The effect of perceived web quality on e marketplace trust a case of vietnam

This study aims to explore: 1 Whether or not perceived Website quality whose components are technical quality, general content quality, specific content quality, appearance quality Aladw

VIETNAM NATIONAL UNIVERSITY, HANOI

VIETNAM JAPAN UNIVERSITY

TRAN ANH QUAN

THE EFFECT OF PERCEIVED WEB QUALITY ON E-MARKETPLACE TRUST:

Hanoi, 2018

ABSTRACT

Vietnam recently has witnessed a rapid rise in e-commerce Based on the logic that web quality is important in a self-service environment It is important to understand the impact of web quality to customers’ trust In this study, the effect of web quality

is tested in relation with trust in Vietnam e-commerce context The purpose of this study is to answer two questions, whether or not web quality has an impact on customers’ trust And in the sub-dimensions of web quality, which would have the biggest impact The finding is highly practical and applicable for Vietnamese firms

or foreign firms that are wanting to build an e-commerce business in Vietnam

ACKNOWLEDGEMENT

This study cannot be finished without the tremendous help from individuals and

institutions First of all, I would like to express my gratitude toward Vietnam Japan

University and JICA for providing unparalleled facilities and opportunities to study

and gather knowledge in incredible environments Furthermore, I would like to give

a big thanks to the two wonderful advisors Professor Motonari Tanabu and Doctor

Mai Anh for giving me invaluable knowledge on researching However, this thesis

can never be completed without the help from teachers from Vietnam Japan

University and friends

Last but not least, I would like to give my biggest gratefulness to the support from

my family, helping me to take the chance I got and supported me regard of the

decisions I made

Tran Anh Quan

TABLE OF CONTENTS

LIST OF TABLES

LIST OF FIGURES

INTRODUCTION 1

CHAPTER 1: Literature Review 4

1.1 E-marketplace 4

1.2 Web quality 4

1.3 Trust 5

1.4 Website adoption 7

1.5 The gap in literature of web quality on trust 9

CHAPTER 2: Research model and data collection 10

2.1 Research model and Hypotheses 10

2.1.1 Web quality 10

2.1.2 Technical quality 11

2.1.2 General Content quality 11

2.1.4 Specific content quality 12

2.1.5 Appearance quality 13

2.2 Data collection 13

2.2.1 Data collection method 13

2.2.2 Questionnaire design 14

2.2.3 Data collection process 16

2.2.4 Data analysis procedure 16

CHAPTER 3: Analysis and results 18

3.1 Data preparation 18

3.2 Characteristics of participants 18

3.3 Data Analysis and Results 20

3.3.1 Choosing acceptance level for criteria 20

3.3.2 Data analysis 20

3.3.3 Model testing 25

CHAPTER 4: DISCUSSION, LIMITATIONS AND FUTURE RESEARCH 27 4.1 Discussion 27

4.2 Limitations and future research 29

CONCLUSION 30

REFERENCES 31 APPENDICES

LIST OF TABLES

Table 3.1 Frequency table of the participants profile 19

Table 3.2 Summary of criteria for evaluating measurement models 20

Table 3.3 Descriptive statistics and reliability, correlation 23

Table 3.4 Regressions results 25

LIST OF FIGURES Figure 2.1 Proposed research model 10

Figure 3.1 Regression results 26

INTRODUCTION

Rationale and objective of this research

For the recent years, Vietnam has witnessed a big and steady rise in the number of people using internet On 4/4/2017, Mr Hoang Phuoc Thuan, Head of National Security, Department of Internet Security, Ministry of Public Security announced in World Security Conference 2017, Vietnam had 58 million people using the internet, increased 6% compared to 2016, placed itself in the top 30 countries in the world with the most IPv4 addresses (Minh, 2017)

And with the great increasing number of internet users, economists also predict the increasing usage of E-commerce Based on the analysis of Vietnam E-commerce Association (VECOM) in 2017, Vietnam’s E-commerce has risen by more than 25% and would continue this trend from 2018 to 2020 Furthermore, the revenue for E-commerce websites also increased by 35%

However, with big potential, comes great competition, many companies have to withdraw from Vietnam market due to fierce competition For example, Zalora, a big e-retailer who had business in many countries of South East Asia like Vietnam, Thailand, Indonesia, Philippines had to shut down in Vietnam market in 2016, before that, many players also disappeared from Vietnam market as well, for example: Deca.vn, Beyeu.com, Lingo.vn, etc (My, 2017)

There are of course numerous factors affecting the success of an E-commerce company, among them having great rate of usage is an important factor (DeLone & McLean, 2016) And within many factors that can have an effect on the usage of a service, trust is one of them (Lu, Yang, Chau, & Cao, 2011)

From long before and even until recently, the role of trust in business in general and E-commerce in particular has been a wanted subject for researchers, many research has been conducted in the name of this subject (Gefen, 2000; Hallikainen &

Laukkanen, 2018; Gefen, 2003) However, in Vietnam market, the lack of trust is still a fence that wards off customers from using E-commerce services, according to Azhar Bin Adnan -chief marketing officer of Vietnam for leading US based CMC

environment is still viable for Vietnamese researchers

To increase the level of trust from customers, from the firm’s perspective, one must understand the antecedents of trust in E-commerce (San-Martín & Camarero, 2012) proposed a model including 4 factors that affect trust in E-commerce shopping context, including web design as one of the factors Moreover, (Teo & Liu, 2007) also states that website assurance and Integration also have effect on trust And while they seem to be separated, web design, assurance and integration are features

of website quality (Aladwani & Palvia, 2002) In other word, Website quality has an influence on customer’s trust To prove this point further, Hwang & Kim (2007) did found perceived website quality has influence on customers’ trust after analyzing the data given by 325 students in north of the U.S Moreover, (Pennington, Wilcox,

& Grover, 2003) said that one of the main questions in building e-trust is how to build and design the website to enhance the trust of a potential online customer, which means researching on how Web quality would affect customer’s trust in Vietnam is important

This study aims to explore:

(1) Whether or not perceived Website quality whose components are technical quality, general content quality, specific content quality, appearance quality (Aladwani, 2006) influents customer’s Trust in Vietnam E-commerce context (2) In different components of Website quality, which has the most significant impact on customer’s trust Based on the result of this study, firms can analyze and customize their website to fit better with Vietnamese’s preference

Scope of the study

This study investigates the effect of perceived web quality on perceived marketplace trust In some countries, such as Jordan (Al-Debei, Akroush, & Ashouri, 2015), North America (Hwang & Kim, 2007) or Malaysia (Nilashi, Jannach, Ibrahim, Esfahani, & Ahmadi, 2016), web quality has significant positive impact on customers’ trust in E-commerce However, the similar case in Vietnam is unclear So, the geographical scope of this study will be in Vietnam The data of this study is collected in May 2018, concluded the time scope for this study is within the year of 2018 itself

e-Structure of the study

This study will be presented as follow: In the beginning of the study, the author will introduce the rationale, research questions and scope of the study In chapter 1, literatures support for the study’s theoretical foundation will be promptly presented Next, research model and hypotheses will be followed Chapter 2 will be about research methodology Chapter 3 will have data analyzing results presented, Chapter 4 will have discussion, limitations and implication Lastly, the conclusion

of this study will be presented

CHAPTER 1: LITERATURE REVIEW

1.1 E-marketplace

E-marketplace (EMP) is an online environment on the internet where different buyers and sellers play as market players to exchange goods and services by electronic means (Giovanoli, Pulikal, & Grivas, 2014) E-marketplace is often divided into four categories, “private marketplace”, “public marketplace”,

“consortia marketplace” and “community marketplace” (Sundareswaran, Squicciarini, & Lin, 2012) Private marketplaces are usually owned by one company, with function is to connect a set of limited pre-qualified buyers or sellers Public marketplace, also called independent marketplace or open marketplace, is the type of marketplace which there is little qualification requirements Public marketplaces are owned by a neutral third party, such as Lazada group or Ebay Consortia marketplace is owned by several large buyers and sellers or joined industry leaders Convisint, originally owned by Ford, GM and DaimlerChrysler, is probably a well-known example of consortia marketplace These marketplaces are usually open to other equity partners after setting up their initial foundation (Wang

& Archer, 2007) And lastly, there is community marketplace, which is owned by multiple players from different background

There are three main functions of EMP, matching buyers and sellers, facilitate exchanges of good, information, services, associate payment and providing adequate infrastructure to keep the market function efficiently (Giovanoli et al., 2014), with the first being the major role for EMP, while the second and the third are also very important to keep EMP run smoothly

1.2 Web quality

Perceived web quality refers to the quality and overall performance of an online shopping web site, and it is a measure of the extent to which the web site design and processes are simple, smooth, reliable, and effective (Petter, DeLone, & McLean, 2008) In the area of IS, there has been quite a number of research about Web

Quality throughout the years, most of them talk about usability attributes and design

of websites Bell & Tang (1998) mentioned factors in technical quality to web design like access to the web, navigation, graphics, structure, unique feature, user friendliness and usefulness On another hand, Johnson & Misic (1999), added to the collection their findings, including speed, navigation, wording, color, style, and especially, contact information (email, phone number, address and so on) In the same year, Rose et al added download speed, web interface search function and security The year 2000 is a big year for web quality research, in this year, Liu & Arnett, 2000 and Wan (2000) added a lot of categories and dimensions for Web quality, including identified information, service quality, system use, playfulness, system design quality, information, friendliness, responsiveness, and reliability Furthermore, Huizingh (2000) divided Web quality into 2 big dimensions: content and design Olsina & Rossi (2002) proposed and used a quantitative way to assess web quality called Website Quality Evaluation Method (QEM), in which the characteristics of a website would be evaluate are usability, functionality, reliability, efficiency, and derived sub characteristics However, all the above studies show different attributes separately, without any unification To solve that problem, in

2002, Aladwani and Palvia unified the above characteristics of web quality that were found by peer into a holistic view, introducing the perceived web quality with

4 main dimensions: Technical quality, general content quality, service content (or specific content) quality and design quality These dimensions accounted for 67%

of the variance in perceived website quality (S Kim & Stoel, 2004)

1.3 Trust

Same as any traditional business, behind the interface of a website of E-commerce, there is an entity that run the business, and trust is crucial in buyer-seller relationships (Gefen, 2003) Because there are several infamous harmful behaviors

of bad E-commerce sites, such as conveying inaccurate information or unauthorized tracking of transaction, users tend to avoid E-vendors that they do not trust (Reicheld, F.F dan Schefter, 2000) In short, trust is vital for a business’ success,

especially in online business since the uncertainty in this environment is higher than

in offline business (Y Chen & Barnes, 2007) Throughout the history, trust has been conceptualized by many researchers, D Gefen (2003) made a summary of previous conceptualizations along with the measurements to measure the operation

of those constructs In which until 2002, there has been 43 conceptualizations of trust in research studys In which trust can be understood as the “willingness of a party to be vulnerable to the actions of another party based on the expectation that the other will perform a particular action important to the trustor, irrespective of the ability to monitor or control that other party” (Mayer, Davis, & Schoorman, 1995)

It also can be understood as a strong defined aspect of many economics interaction, implying a thinking that a certain result will occur Moreover, trust also reduces risk that believer will fall victim to an opportunistic behavior (Warkentin, Gefen, Pavlou,

& Rose, 2002) After 2003, the definition of trust is hardly revisited Schoorman, Mayer, & Davis (2007) revisit the model they made for trust, commenting and making way for future research on the subject In this conceptualization, Mayer proposed Factors of Perceived Trustworthiness including three factors: Ability, Integrity and Benevolence And in the field of E-commerce, Gefen (2002) proposed and validated the same conceptualization for E-trust, stating that e-trust is a multi-dimensional construct combining specific beliefs (sometimes labeled as trustworthiness) and further investigation is meaningful

Ability, also referred to be “competence” (Mcknight, D H., V Choudury., 2002), is the skills and knowledge of service’s provider to provide a stable and desirable performance that meet customers’ need (Yu, Pay Ling; Balaji, M.S.; Khong, 2011)

In E-commerce context, ability is the belief that a player will have enough understand of the market they are in and what kind of the product/service that they are providing, or in another word, ability contain a belief that the other party can promptly do what is expected from them (Mcknight, D H., V Choudury., 2002) Companies with good ability can reduce the uncertainty that are associated with e-commerce (Gefen & Straub, 2004) Ability is also domain specific (Mayer et al.,

1995), meaning that a company that is good at this domain doesn’t necessary good

at other domains An online apparel store may be really good at customer service, but might not as equally good at general knowledge as a book store is

Integrity, according to Mayer et al., (1995), is the perception from the trustor that, the trustee will adhere to a set of principles that the trustor (as an individual) find acceptable It is understood that integrity is the trustee party being honest, fulfil their promises, act ethically and maintain commitment, it convinces individuals that the exchange that they will engage in will be fulfilled trustfully, and therefore, like ability, integrity reduces the uncertainty involved in e-commerce (Gefen & Straub, 2004) So it can be concluded that, in e-commerce context, integrity reflect a customer’s belief in which the orders they made will be deliver promptly as agreed, their information will be secured through all the time using the trustee’s service Benevolence is the perception that the trustee that provide goods and services will genuinely do good for their customers, truly care for the customers’ welfare beside their profit motivation According to McKnight (2002), benevolence reflects the good and services providers’ caring and motivation to act in customer’s interest Benevolence has its root deep in the relationship of sellers and buyers since it reduces the perceived uncertainty and risk that come with opportunistic behaviors (Ba & Pavlou, 2002) In E-commerce context, benevolence, it can be understood that the online vendor does not take advantage of their customers information, and does not act unscrupulously toward them Benevolence is also a relationship between online vendors and their customers, not reflecting the vendors’ kindness and good intention toward everyone (McKnight & Chervany, 2001)

1.4 Website adoption

Web quality on trust is placed in the website adoption site of research, which usually involving purchase intention, trust, customer behavior by using website as a tool for shopping Gefen et al., (2003) made an integrative model for trust in online shopping, combining the knowledge of previous researchers such as Straub (2000),

Liao & Cheung (2001), Shim, Shin, & Nottingham (2002) and Ranganathan & Ganapathy (2002) In which they examine factors from both individual side and business side, such as media richness, convenience, ease of use, system quality, service quality, transaction speed, information content, design, security, and privacy

on firm’s side and user experience, involvement, beliefs, IT education, familiarity

on consumer’s side Aladwani (2006) used the four factors web quality to test of the link between web site quality and forward enterprise integration with web consumers Hwang & Kim (2007) used Aladwani’s model to test effects of perceived Web quality with service contents on enjoyment, anxiety, and e-trust of self-service systems Kuan, Bock, & Vathanophas (2008) tested the effect of website quality on customer initial purchase and continued purchase at e-commerce websites Qutaishat (2012) examined whether or not website quality would attract people to use E-government services in Jordan Lee & Levy (2014), examined the effect of information quality on trust in e-government system Al-Debei, Akroush,

& Ashouri (2015) tested the effect of trust, perceived benefits, and perceived web quality on consumer attitude toward online shopping Nilashi (2016) used a different conceptualization of web quality, including information quality, usability and service interaction to test its relation to trust Mcknight (2017) expanded the literature, not online examining the relation between information quality, system quality on trust, but also on distrust And Hallikainen & Laukkanen (2018) investigated the effect of national culture on e-commerce’s trust, meaning e-commerce trust is still an excellent topic for researchers Furthermore, in the context

of Vietnam, there has been not been many research related to this topic For website quality, there have been research, mostly on the relationship between web quality and loyalty, buying intention and satisfaction (Le & Pham, 2015; Võ Lý Thị Nhị, 2014) And they are in banking and tourist industries There has not yet any study

on the relationship between web quality and customers’ trust in E-commerce context in Vietnam Thus, this study may prove to be the pioneer of this particular niche in Vietnam research environment

1.5 The gap in literature of web quality on trust

The above researchers usually take Aladwani’s conceptualization of Web quality as

a whole and to see how influential it is on different aspects, including anxiety, enjoyment and e-trust (Hwang & Kim, 2007), electronic word of mouth, trust and intention toward online purchase (Al-Debei et al., 2015) It can be seen that they approached figuring out the relationship between web quality and trust as a whole, without focusing to see within the four dimensions of web quality, each of them will have how much of an influence on trust On the other hand, research focusing

on the four dimensions of web quality have touched different aspects such as attitude toward the website and intention to purchase (Aladwani, 2006), intention to use E-government services (Qutaishat, 2012) or initial purchase versus continued purchase (Kuan et al., 2008) In addition, both Qutaishat and Kuan’s research do not include appearance quality as one of the factors and added service quality as another factor To sum up, the previous literatures have not addressed the relationship between perceived web quality and trust using Aladwani’s four dimensions model, researchers use different combination aspects of the model, sometimes information and system quality (McKnight, Lankton, Nicolaou, & Price, 2017), another time information system only (Lee & Levy, 2014) For Vietnam context, there has not been any research on this particular topic This study will address the above-mentioned issue

CHAPTER 2: RESEARCH MODEL AND DATA COLLECTION

2.1 Research model and Hypotheses

The main purpose of this study is to explore whether or not the perceived web quality would have anyhow influent the trust in e-commerce websites in Vietnam context As mentioned earlier in the first chapter, trust is still a problem fencing off people using e-commerce services Furthermore, within the four dimensions of perceived web quality proposed by Aladwani, which will have the most significant effect on customers trust in Vietnam, hence, the research model is proposed as below, along with relationship and hypotheses:

Figure 2.1 Proposed research model (Adapted from Aladwani, 2006; Heli, 2018)

2.1.1 Web quality

In this study, Web quality is the combination of the four different dimensions: technical quality, general content quality, specific content quality, and appearance quality For recent years businesses, a wide reach of IT (Information technology) has affected the retail industry as a whole Many small e-retailers has been

established or shut down, even websites that help people to open their own commerce websites are established (shopify.com) which currently has more than

e-3600 stores operating every day There has been research showing that web quality has direct impact on trust (Al-Debei et al., 2015; Nilashi et al., 2016) Based on their findings, a following hypothesis is proposed:

H1: Web quality will have positive impact on customer’s trust in E-marketplace site

2.1.2 Technical quality

For anyone using the internet, they can see how companies and users interact through different websites On a website, information and medias are shown to attract customers, but to increase the chance of getting opportunity for interactive business transaction, technical characteristics of a website, including ease of navigation, security, speed, interactive function and so on, become important as an aspect of website quality (Aladwani & Palvia, 2002) And since E-commerce is based mostly on online environment, the quality of those characteristics is also need

to be good The technical dimension is related to how the content and services are assembled and made available on a website (Rocha, 2012) Throughout the years, the advancement in technical is undeniable, however, the old findings of previous researchers are still remained true Gao & Bai (2014) found that ease of use (including ease of navigation and speed) has impact on intention to use internet of things Bonsón Ponte, Carvajal-Trujillo, & Escobar-Rodríguez (2015) proved that perceived security has influence on customer’s trust in online purchasing Hence forth, websites with good quality can deem to be worthier in the eyes of customers comparing to other websites with inferior technical quality Thus, a hypothesis is proposed

H2: Technical quality will have positive impact on customer’s trust in marketplace site

E-2.1.2 General Content quality

From decades ago, studies have shown that websites with sub-standard general

content quality (completeness of content, accuracy, clarity of content, concise and usefulness of content) would be a waste of resources and decrease the chance of getting potential customers (Liu and Arnett, 2000; Misic and Johnson, 1999) One

of the main things that people will looks at when shopping online is the information, from product detail’s information to retailer’s information, and even other facts from the websites and products Content deliver information, and if the information does not meet the expectation of customers, then that will signal a negative reaction from people visiting those websites It’s very hard to gain customer’s trust if the website provides them with ambiguous, unusable, or inaccurate information And even in recent studies, content quality is still contributing a great influence on attracting customers that potentially will deliver transactions for companies Calisir

et al., (2014) examined the effect of content quality on the intention to use a based learning system In which content quality have significant impact on perceived usefulness which is the strongest predictor of behavioral intention to use that system Nilashi et al., (2016) found out that Web quality (with information quality) has positive impact on trust on Amazon.com and Lazada.com McKnight et al., 2017), Lee & Levy (2014) proved that information quality have significant impact on trust from customers Using Aladwani and Palvia’s model in 2002, in which to create a website with favorable customer’s intention the information must

web-be useful, complete, clear, current, concise, and accurate, the next hypothesis is proposed, arguing that general content quality would have a positive influence on e-marketplace trust

H3: General content quality will have positive impact on customer’s trust in E- marketplace site

2.1.4 Specific content quality

Similar to general content quality, specific content quality also plays a big part in attracting potential customers However, while general content quality focuses more on the quality of contents that are shown on website, specific content focuses

on a group of specific information that should be found on e-commerce websites,

such as firm contact information, firm general information, products/services details, customers’ policies, customer support (Aladwani, 2006) This is undoubtedly correct in many senses because within a mass of information put on by different seller/resellers, it should be more trustworthy that you can easily find information that could reduce your risk on those sites So, a hypothesis was made, controverting that specific content quality would have positive influence on customer’s trust in e-commerce marketplace sites

H4: Specific content quality will have positive impact on customer’s trust in E- marketplace site

2.1.5 Appearance quality

The final component of Website quality (web quality for short) is the appearance, which refers to characteristics such as attractiveness, organization, proper use of fonts, colors, and proper use of multimedia (Aladwani, 2006) Bart et al., (2005) proven that A visually appealing website demonstrates the e-vendor's capability and professionalism, which would engender online trust Naturally, it’s just hard for users to continue looking for information, or simply just keeping attention on a webpage with unfriendly colors, or with bad or inconsistent style E-commerce sites, aside from focusing on technical quality only, should pay attention to appearance also to get higher chance to successfully accomplish their goals, whether it is to bring more customers to their site, generate more transaction or to initiate online trust from customers Hence, the fourth hypothesis is proposed as follow

H5: Appearance quality will have positive impact on customer’s trust in marketplace site

E-2.2 Data collection

2.2.1 Data collection method

There are two main different data collection method, qualitative and quantitative For qualitative data collection method, the data collected usually is not presented in

numeral form, but through another means like form of words or picture, because it

is used to explain the complex nature and to create and find new concept simultaneously with the data collected On the other hand, quantitative data is formed by number collected by precise measurement The data is used to capture the detail found in empirical social world, and to analyze and express what is found

in the number (Neuman, 2014)

In this study, a questionnaire survey of quantitative method was chosen due to several reasons First, the study is used to test the relationship between different variables (web quality and trust), second, it is effective for collecting a large amount

of data And lastly, the use of questionnaire is less time consuming and more economic efficient

2.2.2 Questionnaire design

In this section, the formation of questionnaire is introduced The questionnaire for this study is developed based on previous studies’ measurement items with some modifications to fit with the scale of the study and the context of Vietnam The questionnaire is composed of two sections with six different parts: questions about participants’ general anonymous information, questions measuring technical quality, questions measuring general information quality, questions measuring specific content quality, question measuring appearance quality and questions measuring trust There is a small description for each part so participants can understand what they will be asked Each part of the questionnaire will be explained in detail

The questionnaire was made using google form and is an online questionnaire At the beginning of the questionnaire, an introduction is written so that participants are informed about what the research is about, the purpose of this study and information

of the author for any question given by participants

In the first section, there are questions asking for participants’ general information such as their age, gender, employment status, experience with e-marketplace and frequency of shopping online

Next, the main questions of the questionnaire are introduced In each part, there is a description telling what participants need to do in order to complete the part, the measurement items are measured using five-point Likert scale, with the range from (1) “Strongly disagree” to (5) “Strongly agree” Likert scale is a widely use measurement scale (Joshi, Kale, Chandel, & Pal, 2015), it is easy for participants to understand and is very straight forward to build and administer The five-point scale was chosen due to the ease of use and clear and simple user interface In the second section, first, participants are asked about the technical quality of their mainly use e-marketplace sites There were six questions in this part Question 1, 2 and 3 measure facility, availability and speed, question 3, 4, 5 and 6 measure automatic functionality and language support Continuedly, participants moved to another part, answering questions about general information quality In this part, there are five questions, they are used to measure information usefulness, completeness, clearness, timeliness and accuracy Similar to the second part, there are five questions in the third part of the second section Question 1 and 2 in this part are used to ask participants about the readiness of firm’s general and contact information Question

3 is used to ask about the information of products and services Question 4 and 5 are used to ask about the adequate and readiness of information related to customers’ policies and support In the followed part, participants are asked by five questions measuring websites’ appearance quality The first question in this part is used to ask participants about the attractiveness of websites Next, participants are asked about how good the websites’ layout are in the second question Question 3 and 4 are set

to measure the proper use of font and color And the last question is about multimedia proper uses The last part of the second section is totally focused on trust There are six questions related to benevolence, integrity and ability of vendors that participants would have to choose from

In conclusion, a 31-item online questionnaire (without questions about participants’ general information) was used to gather data from the participants (Appendix 1) There are total 22 items measuring technical quality, general content quality,

specific content quality, appearance quality (Aladwani & Palvia, 2002) and 9 items measuring trust through perceived integrity, benevolence and ability (Schoorman et al., 2007), adapted from the research of Hwang (2006)

2.2.3 Data collection process

The Vietnamese version of questionnaire’s URL was uploaded on several Facebook university groups in Hanoi, mainly universities study business and economics The target sample for this study was mainly students and young employed people The reasons for choosing this type of sample are due to the internet and online purchase familiarity of young people The participants in this survey are people who are capable of using the internet through PC or mobile phone, who have previously purchased through the internet and used e-commerce websites

Data collection was done in 3 weeks, there were around 67 participants attended the survey in each week In those three weeks, participants who joined the survey get a return present, encouraging people to take part in the survey, also as a gratitude toward people who take their time to join the program

A total 199 responses were collected via online mean and all 199 responses are adequate for further analysis

2.2.4 Data analysis procedure

Data analysis, according to Sekaran (2013) has three main objectives The first objective is to getting the feel of the collected data, using descriptive statistics The second objective is to assessing whether or not the data have good quality by using measurement models And the third objective is to test hypotheses proposed for the research To meet the objective, data collected in this study will be analyzed following these steps:

The first thing to do after collecting the data is to prepare the data, also called data preparation step In this step, the data will be cleaned, removing outliners, missing observations and other discrepancies (Trochim, 2002) After that, demographic and

characteristics of participants, along with descriptive statistics analysis will be carried out to help getting the feel of the data After preparing and running descriptive test for the data, measurement models and analysis will be tested using SPSS 25 All proposed hypotheses and research questions will be analyzed, using methods such as EFA loadings, reliability test, correlation test and regression The details of those analyses will be presented and discussed later in this study

CHAPTER 3: ANALYSIS AND RESULTS

Based on what discussed in the last part of Chapter 2, Chapter 3 will be composed

of two parts, data preparation and data analysis While data preparation focuses on cleaning the data, data analysis will focus on testing the adequacy of data and measure the model proposed

3.1 Data preparation

After receiving the online results, raw data was imported into excel file Since there

is no missing variable due to the nature of google form questionnaire with “require answer option”, the next step is to reduce the outliers To detect outliers, an explore statistic test was conducted in SPSS 25 (Appendix 2) After carefully observe the outliers detected, answer number 89 and 187 were removed from the data set, showing sights that the participants purposely filled the questionnaire not wholeheartedly

3.2 Characteristics of participants

Getting the feel for the data is important before officially conduct analysis for hypotheses testing (Sekaran, 2013) In this part, descriptive statistics will be carried out and data normality will be checked, with in mind that when analyzing the results gotten from the data, profile of participants should be taken account for

As can be seen in table 3.1, the data is consisted of 135 female participants (68.53%) and 62 males (31.47%) Furthermore, the current residency of the people responding to the questionnaire is mainly Hanoi city with 182 participants (92.39%),

10 people are currently working in Ho Chi Minh city (5.08%) and the rest are distributed in different northern and central provinces, such as Nam Dinh, Quang Ninh, Thai Binh, Thanh Hoa and Da Nang For the occupation of the participants, most of them are in different universities studying (68.02%), whether it is

ungraduated or graduated programs Among working people, 95.23% are employed, while 4.76% (3 people) are self-employed and have their own business

Table 3.1 Frequency table of the participants profile

Number of participants Percentage

5 Frequency of e-commerce shopping

6 Purchase within 6 months

Source: Author’s analysis of survey data

With the profile statistics described, it can be seen that for young people in Vietnam, most of them would order 1-5 products from e-commerce sites per year (44.16%) Second type will buy more than 10 orders per year (26.9%) It can be understood that normally young people in Vietnam would buy very little goods online due to the nature of uncertainty of product quality, since people would buy from 0 to 5 products per year took 56.85%, larger than people would buy from 6 to more than

10 products (43.14%) However, there are more people who buys more than 10 products per year than the people that buy from 6-10 products per year Which

means if E-commerce sites in Vietnam can be trustworthy and the product quality are good, Vietnamese people wouldn’t hesitate to buy things online

3.3 Data Analysis and Results

3.3.1 Choosing acceptance level for criteria

Before conducting analysis, it is essential to choose the acceptance level of criteria, these criteria and their acceptance level will be used to evaluate the measurement in the next section Table 3.2 will summary the criteria used and their acceptance level

Table 3.2 Summary of criteria for evaluating measurement models

Criteria

Abbreviation Acceptable level Source Kaiser-Meyer-

Factor loading

0.5 or greater are considered very significant

(Hair, Black, Babin,

& Anderson, 2010) Cronbach alpha

be conducted to test correlation coefficient between variables, making sure the scale

is reliable enough for linear regression analysis

Before conducting factor analysis, it is essential to know whether or not your data is suited to do so To answer this question, a Kaiser-Meyer-Olkin (KMO) test is

conducted via the mean of SPSS KMO measure of sampling adequacy is 0.852 (Appendix 3), indicating that the sample is adequate for further analysis (Kaiser, 1974) Next, the test for exploratory factor analysis was carried out by SPSS The goal of this analysis was to discover whether or not there is any cross factor loading

or inadequate item for the five factors of web quality and trust by using Principal Component Analysis extraction method and Varimax with Kaiser normalization rotation method There is no “right” suppress point, it is based on judgement, prior study or the purpose of the study In this study, the purpose is to find the most significant loading explaining factors So, the suppress point 0.5 was chosen (Hair

et al., 2010)

Table 3.3 shows that there is no cross-factor between the seven factors tested, however duo to previous test, GCQ1, GCQ2, AQ5 didn’t make through the suppress point (Appendix 4), so those observe variables will be dropped out of further analysis With GCQ1, GCQ2 and AQ5 dropped, KMO value reduced to 0.846 (Table 3.3), meaning the data is still suitable for analysis

After factor loading analysis, a reliability test to measure internal consistency for different scales was conducted to see the Cronbach alpha of the whole 28 items left First, the initial 19 items for web quality were analyzed as a whole Then, each of the different components was also analyzed As seen in table 3.3, the Cronbach alpha for web quality is 0.879, the Cronbach alpha for the next 4 components of web quality is 0.813 (technical quality), 0.729 (general content quality), 0.815 (specific content quality) and 0.802 (appearance quality) respectively

Second, the same analysis for 9 items of trust are also conducted, resulting in a 0.862 Cronbach alpha for trust items In which benevolence’s Cronbach alpha is 0.845, integrity’s is 0.847, ability’s is 0.726 There is no item whose corrected item-total correlations is smaller than 0.4

The next step is to test how the variable correlate with each other, hence a correlation analysis was done by SPSS, resulting in all independent variables (web

quality and its sub-dimensions: technical quality, general content quality, specific content quality and appearance quality) have low medium moderation correlation with trust and its sub-dimensions Furthermore, both web quality and trust all have strong correlation with its sub-dimensions, in which integrity has the strongest correlation with trust (0.845) and specific content quality has the strongest correlation with web quality (0.807)

***p < 001; **p < 01; *p < 05; N= 197; Number in () is Cronbach Alpha; KMO = 0.846

Source: Summary results from SPSS 25 output for this study