตรและการบญช มหาวทยาลยธรรมศาสตร โดยม

วตถประสงค เพอเกบขอมลผบรโภครวมถงความคดเหนตาง ๆ เพอนำมาเปนขอมลในการศกษา และ

วจยในหวขอ “ความพงพอใจและความตงใจซอซำผานชองทางพาณชยอเลกทรอนกสขามชาตผานปจจยหลกดานความเชอมนและการรบรความเสยง”

ผวจยจงใครขอความอนเคราะหตอบแบบสอบถามนตามความเปนจรงและครบถวน เพอทำ

ใหผลการวจยนสมบรณตามความมงหมายของงานวจย โดยในสวนของขอมลสวนบคคลทไดรบจาก

การทำแบบสอบถาม ผวจยจะเกบรกษาเปนความลบอยางเครงครดและไมเปดเผยตอสาธารณชน ไมวากรณใด ๆ ทงสน และการวเคราะหและนำเสนอขอมลจะทำในภาพรวม ไมระบถงตวตนของ

ผตอบแบบสอบถามเทานน

แบบสอบถาม ประกอบไปดวย 3 สวน ดงน

สวนท 1: นยามและสาระสำคญ

สวนท 2: แบบสอบถามเกยวกบปจจยดานความเชอมน ความเสยง ความพงพอใจและความตงใจซอ

ซำผานชองทางพาณชยอเลกทรอนกสขามชาต

สวนท 3: ขอมลทวไปของผตอบแบบสอบถามและขอมลการซอสนคา

Ref. code: 25615802037480DCY

73

สวนท 1 คำนยามและสาระสำคญ

พาณชยอเลกทรอนกสขามชาต (Cross-border E-commerce) ตามแบบสอบถามตอไปน

หมายถงการซอขายสนคาผานชองทางออนไลนทมลกษณะการจดตงเปนรานคา (online store) หรอ

มลกษณะเปนพนทกลางในการใหบรการซอขายสนคาจากผผลตถงผบรโภค (marketplace) โดยการคาผานทางชองทางดงกลาวจะเกดขนผานหนาเวบไซตและ/หรอแอปพลเคชนทใหบรการบน

อปกรณสอสาร การซอสนคาผานชองทางดงกลาวสามารถสงคำสงซอผานทางหนาเวบไซตหรอ

แอปพลเคชนบนอปกรณเคล อนท (โทรศพทมอถอ แทบเลต ฯ) ผานการเช อมตออนเทอรเนตใน

ขณะใชงาน โดยการซอขายดงกลาวผใหบรการมการจดตงบรษทเพอดำเนนกจการดานพาณชยอเลกทรอนกสในประเทศตนเองและ/หรอในตางประเทศ และการซอขายสนคาตองมลกษณะการ

ขนสงสนคาขามพรมแดนระหวางประเทศ ในแบบสอบถามนจะใชคำวา "รานคาออนไลนขาม

พรมแดน" เพอใหงายตอความเขาใจ

การซ อสนคาผานรานคาพาณชยอเลกทรอนกสขามชาต (Cross-border e-commerce

purchase) หมายถงการสงซอสนคาผานรานคาพาณชยอเลกทรอนกสซงมการขนสงสนคาขามเขต

แดนประเทศระหวางตนทางผขายสนคาในประเทศหนงไปสปลายทางผรบในอกประเทศหนง ทงน

ผ ใหบรการรานคาออนไลนอาจจดต งบรษทอย ในประเทศไทยหรอไมอย ในประเทศไทยกได ตวอยางเชน การสงสนคาผานเวบไซต Amazon.com ซงผจำหนายสนคาอยในประเทศสหรฐอเมรกา

และผรบอยในประเทศไทย มการสงคำสงซอจากประเทศไทยผานหนาเวบไซตหรอแอปพลเคชนและ

ทำธรกรรมออนไลน จากนนมการขนสงสนคาจากรานคาใน Amazon ขามพรมแดนสผรบในประเทศไทย

หรอการสงซอสนคาผาน Lazada.com โดยผขายมสนคาเกบไวทประเทศจน มการสงสนคานำเขามาสประเทศไทย เปนตน

Ref. code: 25615802037480DCY

74

ทานเคยซอสนคาผานรานคาออนไลนทตองสงสนคาขามพรมแดนหรอไม (มการสงสนคานำเขามาใน

ประเทศไทย เปนสนคาจบตองได)

� เคย � ไมเคย (จบแบบสอบถาม)

สวนท 2 แบบสอบถามเกยวกบปจจยดานความเชอมน ความเสยง ความพงพอใจและความตงใจซอซำ

ผานชองทางพาณชยอเลกทรอนกสขามชาต

Ref. code: 25615802037480DCY

75

ปจจยดานความคาดหวงกอนซอสนคา

จากประสบการณการซอสนคาออนไลนขามพรมแดน ใหทานนกถงความรสกกอนการซอสนคาวาทาน

มความคาดหวงตอผขายจากรานคาขามพรมแดนเมอเทยบกบรานคาภายในประเทศในคำถามตอไปน

อยางใดโดยมระดบความเหนดงตอไปน 1 ไมเหนดวยอยางยง

2 ไมเหนดวย

3 เฉย ๆ

4 เหนดวย 5 เหนดวยอยางยง

คำถาม รานคาขาม

พรมแดน รานคาในประเทศ

ระดบความคดเหน 1 2 3 4 5 1 2 3 4 5

มความนาเชอถอ

จะใหความชวยเหลอเตมทเมอมขอสงสยหรอปญหา

จะทำตามขอตกลงในการใหบรการทใหไวตอกน

จะใหขอมลสนคาและบรการครบถวน

จะมปญหาเกยวกบตวสนคา (เชน คณภาพ ตำหนสนคา ใชงานไมได)

จะมความเสยงทางการเงนจากการซอสนคา (เชน คาใชจาย

ในการคนสนคา ถกโกง ภาษ)

จะมปญหาและความไมแนนอนจากการจดสงสนคา

โดยรวมแลวจะมความเสยงจากการซอสนคา

Ref. code: 25615802037480DCY

76

ปจจยดานความเชอมนและความเสยงหลงซอสนคา

จากประสบการณการซอสนคาออนไลนขามพรมแดน ใหทานนกถงความรสกหลงการซอสนคาวาทาน

มความคาดหวงตอผขายจากรานคาขามพรมแดนเมอเทยบกบรานคาภายในประเทศในคำถามตอไปน

อยางใดโดยมระดบความเหนดงตอไปน 1 ไมเหนดวยอยางยง

2 ไมเหนดวย

3 เฉย ๆ

4 เหนดวย 5 เหนดวยอยางยง



ระดบความคดเหน 1 2 3 4 5 1 2 3 4 5

มความนาเชอถอ

ใหความชวยเหลอเตมทเมอมขอสงสยหรอปญหา

ทำตามขอตกลงในการใหบรการทใหไวตอกน (ขอตกลงการใชงาน)

ใหขอมลสนคาและบรการครบถวน

รสกมปญหาเกยวกบตวสนคา (เชน คณภาพ ตำหนสนคา ใชงานไมได)

รสกมความเสยงทางการเงนจากการซอสนคา (เชน คาใชจายในการคนสนคา ถกโกง ภาษ)

รสกมปญหาและความไมแนนอนจากการจดสงสนคา (เชน

รอนานเกนไป สงลาชา วนจดสงไมแนนอน)

โดยรวมแลวมความเสยงจากการซอสนคา

Ref. code: 25615802037480DCY

77

ปจจยดานความการยนยนความคาดหวง ความพงพอใจ และความตงใจซอซำ

จากประสบการณการซอสนคาออนไลนขามพรมแดน ใหทานนกถงความรสกหลงการซอสนคาวาทาน

มความคาดหวงตอผขายในคำถามตอไปนอยางใดโดยมระดบความเหนดงตอไปน

1 ไมเหนดวยอยางยง 2 ไมเหนดวย

3 เฉย ๆ

4 เหนดวย

5 เหนดวยอยางยง



ระดบความคดเหน 1 2 3 4 5 1 2 3 4 5

โดยรวมแลวเปนไปตามทคาดหวงไว

โดยรวมแลวดกวาทคดไว

โดยรวมแลวไมมความเสยงอยางทคดไว

โดยรวมแลวนาเชอถออยางทคดไว

รสกพงพอใจตอประสบการณการซอสนคา

ไดรบประสบการณทดจากการซอสนคา

โดยรวมรสกประทบใจกบการซอสนคา

โดยรวมทานรสกอยางไรกบประสบการซอสนคา

จะซอสนคาออนไลนขามพรมแดนอกครงในอนาคต

วางแผนวาจะซอสนคาขามพรมแดนอกครงในอนาคต

ตงใจจะซอสนคาออนไลนขามพรมแดนในอนาคต

จะซอสนคาขามพรมแดนตอไป

Ref. code: 25615802037480DCY

78

สวนท 3: ขอมลทวไปของผตอบแบบสอบถามและขอมลการซอสนคา เพศ อายปจจบน _________

� ชาย � หญง

ระดบการศกษา

� ปรญญาตร � ปรญญาโท

� ปรญญาเอก � มธยมศกษา / ประกาศนยบตร

� อน ๆ

รายไดตอเดอน

� นอยกวา 25,000 � 25,001 – 50,000

� 50,001 – 75,000 � 75,001 – 100,000

� มากกวา 100,000

อาชพ

� พนกงานประจำบรษทเอกชน � นกเรยน / นกศกษา

� ขาราชการ / พนกงานรฐวสาหกจ � ธรกจสวนตว / อาชพอสระ

� อน ๆ โปรดระบ ____________________________________

ความถในการซอสนคาตอเดอน _________________

ประเภทสนคาทซอ

� เสอผา รองเทา เครองแตงกาย

� เครองสำอาง อปกรณเสรมความงาม อาหารเสรม

� คอมพวเตอร อปกรณอเลกทรอนกส

� เครองใชไฟฟา

� หนงสอ เครองใชสำนกงาน

� อน ๆ โปรดระบ ____________________________________

Ref. code: 25615802037480DCY

79

APPENDIX B

CONFIRMATORY FACTOR ANALYSIS DETAIL

B.1 Measurement Model and Structural Model

Figure B.1 Structural Model and CFA

B.2 Confirmatory Factor Analysis

Table B.1

Confirmatory Factor Analysis Result.

Observed

Variable

Latent

Construct

Std.

Estimate Estimate S.E. C.R. P-Value SMC

CBPRE CBTRE 0.117 0.135 0.072 1.880 0.060 CBTRP CBTRE 0.589 0.490 0.060 8.174 *** CBPRP CBTRP -0.300 -0.404 0.071 -5.715 *** CBPRP CBPRE 0.625 0.604 0.073 8.285 *** DCE CBPRE 0.049 0.050 0.068 0.733 0.463

Ref. code: 25615802037480DCY

80

Table B.1

Confirmatory Factor Analysis Result. (Cont’d)

Observed

Variable

Latent

Construct

Std.

Estimate Estimate S.E. C.R. P-Value SMC

DCE CBTRE -0.085 -0.100 0.076 -1.319 0.187 DCE CBTRP 0.585 0.827 0.107 7.700 *** DCE CBPRP -0.436 -0.457 0.073 -6.247 *** SAT DCE 0.802 0.778 0.061 12.806 *** CON SAT 0.510 0.496 0.076 6.503 ***

CBPRE4 CBPRE 0.674 1.000 0.447

CBPRE3 CBPRE 0.701 1.088 0.082 13.282 *** 0.394

CBPRE2 CBPRE 0.781 1.481 0.139 10.656 *** 0.425

CBPRE1 CBPRE 0.765 1.369 0.130 10.516 *** 0.326

CBTRP1 CBTRP 0.747 1.000 0.558

CBTRP2 CBTRP 0.678 1.084 0.085 12.780 *** 0.459

CBTRP3 CBTRP 0.635 1.182 0.098 12.030 *** 0.404

CBTRP4 CBTRP 0.702 1.135 0.086 13.198 *** 0.493

CBPRP1 CBPRP 0.654 1.000 0.414

CBPRP2 CBPRP 0.593 0.965 0.067 14.417 *** 0.356

CBPRP3 CBPRP 0.919 1.489 0.120 12.450 *** 0.847

CBPRP4 CBPRP 0.688 0.950 0.094 10.152 *** 0.445

CON1 CON 0.715 1.000 0.474

CON2 CON 0.712 1.033 0.082 12.582 *** 0.505

CON3 CON 0.674 0.930 0.077 12.049 *** 0.439

CON4 CON 0.748 1.038 0.080 13.001 *** 0.524

DCE4 DCE 0.773 1.000 0.598

DCE3 DCE 0.768 1.020 0.063 16.218 *** 0.589

DCE2 DCE 0.780 0.968 0.059 16.537 *** 0.609

DCE1 DCE 0.820 0.927 0.052 17.659 *** 0.672

Ref. code: 25615802037480DCY

81

Table B.1

Confirmatory Factor Analysis Result. (Cont’d)

SAT4 SAT 0.751 1.000 0.564

SAT3 SAT 0.698 0.998 0.076 13.090 *** 0.487

SAT2 SAT 0.618 0.901 0.081 11.182 *** 0.382

SAT1 SAT 0.730 1.070 0.079 13.596 *** 0.533

CBTRE4 CBTRE 0.669 1.000 0.447

CBTRE3 CBTRE 0.628 0.957 0.097 9.894 *** 0.488

CBTRE2 CBTRE 0.652 0.918 0.091 10.125 *** 0.617

CBTRE1 CBTRE 0.571 0.616 0.067 9.264 *** 0.586

Ref. code: 25615802037480DCY

82

APPENDIX C

Residual Variance

Table C.1

Residuals Variance Summary. Estimate S.E. C.R. P

z1 0.196 0.029 6.74 ***

z2 0.26 0.039 6.75 ***

z4 0.089 0.012 7.273 ***

z5 0.134 0.02 6.879 ***

z3 0.113 0.015 7.447 ***

z6 0.091 0.015 6.04 ***

z7 0.23 0.031 7.34 ***

e8 0.317 0.03 10.563 ***

e7 0.324 0.032 10.006 ***

e6 0.371 0.05 7.478 ***

e5 0.351 0.044 7.918 ***

e13 0.108 0.01 10.816 ***

e14 0.188 0.015 12.139 ***

e15 0.28 0.022 12.693 ***

e16 0.18 0.015 11.739 ***

e17 0.33 0.026 12.569 ***

e18 0.424 0.032 13.384 ***

e19 0.1 0.033 3.024 0.002

e20 0.247 0.026 9.498 ***

e25 0.231 0.021 11.096 ***

e26 0.251 0.023 11.159 ***

e27 0.252 0.021 11.878 ***

e28 0.206 0.02 10.3 ***

Ref. code: 25615802037480DCY

83

Table C.1

Residuals Variance Summary. (Cont’d) Estimate S.E. C.R. P

e12 0.183 0.015 12.064 ***

e11 0.197 0.017 11.756 ***

e10 0.163 0.014 11.547 ***

e9 0.114 0.01 10.973 ***

e24 0.198 0.019 10.606 ***

e23 0.269 0.023 11.685 ***

e22 0.337 0.028 12.181 ***

e21 0.257 0.023 11.026 ***

e4 0.242 0.022 10.778 ***

e3 0.276 0.024 11.579 ***

e2 0.223 0.02 11.122 ***

e1 0.154 0.012 12.424 ***

Ref. code: 25615802037480DCY

APPENDIX D MEASUREMENT MODEL CORRELATION

Table D.1 Measurement Model Correlation Matrix.

Variable 1 2 3 4 5 6 7 8 9 10 11 12 CBTRE1 1.000 CBTRE2 0.413 1.000 CBTRE3 0.349 0.408 1.000 CBTRE4 0.395 0.411 0.421 1.000 SAT1 0.100 0.153 0.102 0.003 1.000 SAT2 0.079 0.157 0.058 0.014 0.550 1.000 SAT3 0.050 0.115 0.107 0.067 0.498 0.559 1.000 SAT4 0.148 0.163 0.129 0.118 0.538 0.443 0.539 1.000 DCE1 0.223 0.239 0.221 0.179 0.492 0.421 0.431 0.498 1.000 DCE2 0.168 0.200 0.120 0.136 0.406 0.385 0.445 0.444 0.629 1.000 DCE3 0.163 0.207 0.195 0.129 0.477 0.345 0.402 0.461 0.631 0.673 1.000 DCE4 0.202 0.208 0.152 0.138 0.466 0.436 0.382 0.438 0.615 0.617 0.586 1.000 CON4 0.128 0.121 0.075 0.031 0.196 0.133 0.150 0.207 0.235 0.255 0.210 0.211

84

Ref. code: 25615802037480DCY

Table D.1 Measurement Model Correlation Matrix (Cont.) Variable 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 CON3 0.505 1.000 CON2 0.494 0.529 1.000 CON1 0.571 0.442 0.495 1.000 CBPRP4 -0.061 -0.076 -0.098 -0.180 1.000 CBPRP3 -0.115 -0.145 -0.172 -0.141 0.458 1.000 CBPRP2 -0.090 -0.090 -0.075 -0.143 0.398 0.537 1.000 CBPRP1 -0.028 -0.117 -0.093 -0.072 0.463 0.588 0.618 1.000 CBTRP4 0.139 0.080 0.168 0.133 -0.086 -0.104 -0.050 0.019 1.000 CBTRP3 0.079 0.012 0.121 0.041 -0.011 -0.003 0.064 0.117 0.481 1.000 CBTRP2 0.182 0.123 0.159 0.164 -0.090 -0.155 -0.149 -0.059 0.480 0.417 1.000 CBTRP1 0.179 0.070 0.128 0.165 -0.080 -0.198 -0.127 -0.082 0.519 0.456 0.521 1.000 CBPRE1 -0.062 -0.070 -0.039 -0.084 0.320 0.395 0.308 0.371 0.201 0.262 0.126 -0.006 1.000 CBPRE2 -0.057 -0.092 -0.059 -0.099 0.262 0.406 0.286 0.337 0.185 0.309 0.123 0.003 0.725 1.000 CBPRE3 -0.046 -0.095 -0.087 -0.131 0.332 0.310 0.279 0.345 0.089 0.168 0.049 0.025 0.531 0.575 1.000 CBPRE4 -0.055 -0.094 -0.066 -0.132 0.376 0.391 0.298 0.380 0.052 0.089 0.060 0.002 0.506 0.502 0.564 1.000

85

Ref. code: 25615802037480DCY

86

BIOGRAPHY

Name Mr. Natthakorn Khayaiyam Date of Birth June 28, 1989 Educational Attainment April 2012: Bachelor of Arts in English Work Position April 2017 - Present:

Data Scientist and Engineer Krungthai-AXA Life Insurance

Work Experiences October 2016 - March 2017: Business Support Analyst Deutsche Bank A.G. Bangkok January 2013 - May 2016: Data Specialist and Client Reporting Market Pulse International, Thailand.

Ref. code: 25615802037480DCY

consumer satisfaction and repurchase intention from cross

Documents