Perceived customer value of vietnam airlines online ticketing master project in business administration

The study of Chen and Dubinsky 2003, which suggested a preliminary investigation on perceived customer value in e-commerce, got some interesting findings.. Research Objectives This rese

NGUYEN THE HOANG

PERCEIVED CUSTOMER VALUE

OF VIETNAM AIRLINES ONLINE TICKETING

MASTER PROJECT MASTER IN BUSINESS ADMINISTRATION

(PART-TIME)

Tutor’s Name: Dr NGUYEN Duc Tri

Ho Chi Minh City

(2010)

contributions of others are involved, every effort is made to indicate this clearly, with due reference to the literature, and acknowledgement

of collaborative research and discussions

The work was done under the guidance of Dr NGUYEN DUC TRI, under the framework of the the Master program in Business Management – The join Masters Program between Ho Chi Minh City Open University (Vietnam) and Solvay Brussels School (Belgium) The work was completed in December 2010, in Ho Chi Minh City - Vietnam

(candidate’s name and signature)

NGUYEN THE HOANG

In my capacity as supervisor of the candidate’s thesis, I

certify that the above statements are true to the best of

my knowledge

(guide’s name and signature)

Dr NGUYEN DUC TRI

Date: ……

This thesis could not have been done without the contribution and encouragement from many people…

First of all, I would like to express my profound gratitude and great appreciation to my supervisor Dr NGUYEN DUC TRI for his valuable guidance, advices and recommendations throughout the research study

I would like to express my great appreciation to the staff of the Passenger Sales & Marketing Department, Southern Regional Office of Vietnam Airlines, who I had chance to work with during my short internship at the early stage of this study Deep appreciation and thanks are reserved to Mr HUNG – Supervisor at the Department – for his fruitful discussions as well as providing me with the necessary information Special thanks also to

be sent to Mr KHANH – Head of Human Resource Department of Southern Regional Office – for facilitating all the administrative arrangement for my internship, and Mr SON – Head of the Sales and Marketing Department – for accepting me as a trainee in his Department

I would like to express my sincere thanks to my classmates MBAVB3 who have shared with me their impressive knowledge and skills during the challenging courses of the MBA program I do specially want to express my loving thanks to all the members in my dear group (4A+4B) who were always besides me whenever the “team work” assignments have been carried out

My greatest debt of gratitude is to my wife, Thao, who was sharing with me all the difficulty occurred during this MBA study My MBA study could not be completed without her moral encouragement and support

Finally, I would like to thank the members of the Examination Committee for taking time and giving valuable comments to improve this study

HCM City, December 2010

NGUYEN THE HOANG

ticketing channels to their direct channels such as website which is considered as a prominent marketing channel and the most cost effective for Airlines The biggest Airline

in Vietnam – Vietnam Airlines – has also followed this strategy by recently launching ticket selling via its website

However, the emergence of online booking can lead to significant changes in consumers’ references and behavior Therefore better understanding key factors influencing on “perceived customer value”, which has been found to be a powerful predictor of purchase intention, becomes a crucial issue for e-marketers

In such context, this research follows the study of Chen and Dubinsky (2003) A more simplistic model was suggested to indentify key antecedents of perceived customer value in B-to-C online ticketing The model was empirically tested with Vietnam Airlines’ e-clientele This choice has double purposes Firstly, worthy implications deducted from the results expected can help better implement tactics for further development of Vietnam Airlines’ e-commerce Secondly, the investigation can give further empirical support to better ascertain the validation of the original model suggested by Chen and Dubinsky

A quantitative research was carried out Structured questionnaire was administered

to collect data from Vietnam Airlines’ e-customers by both methods: “online questionnaire” and “personally administered questionnaire” surveys The analysis techniques which include Reliability, Factor and Multiple Regression analyses were used

to valid the construct measurement and test the hypotheses suggested Revision of research questions, hypotheses, and research model in the case of investigation was carried out

The results obtained by the research shows that “valence of experience” and

“perceived product quality” are predictors of “perceived customer value” They are positively and significantly related to the “perceived customer value” Furthermore, it is also shown that the factors, which are related to the website’s performance, including

“relevant information”, “ease of use”, and “customer service” are the predictors of customers’ “valence of experience”

………

………

………

………

………

………

………

………

………

………

………

………

………

………

………

1 CHAPTER 1: INTRODUCTION

1.1 Research background

Facing with high fixed costs, airlines strive to find an efficient and effective distribution strategy of tickets (Toh and Raven, 2003) Distribution through brick-and- mortar travel agencies is expensive because it requires airlines to pay not only commissions to travel agencies but also transaction fees for the computerized reservation systems (Bilotkach and Pejcinovska, 2008) Working with online travel agencies (OTAs), such as Expedia and Travelocity, offers airlines a broader consumer base than if they distributed solely through their brand.com websites which can help to save airlines distribution costs Furthermore, previous studies showed that there was a contention of shares amongst the channels of airlines For example, during 2000-2001 in Taiwan, the market share data showed that, traditional tour agents have lost at least half of their customers to the ticketing websites (Shon et al., 2003) These show many challenges for sales channel management that directly affect the profits of airlines

Airlines’ websites provide a distribution channel linking the Airlines more closely and directly to their potential customers Distribution through own website is generally regarded as the most cost effective for Airlines (Lubbe, 2007).

To consumers, the benefits of the Internet are tremendous As an alternative channel, web shopping is convenient and time saving Consumers can compare prices and products or services’ features across suppliers by using rich and free information available Though, previous works on web-shopping, has raised the e-suppliers’ concerns about the low purchasing rate and moderate satisfaction of online shoppers (Moon and Frei, 2000) Customers now have much more bargaining power, lower switching costs, and high number of choices Thus understanding what leads to online customers’ purchase intentions has become an important research topic (Barsh et al., 2000)

“Perceived customer value” has gained much attention from maketers and researchers because of its important role in predicting purchase behaviour (Cronin et al., 2000) From the standpoint of marketing strategy, creating “customer value” means meeting customers’ needs and increasing customer satisfaction (Porter, 1985) Customer value management has been used widely by firms to differentiate themselves from competitors

Previous studies have shown that perceived customer value can change with consumption context Thus, buying on the Internet, may lead to a change in perceived customer value relative to other traditional channels Despite of its importance, a few systematic researches could be found to understanding how an online shopping context could affect perceived customer value, especially in typical service industry such as Airlines The study of Chen and Dubinsky (2003), which suggested a preliminary investigation on perceived customer value in e-commerce, got some interesting findings A conceptual model of perceived customer value, involving several influencing factors, was suggested However, as suggested by the authors that further empirical supports need to be done for future works to ascertain whether the model is posited

In such a context, investigating appropriate models, to better understand which key factor affecting on perceived customer value, and correctly applying them in dynamic service industries, such as Airlines, in emerging markets, such as the one of Vietnam, will help to bring tremendous benefits for both firms and customers Several implications could

be withdrawn to help the firm, such as Vietnam Airlines, to well implement for further development of its online channels, and hence increase its competitiveness over competitors

1.2 Research problem

2.1.1 Brief introduction about Vietnam Airlines

Vietnam Airlines is owned by the government of Vietnam The airline is headed and overseen by a management team, members of which are selected by the Prime Minister of Vietnam The airline branches include: Southern, Central and North branches Currently the airline is headquartered in Hanoi

Vietnam Airlines has an extensive network throughout East Asia, Southeast Asia, Europe and Oceania With more than 290 daily flights, the airline flies to 20 destinations domestically, and to another 28 internationally Operating the young fleet of 70 modern aircrafts (Airbus and Boeing mainly), the airline expects to further expand its fleet to 115 and 165 aircrafts by 2015 and 2020, respectively1

Vietnam Airlines is currently a member of SkyTeam and codeshares with most SkyTeam members Vietnam Airlines also has codeshares with four current Oneworld

1

http://www.vietnamairlines.com.vn/

members – American Airlines, Cathay Pacific, Japan Airlines, and Qantas – and is part of Cathay Pacific's Asia Miles program Joining SkyTeam marks a favorable step toward Vietnam Airlines’ future growth It will bring passengers convenient and seamless service

of international standards to a global network covering more than 850 destinations in 169 countries

Vietnam Airlines holds about 40 per cent of the market share of international tourists flying to and from Vietnam This is significant because Vietnam Airlines receives two-thirds of its profits from international passengers In domestic market, Vietnam Airlines and VASCO (its subsidiary) has an 80% share of the aviation market, with the rest covered by Jetstar Pacific and the others (Vietjet Air, Mekong Air – a new carrier)

In 2009, the airline's revenue was $US1.3 billion, compared to $US1.56 billion it earned the previous year Vietnam Airlines carried 9.3 million passengers during this period It is reported that Vietnam Airlines’ passenger capacity for 2010 has risen up to 30% over the same period of the previous year ( Figure below ) Total sales are estimated to obtain a high rate of ~18% per year from 2010 to 2015 2

Figure 1 Top 30 airlines in Asia (Anna.aero, May 2010)

The sales resulted from the Southern branch of Vietnam Airlines, located in HoChiMinh City, hold approximately ~50% of the total of the company The remains hold

by the North and Central branches The organizational charts of the corporation and of the Sales Department (passenger) are given in the Appendix 12

Vietnam Airlines has developed a multichannel sales network including both offline and online ones ( Table below ) Offline channels are traditional channels which still hold a large portion of total sales (~ 97% total sales, YTD 20103) The online channels, which include both B-to-C and B-to-B customers, bring only the remains of ~3% The B- to-C e-ticket selling is done via the Airline’s website The later (B-to-B) is performed with

a particular web-based site developed for corporation clients, called “web portal” Some internal sources show that the company is now developing an ambitious plan to foster sales via online channels, especially on web-portal4

Table 1 Classification of key sales channels of VietNam Airline

(5) Tour Agents (6) Independent sales Agents

*Sales estimated: ~ 85%

2.1.2 Identification of research opportunity

Vietnam achieved around 8% annual GDP growth from 1990 to 1997 and continued at around 7% from 2000 to 2005, making it one of the world's fastest growing economies The country’s GDP growths by 8, 9, 6 and 5 percent for the year of 2007,

2008, 2009 and 2010 respectively despite of the difficult period during the financial crisis6

The penetration of Internet user in Vietnam has been estimated to be 35% Approximately $US 6.3 billion has been estimated to inject into the internet market to meet

the expectation in 2010 It can be seen ( Table below) that for 10 years (2000-2010), the penetration of Internet users has been grown up to 27%

Table 2 Internet Usage and Population Statistics (Internet World Stats)

Furthermore, some recent reports has identified that Vietnam’s air travel market is one of the most potential markets in Asia with double-digits growth rate (Figure below)

Figure 2 Top Asian air travel market (Anna.aero, May 2010)

In 2009, Vietnam Airlines has officially launched the e-ticket selling via the website of the company for one year ( VOA news, 2009 ) With such new channel, passengers can make their bookings, and buy e-tickets using credit cards or via ATM services Customers also save time by having easier access to all the information related to their flights, bookings and ticket prices To ensure the highest security for payments, Vietnam Airlines has cooperated with chartered credit card organizations to take measures

to limit risks in conducting e-transaction

After one year of operation, sales via the website channel (B-C) has just been estimated to obtain approximately nearly 1% of total sales Furthermore, the sales brought

by all the online channels (B-C and B-B) has been targeted to attain growth at 2-3% annually7, which is much lower that the average growth rate estimated for total sales of VNA (~18% has been estimated for period 2010- 2015)

Literature review gives us some good “benchmark” cases study about the development of channel portfolio of Airlines KLM (Royal Dutch Airlines) has developed

a multichannel network comprising 4 channels: Airline’s website, OTA (Online Travel Agency), Airline branch, and independent travel agencies ( Harison and Boonstra, 2008 ) The two first channels compose “online channels”, and the later ones define “offline channels” of the Airline The Airline has clearly defined the target for whole portfolio of the channels They have stated that, starting from 12% contribution to total sales in 2005/2006, the online channels (including website and OTA) would obtain the target set up

to 40% in 2008/2009 And as the result, the share contribution of the “offline channels” has been set to be reduced from 88% down to 60% of total sales In fact, the KML’ managerial board has immediately taken actions in shifting “offline” channels to “online” channels in order to reduce its costs by direct sales, and use the internet as a prominent marketing channel These measures have helped the company survive in the market with intense competition, and maintained its major position in this rapidly changing business environment Furthermore, Shon et al (2003) has shown that, in Taiwan, just after 18 month of website channel launching, this prominent online channel gains up to 15% of sales by cannibalize haft of portion previously hold by traditional “offline” tour agents

Under such favorable market potential, promoting sales development through online channels in general, and via website in particular, becomes imperative for Vietnam Airlines in a long run This will help the corporation saves costs for traditional intermediary agents (e.g commission may be paid up to 10% of Airline costs), as well as takes advantages of online channels in creating direct communication with existing and potential customers

As an alternative option, web shopping brings customers convenient and time saving; consumers can easily compare prices, products and service features across

7

Unofficial Internal Source of VNA

available suppliers in the market It has been insisted in several previous studies that Internet as an information and distribution channel can lead to significant changes in consumers’ preferences and their buying behavior (Teichert et al., 2008) Therefore, better understanding of what leads to customers’ online purchase intention has become a crucial issue Recently, “perceived customer value” has gained much attention from e-marketers and researchers because of its key role in predicting purchase behavior; as a result, it helps e-sellers achieve sustainable competitive advantages ( Cronin et al., 2000 )

The brief discussion mentioned-above provides an overall identification of business opportunity for Vietnam Airlines in developing quickly but sustainably its prominent online channel (e.g the website) This gives the motivation for this research in

investigating appropriate models to better understand key factors influencing on perceived customer value The business implication, which can be resulted from the research, will

be a solid background for the Airline’s online-sales development strategy

1.3 Research objectives and questions

2.1.3 Research Objectives

This research at hand aims to:

(1) empirically validate the model suggested by Chen and Dubinsky (2003) in

Airlines industry in Vietnam, in order to better understand and confirm the key antecedents

of perceived customer value in a B-to-C online context A more simplistic model will be tested with a specific Airlines’ clientele – the Vietnam Airlines’

(2) empirically confirm hypotheses regarding inter-relationships amongst factors; (3) analyze possible implications in development of website channel for Vietnam

Airlines in the purpose of improving sales via this potential channel

2.1.4 Research questions

The inter-relationships amongst several factors are mentioned in the model of Chen and Dubinsky (2003) Some research questions for the study at hand could be raised as follows:

Question 1: Whether the factors “Ease of use of website”, “relevant information”, and

“Customer service” related to “perceived customer value”?

Question 2:

(2.a) How well the factors “Ease of use of website”, “relevant information”, and

“Customer service” predict the “valence of experience”? and how much variance in

“valence of experience” can be explained by scores on these factors?;

(2.b) Which is the best predictor of “valence of experience?

Question 3: Whether the factors “valence of experience”, “perceived product quality”,

“perceived risks”, “product price” related to “perceived customer value”?

Question 4:

(4.a) How well the factors “valence of experience”, “perceived product quality”,

“perceived risks”, “product price” predict the “perceived customer value”? and How much

of variance in “perceived customer value” could be explained by these factors’ scores?

(4.b) Which is the best predictor of “perceived customer value”?

1.4 Research model

The study of Chen and Dubinsky (2003) , which suggested a preliminary investigation on perceived customer value in B-to-C e-commerce, found some interesting findings A conceptual model of perceived customer value, involving several influencing factors, was suggested However, as suggested by the authors that further empirical supports need to be done for future works to ascertain whether the model is posited

The research at hand will validate the model proposed by the above authors ( Chen and Dubinsky, 2003 ) A more simplistic model, suggested as the figure below, will be investigated Definition of all the constructs involving in the research model will be

elaborated later in the Chapter 2 More detailed discussion of each concept is described in

that chapter

Figure 3 Suggested research model (simplified from Chen and Dubinsky, 2003)

Several hypotheses, which have been developed to investigate the relationships amongst these factors mentioned-above, are listed in the below Preliminary rationales for the above hypotheses will be elaborated later in the Chapter 2

Hypothesis H2: “valence of experience” of online customer is positively related to his/her

“perceived customer value”

Hypothesis H3: “perceived risks” is negatively related to “perceived customer value” Hypothesis H4: “product price” is negatively related to “perceived customer value”

Hypothesis H5: “perceived product quality” is positively associated with “perceived

customer value”

Perceived Customer Value

Valence of Experience

Perceived Risk

Product Price

Perceived Product Quality

Relevant information Customer

services

Ease of Use of Website

1.5 Research methodology

The research at hand will be limited in a confirmatory research by using the preliminary model suggested by Chen and Dubinsky (2003) with a more simplistic model represented in the Figure 3 Vietnam Airlines’ e-clientele defines the population for this research

In this research, quantitative method is applied to examine the research questions and hypotheses A structured survey questionnaire was administered to collect primary data from e-customers of an Airline in Vietnam Vietnam Airlines was selected for the research The e-clientele of the Airline, who bought air-tickets from its website at least 1 time, was asked to answer the questionnaire Data were collected using a combination of online survey and “personally administered questionnaire” survey which was conducted by the author and a team of students from Vietnam Aviation Academy (VAA) The team was well trained before carrying out interviews With same content, two versions of questionnaire (online and paper versions) were designed to facilitate the efficiency of each kind of interviews

Finally, the administration of data, as well as the analysis related to statistic validation of the investigated model, will be carried out by using SPSS software (version 16)

1.7 Structure of the study

This thesis contains 6 chapters Table 1 shows the title of each chapter The content

of each chapter is briefed in the following paragraphs

Table 3 Structure by chapters of the thesis

CHAPTERS TITLES Chapter 1 Introduction

Chapter 2 Literature Review

Chapter 3 Research methodology

Chapter 4 Data analysis and findings

Chapter 5 Discussion and implication

Chapter 6 Conclusion

Chapter 1: “Introduction” - A preliminary chapter provides an overview on research

background for this study The business opportunity of Vietnam Airlines’ online channel development is discussed The research model, together with research questions and hypotheses developed, are proposed Brief introduction about the research methodology used for this study at hand is presented Finally, delimitations of this study are discussed

Chapter 2: “Literature Review” provides detail of review on previous works Firstly, an

overview on business environment in Airline industry is given This helps to better understand key characteristics of competitive landscape of the industry Secondly, multichannel as a business development strategy and related issues in channel management is discussed Thirdly, the complex change of consumer buying behavior in e-commerce context, and the importance of “perceived customer value” are raised Next, conceptual definition of key factors affecting on “perceived customer value” are discussed Finally, a brief on theoretical framework for this study is given

Chapter 3: “Research methodology” details the research process; research activities

included in each stage are described The Chapter also gives details of the questionnaire items with sources of reference, and the survey development Justification of sampling design is also given Finally, the statistical procedures with SPSS programs used in the research are provided

Chapter 4: “Data analysis and findings” presents the key statistical analysis results

obtained by this research The demographic profile of the respondents is given The results of reliability and factor analyses are presented The steps to perform the analyses on two multiple regression models are detailed; the results of assumption checking (multicollinearity, outliers, normality, etc.) and evaluation of equation as well as of independent variables are presented Revision on research questions and hypotheses posed at the beginning of the study (chapter 1) is carried out; and as a result, a confirmed research model applying for the case under investigation was suggested at the end of the chapter

Chapter 5: “Discussion and implication” focuses on discussion of the results obtained in

the previous chapter The underlying implications are analyzed to better understand the underlying issues of statistical results Some preliminary insights and rationales for the expected findings as well as the unanticipated ones are suggested

Chapter 6: “Conclusion” provides a general conclusion about the result obtained and the

contribution of the research Limitation and future works are analyzed and recommended

2 CHAPTER 2: LITERATURE REVIEW

2.2 Business environment in the airline industry

The airline industry comprises passenger air transportation, both scheduled and chartered, but excludes air freight transport ( Datamonitor, 2009 ) In the past, the airline industry was characterized both by high growth rates and governmentally regulated protectionism The liberalization of the airline industry in most countries globally has lead

to the development of highly competitive market due to the low entry barriers for new entrants ( Teichert et al, 2008 ) Market entrance of low cost carriers with significantly high growth rates is the most obvious evidence for increasing completion in the industry In Vietnam in just over two years, four foreign low - cost airlines have conquered Vietnam’s airline market8 ; this fact leads local airlines face to fiercer competition with foreign airlines especially on international routes

The global financial crisis and recent recession has impacted heavily on the airline industry Mergers and acquisitions are becoming more prominent in order to save struggling companies The threat of bankruptcy is strong in this industry (e.g Zoom, Silverjet, XL, Frontier, ATA, Oasis Hong Kong Airlines and SkyEurope) ( Datamonitor,

2009 )

The Figure below represents the value chain of Airline industry The airline industry possesses some specific characteristics The key suppliers of an Airline include fuel suppliers, aircraft manufacturers, and skilled employees The key buyers will be taken

as leisure and business travellers, the latter considered as B2B Rivalry between airlines is strong, as customer loyalty in this industry is moderately low Some previous studies show that this industry is price sensitive ( Datamonitor, 200 9) Many consumers seek out cheap deals, leading to intense price competition between industry players Alternative modes of transport, such as trains, can be expensive and slow substitutes for air travel, but it is possible that technologies such as videoconferencing may pose a threat to business air travel Players can regain some power by differentiating their services, for example offering the transportation of freight which means that airlines are less reliant on passenger fares, while passenger loyalty schemes have the effect of generating switching costs for

8

http://www.baokhanhhoa.com.vn/english/econo_tour/2006/11/168709/

buyers As a customer who uses the services of a competitor loses the benefits of the loyalty scheme, the equivalent of a switching cost has been incurred

Figure 4 airline industry value chain ( Buhalis, 2004 )

Due to a fiercer competition environment and the success of LCC (Low Cost Carriers), it was showed that some network carriers (namely full service airlines) were obliged to revise their business model by adapting to the LCCs’ strategy with significant cost advantage ( Teichert et al, 2008 ) However, the cost reducing measures of the network carriers are shown ineffective; business customers buy products that do not satisfy their quality expectations, and leisure travelers receive offerings that surpass their quality expectation, but not fulfill their price expectation

In addition, facing with high costs, Airlines are attempting to shift consumers from traditional ticketing channels to their direct channels such as website which is considered

as the most cost effective for Airlines ( Lubbe, 2007 ) However, the emergence of online booking means that consumers can compare and contrast the services available on the market Customers are becoming more and more conscious of their needs Websites, and

Internet in general, as information and distribution channels can lead to significant changes

in consumers’ preferences and their behavior ( Teichert et al, 2008).

In summary, Airlines are faced with both major changes in their business environment and changes in customer behavior

2.3 Airline e-commerce and multichannel development

The Airline industry had applied e-commerce technologies since the early 1970s, nearly 30 years before the web driven e-commerce activities turned into so popular in the mid-1990s Airlines have been applying online technologies for both ticketing and booking Starting from the early-1980s, the computer reservation system (CRS) has played

a very important role in airline services With some other added functions, car rental and hotel booking systems were introduced The entire system has thus turned into a global distribution system (GDS), which is the main background behind current travel websites

on the Internet These systems have not only reduced airline operating costs, but also represented the computerization and globalization of the air transportation industry

Related to airline e-commerce is the development of frequent flyer programs (FFPs) Since the mid-1980s, FFPs have been very popular among large carriers, and a number of databases with customers’ information have been created since that time With more information added, these databases were soon to become critical factors for airline relational marketing and CRM activity success These factors are also very important elements of today’s e-commerce ( May, 2000 ) The e-marketplace for the airline ticketing business became very popular in a short time And it is expected to soon replace traditional travel agents as the major ticketing channel of airlines

Different Airlines use different design of channel strategy development The Figure below represents a sales network schema was suggested by Dembrower et al (2003) Unlike Airline “A”, Airline “B” can use website operated by itself to sell tickets in addition

to the CRS channel9 Some chose to only sell tickets via their own website, like Airline

“C” The internet also gave the chance to create new form of collaboration between Airlines, where they created a “joint venture” website – “collaborative Web” Today, traditional Airlines use several sales channels simultaneously and buyers alternate their choice among the channels

9

Four major CRS companies: Amadeus, Galileo, Sabre and Worldspan (Dembrower et al., 2003)

Figure 5 Sales channels with alternatives that Airlines can use (Dembrower et al., 2003)

In response to radical changes in the air transport market, some Airlines introduced

a fully integrated e-business option into its ticketing process; an example was the case of Royal Dutch Airlines (KML) in 1995 ( Harison and Boonstra, 2008 ) Royal Dutch Airlines (KLM) was founded in 1919 and in 2004 merged with Air France becoming a division within the Air France/KLM group In 2004–2005 the KLM group carried more than 20 million passengers and more than 600,000 tons of cargo Reducing its costs by direct sales and using the Internet as a prominent marketing channel were immediate actions taken by KLM’s management, and these measures helped the company survive the intense competition in the market and maintain its major position in this rapidly changing business environment KLM bases its marketing and sales activities on four main distribution channels that include its own website, online travel agencies (e.g Expedia.com and Kayak.com), KLM branches and ‘‘physical’’ travel agencies ( Table below )

Table 4 Four-distribution channels of KLM (Royal Dutch Airlines)

Great benefits from online channels are clear However, it was pointed out in the literature that a conflict among channels, including offline versus online channels, could be problematic ( Shon et al., 2003 ) The Figure below shows the market-share data of different ticketing channels for 2000–2001 in Taiwan The proportion of tickets sold through direct sales channel is consistently the highest Around 70–75% of passengers pay the fares at the airline counters in the airport and/or in the city office, thus contributing a great deal to the airlines’ operating revenues The remaining 30% of the passengers had been tour agency customers at the beginning of 2000 before ticketing websites were created After 18 months’ operation, direct sales passengers were still contributing 70% of the market share; while Websites shared half of the remaining ticket sales That is to say, traditional tour agents have lost at least half of their customers to the ticketing websites This would have certainly led to some issues of contention between the new and existing channels

Figure 6 Market share of ticketing channels in Taiwan after 18 months of ticketing website launching ( Shon et al., 2003)

2.4 Consumer behavior towards online shopping

Traditional channels, such as travel agents, rely heavily on the commissions they

received from ticket sales to stay in business The commission rate for distributing air tickets accounted for 8-10% of Airline costs ( Alamdari, 2002 ) The Internet has changed the paradigm of air ticket distribution and offers new alternatives New channels through the Internet affect purchasing behavior for air tickets and change the role of existing

distribution channels

Previous work shows that consumers have increasingly favorable attitudes toward line shopping ( Lohse et al, 2000 cited in Chen and Dubinsky 2003 ) Buying on-line is thus becoming more and more acceptable to many people

on-The benefits of the Internet to consumers are tremendous As an alternative channel, Web shopping is convenient and time saving; with rich, free information available, consumers can easily compare prices and product features from several suppliers The Internet has also raised consumers' expectations of retailers

A low barrier to entry has brought more players into the retail business, so competition

in B-to-C commerce has intensified ( Porter, 2001 ) Because consumers now have more bargaining power, lower switching costs, and an increased number of choices available, understanding of on-line shopping behavior processes, as well as understanding what leads

to on-line shoppers' purchase intentions have become an important topic meriting research attention (Korgaonkar and Wolin, 1999; Webster, 1998 cited in Chen and Dubinsky 2003) Amongst the past works, the study of Chen and Dubinsky (2003) seeked to partially

address this important issue and did so vis-a-vis perceived customer value

Perceived customer value has recently gained much attention from marketers and

researchers because of the important role it plays in predicting purchase behavior and achieving sustainable competitive advantage ( Cronin et al, 2000; Dodds et al., 1991 ) From the standpoint of marketing strategy, creating customer value in consumer marketing means meeting target customers' needs and increasing customer satisfaction ( Porter, 1985 ) Customer value management has been used widely by firms to differentiate themselves from competitors ( Hoffman, 2000 cited inChen and Dubinsky 2003)

Previous research has showed the multidimensional and context-dependent nature of

perceived customer value (Bolton & Drew, 1991; Holbrook, 1994 cited in Chen and Dubinsky 2003) That is, perceived customer value can change with the circumstances of the person and consumption situation Thus, buying on the Internet – a new consumption

context - when comparing with traditional ones - may well lead to a change in perceived

customer value, as well as the factors influencing perceived customer value Despite its

importance, a few systematic body of literature suggests how e-commerce shopping

environment does affect perceived customer value

In such context, Chen and Dubinsky (2003) developed a model elaborating the

e-commerce An exploratory investigation was conducted to validate model As suggested

by the authors that further empirical supports need to be done for future works to ascertain whether the model is posited

The research at hand will be limited in a more simplistic version of the full model ( Figure below ) suggested by Chen and Dubinsky (2003) Then, in the subsequent paragraphs, the literature review will focus on the concepts of the constructs which involve

in the model

Figure 7 Conceptual framework of perceived customer value in an e-commerce context

(Chen and Dubinsky, 2003)

2.5 Antecedents of Perceived Customer Value

2.5.1 Concept of Perceived Customer Value

Perceived customer value has been recognized as an important concept in marketing

research It has often been viewed as essentially a trade-off between relative quality and relative price (Cravens, Holland, Lamb, & Moncrief, 1988 cited in Chen and Dubinsky,

2003) However, it has been criticized that this simplification ignores some important constructs (such as: shopping experience, risk), and may be misleading in measuring perceived customer value (Sinha and DeSarbo, 1998 cited in Chen and Dubinsky, 2003)

Based on a synthesis of previous definitions, “perceived customer value is defined as a consumer's perception of the net benefits gained in exchange for the costs incurred in obtaining the desired benefits” (Chen and Dubinsky, 2003)

As suggested in some previous studies that evaluating evaluating perceived customer

value from the perspective of the consumption experience is important (Anderson & Narus, 1998 cited in Chen and Dubinsky, 2003 ) Chen and Dubinsky (2003) suggest that there are four major elements involved in the prepurchase stage which have significant influence on a consumer's value perception and purchase intention in a B-to-C e-commerce

setting These factors include valence of experience, perceived product quality, perceived

risk, and price Each of these can either positively or negatively influence perceived customer value in an online setting

2.5.2 Valence of Experience

“Valence of experience is defined in the research model as a consumer's emotional or

attitudinal state aroused by the prepurchase on-line shopping experience.” (Chen and Dubinsky, 2003)

Consumer perception is considered a process of sensing, selecting, and interpreting stimuli from the “external”, physical world into the “internal” world ( Wilkie, 1994 ) As a result, one can infer that external signals, such as what consumers experience when they are shopping online, can influence consumers’ internal perceptions of customer value Prior works support this argument ( Donovan et al 1994 ) In the model proposed in this

study, a similar positive influence of valence of experience on perceived customer value is

expected

Electronic technology makes the on-line shopping experience different from what happens in the bricks and mortar (traditional) business format With online purchasing, the physical store environment no longer exists, as the shopping experience is converted into a human-Web-site interaction However, consumers need much more efforts, and easily get frustrated and annoyed Therefore, how an e-retailer's website could foster a favorable purchase experience is crucial Three key factors are identified in the study of Chen and Dubinsky (2003): relevant information, ease of use of the Web site, and customer service

The following paragraphs will respectively give the definition of these factors, as well as discussion surrounding the concepts found in the literature In fact, this will serve as the definition of the constructs involving in the research model suggested for this study

2.5.4 Ease of use of the website

“Ease of use of the website is defined as the degree to which the user expects the use of

the website to be free of effort” ( Chiou et al., 2010 )

Because information processing requires cognitive effort, especially when the information displayed is not readily comprehensible ( Coupey, 1994 cited in Chen and Dubinsky 2003 ), a Web-site design that does not facilitate information processing may cause negative effect Analogous to a physical store with a poor layout and store environment, an unfriendly on-line user interface may lead consumers to feel confused, feel a loss of control in the interaction, and ultimately develop negative feelings about tbe on-line shopping experience This could result in their abandoning the purchase process or moving to an alternate Web site On the other hand, an ideal human-Web interaction will lead to positive consequences, longer staying time, and more exploratory behavior ( Novak

et al., 2000 cited in Chen and Dubinsky 2003 )

24 hours a day, and whether the company is responsive can all comprise customer service

Previous research also shows that most consumers prefer some form of human interaction with e-commerce

2.5.6 Perceived Product Quality

Consistent with previous research, “perceived product quality is defined in this research as the consumer's judgment about a product's overall excellence or superiority”

(Dodds et al., 1991; Sweeney et al 1999; Chen and Dubinsky, 2003)

Products consist of “cues” which can be classified into two categories:

“extrinsic” and “intrinsic” While intrinsic cues represent product-related attributes relating to physical properties of the product, extrinsic cues represent attributes that are not part of the physical product (e.g brand name, price, packaging, etc.) Previous research suggests that consumers use “extrinsic” attributes to infer “product quality” (Bolton and Drew, 1991 cited in Chen and Dubinsky, 2003) E-commerce provides consumers with only a visual display of goods and services As such, consumers perceive a lower level of tangibility because of the lack of demonstrable proof about the performance of a product

In this situation, “extrinsic” attributes are expected to have a strong influence on perceived

quality ( Teas and Agarwal, 2000 cited in Chen and Dubinsky, 2003)

It is also noted here that consumer's perception of quality is different from objective

quality The latter describes the actual technical characteristic of the product that is

measurable Conversely, perceived product quality is rather a higher-level abstraction, a

global assessment, and highly subjective owing to the specific consumption setting (Zeithaml, 1988 cited in Chen and Dubinsky, 2003)

Prior research suggests that perceived product quality serves as a mediator in the linkage between extrinsic cues and perceived customer value (Dodds et al., 1991) Chen and Dubinsky (2003) suggest a positive association between perceived product quality and perceived customer value

2.5.7 Perceived Risk

“Perceived risk is the consumer's perception of the uncertainty and concomitant adverse consequences of buying a product or service” ( Dowling and Staelin, 1994 cited in Chen and Dubinsky, 2003 )

Risk, as perceived by the e-customer, has been identified as one of the major barriers to online shopping and thus major e-business firms have taken steps to address risk concerns with security technologies, awareness campaigns, and assurance policy statements ( Chang

et al., 2005 cited in Udo et al, 2010 ) Previous studies suggest that perceived risk is an important variable that needs to be examined vis-a-vis perceived customer value Nicolas and Castillo (2008, cited in Udo et al, 2010) suggests that perceived risk influences shopping behavior and e-purchasing intentions: the higher the perceived risk, the less likely an e-customer’s intention to purchase Broydrick (1998, cited in Chen and Dubinsky, 2003) maintains that removing risk is an important means of enhancing perceived customer value However, exploratory work of Chen and Dubinsky (2003) found that the hypothesis regarding the inverse association between perceived risk and perceived customer value was not support in their context of research In this research, we will incorporate similar hypothesis (e.g negative association between two constructs) with the purpose of

experimentally “confirming” the result obtained by these authors

2.6 Conclusion about theoretical framework for the research

The foregoing discussion offers a broad theoretical background for the study at hand

In an B-to-C e-commerce context, cognitive effort of the purchase occurs prior to actual buying behavior Therefore, knowing how to favorably influence customers in the prepurchase stage would be crucial for e-retailers Perceived customer value has been found to be a powerful predictor of purchase intention Thus, identifying key factors which

are critical for “shifting” browsers into buyers, acquiring new customers, as well as retaining current customers should be of the great interest to e-marketers

The research framework proposed for this research is based on the preliminary research model suggested by Chen and Dubinsky (2003) The research model, which is presented in the Figure 3 , identifies key factors which could influence on “perceived customer value” in

an online ticketing context Several antecedents and mediating variables are included Set

of hypotheses is developed thanks to the foregoing literature review of conceptual and empirical works As insisted by Chen and Dubinsky , what distinguishes this kind of model from the previous works is the inclusion of the factors which are solely appropriate to online shopping effort In other words, this research seeks to identify variables which are unique in an e-commerce context which might affect on perceived customer value of online customers

3 CHAPTER 3: RESEARCH METHODOLOGY

This chapter will describe the research methodology used in the study It describes in detail the research process, questionnaire design, and as well as sampling design The sources used for development of items, which measure the corresponding constructs, will

be reported Finally, survey method and statistic techniques for data analysis will be presented

3.1 Research model

The research at hand is limited in a confirmatory research by using the preliminary model suggested by Chen and Dubinsky (2003) with a more simplistic model as shown in the Figure 3

The authors identify key factors that influence perceived customer value in a

B-to-C e-commerce setting Several antecedents and mediating variables of perceived customer value are included in the model four major elements involved in the prepurchase stage that have significant influence on consumer's value perception and purchase intention in a B-to-

C e-commerce setting These factors are valence of experience,perceived product quality, perceived risk, and product price Each of these can either positively or negatively influenc perceived customer value in an on-line setting Valence of experience is defined in this model as a consumer's emotional or attitudinal state aroused by the pre-purchase on-line shopping experience The key factors identified that influence what consumers experience when they are shopping online in the model include: relevant information, ease of use of the Web site, and customer service The definitions of these factors involved

in the model will be detailed in subsequence paragraphs

3.2 Research process

A questionnaire was designed to collect primary data to experimentally valid the research model A three-stage approach has been used in the research process: (1) generation of items; (2) pre-test survey; and (3) main survey (see Figure below)

The core activity of the first stage (1) is to do intensive literature research Starting from the identification of business opportunity, research problem and theoretical framework for the study were identified through the work of literature research A research model, involving key constructs identified, for the study was well defined at this stage

Items required to measure each construct were generated by using different research sources (mentioned in the paragraphs below) ; the wording of items from previous works found in the literature were carefully and appropriately revised for this research Then the first draft of questionnaire under hardcopy and online format (using free tool of Google document) was completed

The population of the survey is defined as the clientele of Vietnam Airlines (VNA) who have bought e-ticket(s) via the website of the company at least one time The stage (2) involved a pre-test survey (face validity) A sample of 30 individuals was arranged for answering the first draft of questionnaire, and then interview 5 persons (face-to-face or via telephone and email) to see if any items caused confusion or the wording of items was clear and understandable Because the method for collecting data in the main survey were mainly based on online electronic questionnaire, which was designed by using a tool of Google document, a pilot of method and tactics for data collection and running with SPSS was also carried out at this stage At the end of the step, a final version of questionnaire, under both hardcopy and online electric format, was well prepared for the next step (3) of the main survey

The stage (3) involves activities for the main survey of the research Multimethods (personally administered questionnaires & electronic questionnaires) of data collection were used The collected data were coded into SPSS software to prepare for analysis Three main types of statistic analysis were carried out to validate the hypotheses and explore the questions of the research First of all, the reliability of the constructs was checked by calculating Cronbatch’s alpha coefficient (Cronbach, 1946) The second type

of analysis was the “factor analysis” The factor analysis examines the way in which each respondent completed the items, end suggests that certain items should come together on

“clusters ” (Cavana et al., 2001, page 439) The final step of the stage (3) was regression analyses A series of multiple-regression analyses was performed to determine whether the hypotheses developed in this study received empirical support More details about statistic methods for this research will be detailed in the paragraphs below

Figure 8 Research process of the study

Face validity (&

pilot data running)

Factor analysis

Multiple regression analysis

Literature research Fist draft

questionnaire

(hardcopy &

online)

Item generation

Moving items with low item-total correlations (if

* Assessment of suitability of the data (KMO≥ 0 6 , Bartlett’s test)

3.3 Questionnaire design

Questionnaire survey was designed to collect information of Vietnam Airlines’ customers who bought e-tickets via the website of the company The questionnaire (see Appendix 1 ) includes two parts, in which that “part 1” is the most important helping collecting primary data needed for analyzing the perception of customers The “part 1” includes several sub-sections titled the construct names In fact, these sub-sections measure the constructs involving in the research model by several questions (items) The items needed for measuring each construct were generated by using different research sources; the wording of items found in the literature was revised carefully and appropriately for this research The “part 2” focuses on gathering personal information of the customers (profile information)

The research process previously mentioned explains how the questionnaire was designed and modified from the first draft to the final version The pre-test survey of 30 customers of the Airlines was carried out to check if the items (questions) made confusion and then modify them appropriately The questionnaire was designed under two types of format - hardcopy and online It takes about 10 minutes to answer the whole questionnaire

It is also remarked that the questionnaire was originally designed in English, and then carefully translated into Vietnamese (see Appendix 1 )

The online electronic questionnaire was designed using a free tool of Google (called Google docs) Aesthetic aspect in design of the online version was carefully paid attention

as the author assumes that this can make customers answer the questionnaire enthusiasmly

An example of the online version’s interface is given in the Appendix 2

3.4 Measures

There are 8 main constructs in the theoretical model An intensive review on previous studies provided the basis for developing scale items for each construct The number of items in each construct and the sources of each item are provided in the table below The adaptation of item wording should be carefully carried out for this research with specialization in Airline ticket The English and Vietnamese versions of these measures are included the questionnaire in Appendix 1

Five point likert scales were used in this study, as previous studies have shown that a five point scale is comprehensible to respondents and enable them to express their views (Marton – Williams, 1986 cited in Chen and Dubinsky, 2003) Scale reliabilities were evaluated by calculating Cronbach’s coefficient Alpha (see Figure above ) Once the reliability and factor analyses completed, scores of items were averaged out to represent the corresponding construct

No Scale (construct) Number

5 (1) The website is easy to use

(2) The interface is user friendly (3) It is difficult to operate the website

ea1 ea2 ea310

adapted from Hassan, 2003

2 “Relevant

Information”

5 (1) The quality of information on the website is valuable re1 Chen Z., Dubinsky

J.A (2003) (2) The information on the website is relevant

(3) The information on the website is to the point (4) The information on the website does not help me at all (5) The website provides sufficient information

re2 re3 re411re5

Muylle et al., 2004

3 “Customer service” 5 (1) The means of contact (email address, telephone number, hotline) of

customer support staffs are available and easy to contact

se1 Chen Z., Dubinsky

J.A (2003) (2) Customer support staffs are never too busy to respond to my

requests (3) I receive quick responses to my questions from customer support staffs

(4) I am informed accurately about when service requests will be performed

se2 se3 se4

adapted from

Dabholkar P (1995)

(5) Generally, the quality of interaction between me and the Airline Customer Service is excellent

se5 Chen Z., Dubinsky

enjoyable (3) This website immerses me to exciting product that it offers (4) Shopping tickets on this web site is a very nice time out (5) I enjoy this website for its own sake, not just for the tickets I have purchased

va3 va4 va5

2005 (based on Babin et al., 1994)

5 “Perceived risks” 4 (1) I worry if my online transactions are safe

(2) I worry about how my personal information might be used when I buy tickets online

(3) I worry about my credit card information being stolen (4) I worry about the product quality on the Airlines website

rk1 rk2 rk3 rk4

Udo et al, 2010

6 “Perceived product

quality”

4 (1) The e-ticket is likely reliable

(2) The e-ticket is likely dependable (3) The workmanship on the e-ticket is good

pro1 pro2 pro3

[1] Dodds et al.,

1991 [2] Sweeney et al

1999 (4) The e-ticket is likely to be good quality pro4 Chen Z., Dubinsky

J.A (2003)

7 “Product price” 3 (1) Compared to other tickets with similar features (class, service

quality level) the price I pay for tickets offered by Vietnam Airlines’

website is high (2) Considering the price of tickets bought via Vietnam Airlines’

website, I would say the price is high compared to those bought via other traditional channels (agencies, etc.)

(3) Considering 1 as the cheapest price and 5 as the most expensive one

of available products with similar features, how would you rate tickets purchasing via the Vietnam Airlines’ website?

[2] Sweeney et al

1999

8 “Perceived customer

value”

5 (1) Airline tickets purchased via website are a good economic value

(2) I am happy with the price of airline tickets purchased online (3) The prices of tickets I purchased from the website is too low compared to the quality of Airlines service

(4) Considering that you get what you pay for, the e-tickets bought via website appears to be a good value for money

(5) I get a good value for time and effort I spend to purchase tickets online

cva1 cva2 cva3

cva4 cva5

adapted from:

[1] Chen Z., Dubinsky J.A (2003)

[2] Mathwick et al.,

2001

a) Population

The theoretical research model was limited to investigate the perceived value of Vietnam Airlines’ e-clientele who is defined as individuals bought at least once the air ticket(s) via the website of VNA As explained in previous section, the understanding of key factors affecting on perceived customer value will help to bring tremendous benefits for the firm in terms of great possibility in satisfying customers, and hence increasing sales via this prominent channel Therefore the population of the study is individuals who bought the Airline’s ticket via its website at least once In fact the company has just officially launched the ticket selling via the website for just one year

b) Sampling method

According to the literature, there are two main sampling methods, probability and probability sampling (Cavana et al., 2001, page 257) Taking into account of the extent of generalization desired, the availability of time, as well as the purpose of the study, the

non-“convenience sampling” was selected for this research Thanks to the advantages of the online electronic version of the survey questionnaire, an email attached with a link to the online questionnaire (designed by Google doc) was sent to the list of emails of the author’s friend network that were available to provide their opinions by answering the questionnaire A call for help to the author by spreading out the link of online questionnaire to the network of author’s friends was well written in the email (the feedback of nearly 300 individuals by using this method showed the success of such simple and efficient way) Moreover, a team of interviewers including 4 students from the VAA (Vietnam Aviation Academy) were built up to carry out

“convenient” interviews with e-customers of VNA at the TSN International Airport in Ho Chi Minh city 53 completed questionnaires were collected by this way The collecting data methods will be discussed in more detail in the subsequent paragraphs

c) Sample size

Formula for determining the sample size (n) can be found in the literature (Cavana et al.,

2001, p 257) as follows:

2 2 2

E Z S

n=Where: (S) is the sample standard deviation (normalized to the mean in the case of using a likert scale), Z is statistical value equivalent to the confidence level desired (if being 95% then Z=1,96), E is the level of precision required (5%) As a first attempt to estimate the sample size (n) for this research, a pilot of data running over 100 valid questionnaire answers was carried out

to find “S” values for 2 dependent variables involved in the research model The values of

calculating approximate sample size required for the research: n=119

Moreover the previous studies (Roscoe, 1975 cited in Cavana et al., 2001, p 279) showed several “rules of thumb” for determining sample size, and stated that in multivariate research, including multiple regression analyses, the sample size should be several times (preferably 10 times or more) as large as the number of variables in the study Or, more concretely, it is showed

by Tabachnick and Fidell (2001, p 117), the formula for calculating sample size (n), taking into account the number of independent variables, is stated as follows: n > 50 + 8m (where m = number of independent variables) In this study, we have 3 or 4 independent variables for each regression model, then it requires n > 82 cases

The mentioned above formulas from previous studies do support the validity of the sample size of n=165 cases of this research

3.6 Survey method

In this research, quantitative methods were used to explore the research questions and hypotheses The structured survey questionnaire was administered to collect primary data from e-customers of Vietnam Airlines Data were collected using a combination of “online questionnaire” and “personally administered questionnaire” surveys

In the first method – “online questionnaire” survey, after completing the design of online questionnaire with Google document tool, an email attached with a link to the online questionnaire (designed by Google doc) was sent to the list of emails of the author’s friend network that were available to provide their opinions by answering the questionnaire A call for help to the author by spreading out the link of online questionnaire to the network of author’s friends was well written in the email The feedback of 299 individuals by using this method showed the success of such simple and efficient way However, among 299 online participants, there were only 112 valid answers (37%) for the next step of data analysis relating to the research model

In the second method – “personally administered questionnaire” survey, a team of interviewers including 4 students from the VAA (Vietnam Aviation Academy) were built up to carry out “convenient” interviews with e-customers of VNA at the TSN International Airport in

Ho Chi Minh city All the members of the team were well trained by the author before carrying out the survey There were two steps to follow; first, the interviewers approached voyagers at the TSN Airport and asked them if they could help the research by reserving a time of about 10 minutes for answering the questionnaire in the case they bought e-tickets of Vietnam Airlines (VNA) in the past The purpose of the research, which is to improve the services of ticket selling

was given to the customers to fill in The interviewers could also provide some clarifications sought by the customers At the end of the survey lasting in 3 consecutive mornings (from 9AM

to 12AM) in a week, 53 completed questionnaires were collected

In total, the 2 above mentioned methods can help to collect 165valid cases for the next step

of analysis

3.7 Data analysis techniques

As explained in the above paragraph, Multimethods (personally administered questionnaires

& electronic questionnaires) were used to collect data from the Airline’s customers The data were coded and inputted into SPSS software (version 16) to prepare for analysis A coding book (Appendix 3) was developed to facilitate the input, especially for the completed questionnaires got from the on-site survey at the Airport (mentioned above)

Three main types of statistic analysis were carried out to validate the hypotheses and explore the questions of the research First of all, the reliability of the constructs was checked by calculating Cronbatch’s alpha coefficient (Cronbach, 1946) The scales’ overall Cronbach alpha coefficients were checked if being above 0.7; If not, items with low item-total correlations were removed to improve the coefficients (Pallant J., p 92) The second type of analysis was the

“factor analysis” The factor analysis examines the way in which each respondent completed the items, end suggests that certain items should come together on “clusters” (Cavana et al., 2001, p 439) Several sub-steps were needed for such analysis First, the assessment of the suitability of the data for factor analysis was carried out; the KMO index should have a minimum value of 0.6 for a good factor analysis (Cavana et al., 2001, p 174), and the Bartlett’s test of sphericity should

be significant (p < 0.5) Second, Varimax method was used to detect which items clump together

to build separate constructs The nature of underlying latent variables could be recognized thanks

to the results of this method The final step was regression analysis A series of regression analyses was performed to determine whether the hypotheses developed in this study received empirical support Multicollinearity amongst independent variables should be carefully checked Other issues (such as Outliers, Normality, linearity, homoscedasticity) should also be verified before running to assure a good regression analysis

multiple-4 CHAPTER 4: DATA ANALYSIS AND FINDINGS

The previous chapter discussed the research methodology including the operationalisation of the 8 constructs of the theoretical model developed in Chapter 1, and the research design of the main survey This chapter presents the results of the data analysis (with SPSS version 16)

4.1 Profile of samples

The sample for analysis consists of 165 cases The profile information of customers can be collected thanks to the part 2 of the questionnaire (Appendix 1) The average age of respondents was 29 years 46% were male and 54% were female More than 96% of them got the educational level at University (69.3%) or higher (27%) Nearly 62% had the monthly income of more than 6.5 VND million 34% of them purchased one time the e-ticket(s) via the Vietnam Airline’s website, and 66% did so at least 2 times 93% traveled with the Airline during the past 6 months; and the purposes of travelling were business (28%) and leisure (45%)

More details on descriptive statistics for each variable (in the part 2 of the questionnaire) relating to the profile of respondents are given in the Appendix 4

4.2 Reliability analysis

The reliability of 8 constructs was checked by calculating Cronbatch’s alpha coefficients The scales’ overall Cronbach alpha coefficients were checked if being above 0.7; if not, items with low item-total correlations were removed to improve the coefficients

The details of analysis using SPSS (version 16) were given in the Appendix 5 The summary

of results obtained is given in the Table below In the first run, most of Alpha coefficients of the constructs were obtained above the critical value of 0.7, with an exception of items (pri1-pri3) belonging to the construct “Product Price” in the research model With the help of SPSS output (see in Appendix 5), the low item-total correlation “pri2” was removed from a second analysis After removing this item (pri2), the Alpha coefficient was improved up to 0.71 which is quite satisfied when comparing with the critical value (0.7) The final number of items for each construct were obtained (final column in the Table below) to prepare for the next step of analysis – “factor analysis”

4.3 Factor analysis

The factor analysis examines the way in which each respondent completed the items; and suggests that certain items should come together on “clusters” (if any) There are three main steps in conducting factor analysis presented as below

Step 1: Assessment of the suitability of the data for factor analysis

To be considered suitable for factor analysis the correlation matrix should show at least some correlations of r=0.3 or greater In addition, two statistical measures are also generated by SPSS

to help assess the factorability of the data: Bartlett’s test of sphericity, and the Olkin (KMO) measure of sampling adequacy The Bartlett’s test of sphericity should be significant (<0.05) for the factor analysis to be considered appropriate The KMO index ranges from 0 to 1, with 0.5 considered acceptable and with 0.6 suggested as the minimum value for a good factor analysis (Tabachnick & Fidell, 2001)

Kaiser-Meyer-The Table below represents the summary results of factorability assessment for set of items corresponding to the studied constructs Most of the items belonging to corresponding constructs have the value of KMO above the critical value for a good factor analysis An exception can be seen in the case of items belonging to the construct “Product Price” It is not a big surprise as this construct only possesses 2 items (pri1 and pri3) resulted from the Reliability analysis process presented in the above section

More details on the results got from SPSS’s output can be found in the Appendix 6

Original items (coded

names) values - 1 ALPHA st

run

Items deleted values - 2 ALPHA nd

run

Final items number

Step 2: Factor extraction

Factor extraction involves determining the smallest number of items that can be used to best represent the interrelations among the set of variables There are a variety of approaches that can

be used to identify the number of underlying dimensions The most commonly used approach is principal components analysis And finally, it is up to the researcher to determine the number of item which are considered to best describe the underlying relationship among the items There are a number of techniques that SPSS offers and can be used to assist this analysis: Kaiser’s criterion and scree test The most commonly used techniques is the Kaiser’s or the eigenvalue rule The eigenvalue of a factor represents the amount of the total variance explained by that factor

The details of Factor extraction analyses by SPSS can be found in the Appendix 7 Set of items corresponding to each construct were analyzed The eigenvalues for each item in the set (construct) were listed The results obtained showed that all of studied constructs had only one corresponding component (item) recorded eigenvalue above 1 And these components explain from more than 50% up to 78% of the variance (see Cumulative % column in Appendix 7) of the corresponding constructs

Step 3: Factor rotation and interpretation

Once the number of items have been determined, the next step is to try to interpret them To assist in this process the factors are ‘rotated’ Within the two broad categories of rotational

names)

KMO values Bartlett’s test

(p values)

Ease of Use of Website ea1-ea5 (5 items) 0.831 0.000

Relevant Information re1-re5 (5 items) 0.783 0.000

Customer Service se1-se5 (5 items) 0.853 0.000

Valence of Experience va1-va5 (5 items) 0.858 0.000

Perceived Risk rk1-rk4 (4 items) 0.739 0.000

Product Price pri1, pri3 (2 items) 0.5 0.000

Perceived Product Quality pro1 - pro4(4 items) 0.683 0.000

Perceived Customer Value cva1-cva5 (5 items) 0.701 0.000