Impact of Public Transport’s Ticket Service Quality on Customers’Behavioral Intention: A Case of Bus Rapid Transit (BRT) Line 1 (Kim Ma - Yen Nghia) in Hanoi City, Vietnam

VIETNAM NATIONAL UNIVERSITY VIETNAM JAPAN UNIVERSITY NGUYEN HONG HANH IMPACT OF PUBLIC TRANSPORT’S TICKET SERVICE QUALITY ON CUSTOMERS’ BEHAVIORAL INTENTION: A CASE OF BUS RAPID TRA

VIETNAM NATIONAL UNIVERSITY

VIETNAM JAPAN UNIVERSITY

NGUYEN HONG HANH

IMPACT OF PUBLIC TRANSPORT’S

TICKET SERVICE QUALITY ON

CUSTOMERS’ BEHAVIORAL INTENTION:

A CASE OF BUS RAPID TRANSIT (BRT)

LINE 1 (KIM MA-YEN NGHIA)

IN HANOI CITY, VIETNAM

MASTER’S THESIS BUSINESS ADMINISTRATION

Hanoi, 2019

VIETNAM NATIONAL UNIVERSITY

VIETNAM JAPAN UNIVERSITY

NGUYEN HONG HANH

IMPACT OF PUBLIC TRANSPORT’S

TICKET SERVICE QUALITY ON

CUSTOMERS’ BEHAVIORAL INTENTION:

A CASE OF BUS RAPID TRANSIT (BRT)

LINE 1 (KIM MA-YEN NGHIA)

IN HANOI CITY, VIETNAM

MAJOR: BUSINESS ADMINISTRATION

CODE: 60340102

RESEARCH SUPERVISORS:

DR MAI ANH PROF MOTONARI TANABU

Hanoi, 2019

TABLE OF CONTENTS

CHAPTER 1 INTRODUCTION 1

1.1 Rationality 1

1.2 Research objectives 5

1.3 Research questions 5

1.4 Subjects of research 5

1.5 Scope of research 6

CHAPTER 2 LITERATURE REVIEW 7

2.1 Service Quality 7

2.2 Customer Satisfaction 11

2.3 Behavioral Intentions 13

2.4 Service Quality and Customer Satisfaction 15

2.5 Service Quality and Behavioral Intentions 17

2.6 Customer Satisfaction and Behavioral Intentions 18

2.7 Customer Satisfaction on the relationship of Service Quality and Behavioral Intentions 19

CHAPTER 3 THEORETICAL FRAMEWORK, METHODOLOGY 20

3.1 Theoretical framework 20

3.2 Methodology 20

CHAPTER 4 RESULT ANALYSIS AND FINDINGS 24

4.1 Validity of measurement 24

4.2 Descriptive statistics and correlation 30

4.3 Hypotheses tests and result analysis 32

CHAPTER 5 CONCLUSIONS AND RECOMMENDATIONS 36

5.1 Conclusions 36

5.2 Recommendations 38

5.3 Limitations 39

5.4 Suggestions 40

REFERENCES 41

APPENDIX 1 Questionnaire 46

APPENDIX 2 SPSS results 53

I would like to express my special gratitude to officers of MBA Program for your patience and endless assistance during the implementation of this thesis, as well as for Department of Academics and R&D Promotion for consistent backup of the thesis’ execution

My appreciations and thanks also go to my colleagues and people who have willingly helped me to complete this thesis

Nguyen Hong Hanh

TABLES AND FIGURES

LIST OF FIGURES

Figure 1.1 BRT Line 1 alignment 2

Figure 2.1 Service Quality model by Gronroos (1984) 7

Figure 2.2 Service Quality model by Rust and Oliver (1994) 10

Figure 2.3 Service Quality Model by Cronin & Brady (2001) 10

Figure 2.4 Expectation Confirmation Theory (Oliver, 1980) 12

Figure 2.5 Theory of Planned Behavior 14

Figure 3.1 Conceptual research framework 20

LIST OF TABLES Table 3.1 Questionnaire sources 22

Table 3.2 Demographic Profile of Respondents 23

Table 4.1 Rotated Component Matrix 26

Table 4.2 KMO and Bartlett's Test 27

Table 4.3 Exploratory Factor Analysis and Reliability Estimated Results 29

Table 4.4 Descriptive Statistics and Correlations 31

Table 4.5 Linear regression results 33

Table 4.6 Summary of Hypothesis testing results 35

ABBREVIATIONS

AFC: Automatic Fare Collection

HPC: Hanoi People’s Committee

DOT: Hanoi Department of Transport

TRAMOC: Public Transport Operation and Controlling Center of Hanoi TRANSERCO: Hanoi Transport and Services Corporation

IC Card: Integrated Chip Card

FeliCa: Operating System for IC card invented and marketed by SONY BRT: Bus Rapid Transit

MRT: Mass Rapid Transit (Urban Railway)

TPB: Theory of Planned Behavior

CHAPTER 1 INTRODUCTION

1.1 Rationality

The role of public transport in mega city

In a megacity, transport would be the most critical system among various urban structures The scale for demand growth in transport services is in line with a city’s growth, which is more than any other urban services like water, electricity, internet, etc

It is distributed by space, flow direction and time unevenly In addition, this system also needs to take care of different kind of services which are dependent on individual choices Therefore, this system has uncertainties in terms of demand and service quality (Morichi et al., 2013) Public transport includes Mass Rapid Transit (MRT or Urban railway network), Bus, Bus rapid transit (BRT) and Taxi

However, there are problems in transport systems in developing megacities in Asia Passenger mobility in these cities are high thanks to high economic growth and rapid urbanization Meanwhile, the infrastructure is insufficient to respond to the growing demand Model share in public transport seems declining in comparison to private ones due to increasing affordability

In developing big cities like Hanoi which has the population of about 9.8 million as of

2018, the introduction of public transport like bus or BRT has been considered early approach to gradually provide incentives for passengers to stick more to public transport, decreasing private modes Still, how to encourage more passengers to this system is always a headache to policymakers Among many different policies and strategies to enforce the use of public transport, the improvement in service quality is always crucial matter

Importance of Ticketing system in public transport operation

The operation of public transport includes various services such as transport service, station services and ticketing services, etc Ticketing system is considered one of important factors since passengers firstly interact with a transport system by its ticketing system

The Government always discussed about how to provide incentive policies to assure the ridership for public transport and decreasing the private vehicles There are PUSH and PULL policies, in which, PUSH focused on forcing measures such as imposing high tax or banning private vehicles in certain conditions, and PULL refers to those with more incentives to attract passengers such as improving the service quality

As a part of the overall services of public transport, any improvements in ticketing service quality shall contribute to the realization of the Government’s goals

The introduction of new technology for public transport in Vietnam

BRT Line 1 (Kim Ma – Yen Nghia) is the only BRT line in Hanoi so far, which was constructed by the support of World Bank to Hanoi City This line started its operation

on December 31, 2016 by TRANSERCO as operator This line included 27 stations along the alignment, spreading through Giang Vo, Lang Ha, Le Van Luong, To Huu,

Le Trong Tan and Quang Trung (Ha Dong District) streets BRT buses operate on their own dedicated lane which is marked on the road Dedicated lane for BRT does not allow any other vehicles, this characteristic ensures the frequency of BRT which is designed as 30 minutes per round and 60 turns per day This makes BRT differentiated

to ordinary city bus because it will be faster by using dedicated lane

Source: TRAMOC

Figure 1.1 BRT Line 1 alignment

The Government of Vietnam is getting more and more concerned with introducing high technology in operation and management of public transport New electronic ticket system applying smartcard and QR code tickets is expected to benefit the operation and management of public transport system such as bus services, Bus Rapid Transit (BRT) and future Mass Rapid Transit (MRT) Hence, Hanoi City has recently introduced the first electronic ticket system for BRT Line 1 (from Kim Ma to Yen Nghia) as a pilot to evaluate and understand the feasibility of this application in the context of Hanoi City

The experience of a new ticketing system which is mostly automatic and very advanced would cause different behavioral intentions of passengers in Hanoi City – who have been, for very long time, only familiar with manual paper-based system Due

to characteristics of the local people, it is necessary to know how loyal the passengers would like to be thanks to the introduction of this new service

The research is focusing on ticketing service in public transport, i.e the case of BRT line 1 in Hanoi City, Vietnam In this system, the ticketing was changed from ordinary manual paper-based tickets to smartcard and QR code-based tickets Passengers, after purchasing the card/QR code ticket, can pass through the automatic gates at BRT stations by tapping on the validators/scanners This research will inspect the impact of service quality aspects based on Brady & Cronin model (2001) with constructs mostly related to the ticketing service of BRT Line 1 (Proposed conceptual framework is explained later)

Pilot application of ticket service for BRT Line 1 in Hanoi, Vietnam

The pilot of ticketing system for BRT 01 was launched on October 10, 2018 to inspect the community’s reaction as well as seeing the system performance The pilot was instructed by Hanoi People’s Committee, and organized and invested by a consortium

of technology and operating companies including Viettel Telecom – the biggest network carrier in Vietnam, Hanoi Transport and Services Corporation (TRANSERCO)

– Hanoi’s major bus operator, MK Smart – a card-maker based in Vietnam, and Nhat Cuong Software

The ticket system of BRT issues smart cards or QR code tickets and validate the tickets

by the device at automatic gates installed inside BRT stations The issuance is made by the customer’s online operation, or computer operation by ticket staff at ticket office in Kim Ma station Since this is the transition period from manual ticketing system to automatic system, ticket staff is always allocated in stations to support passengers whenever passengers have problem in operation

According to the implementer and management organization (Hanoi Public Transport Control and Management Center (TRAMOC) under Hanoi Department of Transport), passengers are getting used to the operation of the system, but they have both positive and negative opinions on this new electronic system of ticketing

It is the goal of municipal government to realize e-ticket system for the whole public transport network in Hanoi However, its realization will be executed based on the evaluation’s result of the pilot system, and the subsequent improvement and implementation strategy will be discussed after the pilot

Thus, to understand how service quality of this ticket system would impact the behavioral intentions, i.e the intention to reuse, recommend and to have positive word

of mouth, is meaningful to managers and decision makers in Hanoi Department of Transport Advanced technology, or contactless technology as in this case, will be definitely the trend for application all over public transport system in Hanoi City in particular and in Vietnam in general As a result, decision makers will always need to observe the quality of the service as well as the consequences it may cause to the passengers, since it is always the goal of the City that is to increase the ridership of public transport, and mitigating the number of private vehicles, contributing to easing the traffic congestion in such a crowded city like Hanoi

Hence, the research is not only theoretical meaningful because it would provide empirical evidence on the impacts of ticket service quality on behavioral intentions of

public transport’s customers, but also expected to imply practical managerial recommendations to operators, City’s decision makers and regulators on how to improve the service to realize incentive policies for public transport passengers

1.2 Research objectives

1 To understand and explain the relationship among service quality of ticketing system

in BRT Line 1, customer satisfaction as well as customer behavioral intentions to repurchase, recommend and say positively about this service

2 To inspect the impact of Customer satisfaction on the relationship between Service Quality and Behavioral Intention in both meditating and moderating roles

Target respondents are passengers who have used and/or experienced the ticketing system of BRT Line 1 in Hanoi City since October 2018 (the launching time of the pilot for smartcard and QR code system)

1.5 Scope of research

The research was conducted from January to May in 2019 in Hanoi City, Vietnam

since its subject is the BRT Line 1 in Hanoi City and its ticketing system using

Smartcard and QR code tickets

Estimated number of respondents for the research is 200 to 300 from the beginning of

the research Respondents are ticket users of BRT Line 1 in Hanoi City

CHAPTER 2 LITERATURE REVIEW

2.1 Service Quality

There are various definitions of Service quality upon objectives for studies and studying environment The understanding of service quality is a basis for any improvement solutions to be applied inside firms Thus, the definition of service quality is not only important in view of creating development objectives, but also orienting the activities in firms in order to optimize their strengths in the most effective ways

According to Lehtinen & Lehtinen (1982), service quality shall be evaluated based on two aspects (1) service supplying process and (2) service results Gronroos (1984) also defined service quality with two factors including (1) technical quality and (2) functional quality The technical quality refers to those being served, and functional quality explains how services are given

Figure 2.1 Service Quality model by Gronroos (1984)

Parasuraman et al (1985) showed a model to measure service quality by focusing on perception and expectation There are five (05) gaps between expectation and service performance for measurement between

1) consumer expectation and management perception of the expectation of customer 2) consumer expectation’s management perceptions and service specification

3) service specification and service provided

4) service provided and external communications to customers

5) customer expectation & consumer’s perception about the service

Accordingly, 10 dimensions were initially proposed to measure the gaps:

SERVQUAL model has been becoming one of most used models by scholars and practitioners for years Still, there were researchers argued that this model is not clear for measuring gaps in different levels (Brady and Cronin, 2001)

Researchers continued to study about Service quality and found that SERVQUAL is not comprehensive for various applications (Brady and Cronin, 2001) Dabholkar et al

1996 had argued that this model was not suitable for fields such as retail stores Hence, SERVPERF model was worked out, measuring the performance with same dimensions

of SERVQUAL, instead of “expectation and perception” Cronin and Taylor (1992) who developed SERPPERF explained that service quality is consumer’s attitude and that performance (perceived service) is the only measurement of service quality

In later years, researchers are still further researching about the topic to find the best fit model for different fields and services Hierarchical model developed by Brady & Cronin in 2001 was based on SERQUAL by refining SERQUAL dimensions, as well

as adopting measures of Gronroos’s model (1984) and Rust and Oliver model (1994) From Gronroos model (1984), two dimensions were used They are Interaction Quality which explains relationship between customers and staff (functional quality) by sub-dimensions including Attitude, Behavior and Expertise and Outcome Quality (technical quality) by sub-dimensions of Waiting time, Tangibles and Valence In addition, the service environment is the factor adopted from Rust and Oliver (1994) model, which became Physical Environment Quality in the model proposed by Brady and Cronin (2001)

As a result, the service quality based on hierarchical model by Brady and Cronin (2001)

is generated by three categories, i.e Interaction Quality, Physical Environment Quality and Outcome Quality, and each respectively is explained by sub-dimensions which are also attributes of service quality, i.e Attitude, Behavior and Expertise (for Interaction),

Ambient conditions, Design and Social factors (for Physical Environment) and Waiting time, Tangibles and Valence (for Outcome)

Figure 2.2 Service Quality model by Rust and Oliver (1994)

Figure 2.3 Service Quality Model by Cronin & Brady (2001)

This hierarchical model has strongly literature supported and combined of some models as explained above It is generic and can be applied for a variety of industries Also, it is flexible to include different factors upon varied business (Brady and Cronin, 2001) The model has developed from the idea that service quality perception is

multilevel and multidimensional (Dabholkar et al., 1996) By this, it enables managers

to identify problems of the service right from primary stages of service provision, as well as understanding the needs of customers and the weak points in the service, so that the perception of customer about the service quality would be improved by providing higher quality

So far, this is the most suitable and applicable model for measuring service quality (Ghotbabadi et al., 2015) Thus, in this research, the hierarchical model of Brady & Cronin (2001) was utilized for Service Quality model due to its generic characters for different services Brady and Cronin (2001) in their work also mentioned that this model is flexible to be adjusted with other dimensions or sub-dimensions to match specific services For example, in some highly automated services, the dimension of Interaction (which is explaining the service quality in the relationship between customers and staff) would be omitted In this research, the ticketing system has not been totally automated On the basis of the system’s characteristics, all dimensions in original model of Brady and Cronin (2001) were remained to do the analysis

2.2 Customer Satisfaction

Customer satisfaction is an important factor among others in marketing Scorecard of core marketing is always measured to maximize customer satisfaction (Sunny et al 2014)

There are various views over Customer satisfaction Satisfaction is defined as individual perception on the performance of a product or service with relation to that customer’s expectation (Schiffman & Kanuk, 2009) Kotler & Keller (2006) defined customer satisfaction as comparative feelings of pleasure or disappointment Satisfaction depends much on the difference between perceived result and expectation

If the outcome, in reality, is lower than expectation, customer is not satisfied If the outcome is equal to expectation, customer is satisfied And, customer will be over

satisfied if the outcome is over expectation Expectation is formulated from experience, outsiders (those who have been used the service) and information of service seller Customer satisfaction is concepted as the evaluation of customer need met expectation (Zeithaml and Bitner, 2000) Customer satisfaction is reaction of customers about difference between experience and expectation (Parasuraman et al 1988) This means that experience of customer in using a service and the result after using such service Specifically, customer satisfaction is the perception of customers about a company when their expectation is satisfied or over satisfied throughout the life cycle of product

or service Customers who are satisfied will be loyal and continue to purchase the product or services of the company

Fournier and Mick (1999) conceptualized Customer satisfaction as judgement, psychological feeling in prior and after purchasing, comparison and interaction with service provider

A popular theory to explain customer satisfaction is Expectation Confirmation Theory This theory is developed by Oliver (1980) and it has been used to study customer satisfaction for service quality of service or product of an organization

Figure 2.4 Expectation Confirmation Theory (Oliver, 1980)

This theory included 2 small processes independently affecting customer satisfaction: expectation of the service before purchasing and perception after being experienced According to this theory, customer satisfaction can be understood as the following process:

- Firstly, customers generate in their mind expectations of factors of service quality that provider can bring them before they decide to buy

- Afterwards, service purchasing and using contribute to their trust on real performance

of the service that they are using

- Customer satisfaction is comparison’s results between effectiveness that this service brings to customers among their expectation before purchasing the service and what they receive after using the service, there will be 3 kinds of status:

* Satisfaction is confirmed if the effectiveness of that service is totally refelecting customer’s expectation

* Disappointment is confirmed if the effectiveness of that service is not matching their expectation

* Satisaction is reached if their perception and experience of using the service is beyond what they expect before purchasing the service

This research employs 5-item measurement of customer satisfaction of Wen et al.,

2005 which asking for overall satisfaction of customer over the service they experience The research of Wen et al., 2005 was investigating the loyalty of passengers for intercity bus services in Taiwan

2.3 Behavioral Intentions

Behavioral intention refers to the likelihood of customer to engage in certain behaviors

It is important indicator to know the future behavior of customer In public transport, it

is important to understand behavioral intentions of passengers in using public transport since the loyalty of customers is considered the prime determinant of financial performance in long term Lowering the defection rates of customer is profitable to organizations

It is suggested by business analysts that recruiting a new customer will cost 5 times higher than retaining an existing one (Barsky, 1994) The replacement is costly process since it involves various activities such as advertising, set up, selling, etc Zeithaml, Berry, Parasuraman (1996) suggested that customers those stay with the company for years because of their satisfaction over the service quality will tend to purchase additional services, and especially to spread out positive word of mouth

Theory of Planned Behaviors (TPB) (Ajzen, 1991) assumed that a behavior would be forecasted or explained by intention to conduct such behavior

Behavioral intention is assumed including motivation factors which influence behavior

It is defined as efforts that people try to make to conduct a behavior (Ajzen, 1991) According to Ajzen (1991), behavioral intention is a function with three factors Firstly, attitudes are defined as positive or negative evaluation on behavior to implement Secondly, subjective norms refer to social pressure being felt to conduct or not conduct such behavior Finally, it is perceived behavioral control

Perceived behavioral control reflects the difficulty or easiness in implementing the behavior This depends on the availability of resources and chances for conducting such behavior Ajzen (1991) proposed that perceived behavioral control influences directly on behavioral intention, and if one person is accurately in perceived behavioral control, the behavioral control would forecast the behavior

Figure 2.5 Theory of Planned Behavior

Behavioral intention is split into two types, including favorable & unfavorable behavioral intentions (Zeithaml et al 1996) In which, favorable behavioral intentions consist of specific factors including loyalty, switching intentions, external response willingness to pay more, and internal response

Loyalty of customers toward service is always playing important role in that service’s success Customer’s loyalty for a service shows their trend of using that service and repeating this activity Zeithaml et al (1996) argued that if customers see the service quality good, they tend to think highly of that company and more importantly they show their preference to that service As a result, they will re-purchase and purchase more In other words, service quality of ticketing system of public transport is considered one of factors differentiating this service (public service) with others, creating the competitive advantages for BRT to attract passengers and increasing its share This is an objective that all lawmakers want to archive with an aim of increasing the number of riders on public transport modes and decrease the share of private vehicles

Researchers often view behavioral intentions as customer loyalty’s representing measures (Yang and Peterson, 2004) Behavioral intentions consist of repurchase and intention of recommendation (Zeithaml et al., 1996) Hence, this research employs the behavioral intentions concept by Zeithaml et al (1996) with the measurement of 5 items including intention to say positive thing about the service, to recommend the service to others and to repurchase

2.4 Service Quality and Customer Satisfaction

Service quality is the most affecting factor to customer satisfaction (Cronin and Taylor, 1992) If service providers bring products with quality satisfying the demand of customers, those firms have created the satisfaction among their customers

The relationship between service quality and customer satisfaction is a topic gaining

attention and being argued by researchers continuously in recent decades Numerous researches on customer satisfaction in service sector have been carried Some authors argued that these two constructs are the same, thus, these two concepts would be interchangeably used However, via various researches, service quality is showed as differently to customer satisfaction The judgements of service quality are specific and cognitive, while judgements of customer satisfaction are holistic (Oliver, 1997) According to Zeithalm and Bitner (2000), the customer satisfaction is influenced by different factors such as product quality, service quality, price, context factor and individual factor

In researches of services, service quality has close relationship with customer satisfaction though they are different concepts (Parasuraman et al., 1988) Cronin and Taylor (1992) suggested that service quality leads to customer satisfaction The reason

is that service quality relates to service supply, satisfaction can be only assessed after customers use the service

Customer satisfaction is the consequence, while service quality is the cause Satisfaction is forecasting and expecting construct, service quality refers to an ideal standard Customer satisfaction is general concept, explaining the satisfaction of customers when they consume a service Meanwhile, service quality only focuses on specific factors of that service (Zeithaml & Bitner, 2000) Though there is relationship between service quality and customer satisfaction, few studies have been done to test the explanatory of factors in service quality to customer satisfaction, especially in each certain service (Lassar et al., 2000) This relationship has been tested by Cronin and Taylor (1992) showing that perceived service quality leads to customer satisfaction And, service quality is precondition of customer satisfaction (Cronin and Taylor, 1992) and main factor influencing customer satisfaction (Ruyter et al., 1997)

In summary, service quality is factor influencing the customer satisfaction If service provider brings customers products with quality satisfying their demands, that firm has begun to satisfy its customers As a result, to improve customer satisfaction, the firm/service provider needs to improve its service quality In other words, service quality and customer satisfaction are closely related, in which, service quality is created in prior and playing deciding role to customer satisfaction If service quality is improved uncompliant with the basis of customer demands, customers will never satisfy with such service

Direct relationship between Service quality and Customer satisfaction has been confirmed in numerous studies, thus, the research’s first hypothesis is stated as following:

H1a: Service Quality has positive effect on Customer Satisfaction

2.5 Service Quality and Behavioral Intentions

Cronin et al (2000) studied and concluded that service quality is determinant of behavioral intentions In the research by Lien & Yu (2001), service quality has affected both directly and indirectly to behavioral intentions Managers, hence, should consider service quality as the foundation to build up satisfaction of customer and their behavioral intentions

Cronin and Taylor (1992) found positive influence of service quality on purchase intentions In previous study by Parasuraman, Berry and Zeithaml (1991), positive and significant relationship between the perception of customers on service quality and their willingness of recommending the service of the company was found In addition, Boulding et al (1993) also found positive relationship between service quality and repurchase intention, willingness to recommend and positive word of mouth

Research results from an empirical study of multi-company have shown strong

evidence of the influences of service quality on behavioral intentions (Zeithaml et al., 1996)

In public transport, Lai and Chen (2011) in a study about Kaohsiung Mass Rapid Transit System revealed that service quality positively affects behavioral intention

As a result, the second hypothesis was given to test the relationship between Service Quality and Behavioral Intentions

H1b: Service Quality has positive effect on Behavioral Intentions

2.6 Customer Satisfaction and Behavioral Intentions

Studies which mostly based on TPB have shown that customer satisfaction is identified widely as favorable behavioral intentions’ most important determinant

In service industry, a lot of benefits are brought to organizations by satisfaction Satisfaction has been widely defined as a key intermediate Among the benefits, satisfaction is considered as main driver of favorable behavioral intentions (Ranaweera

& Prabhu, 2003) The study examining the combined effects of customer satisfaction and trust on customer retention and positive word of mouth by Ranaweera & Prabhu (2003) confirmed that satisfaction has strong positive effect on customer retention and positive word of mouth

If the customer was satisfied, favorable behavioral responses will be expected and obtained, such as re-purchase and positive word of mouth (Cronin & Taylor, 1992; Wen et al 2005)

Previous studies have confirmed empirically on the direct positive relationship between satisfaction and behavioral intentions in different industries (Cronin & Taylor, 1992; Liu et al 2005; Anderson et al 1993)

Hence, the third hypothesis is stated as following:

H2: Customer Satisfaction has positive effect on Behavioral Intentions

2.7 Customer Satisfaction on the relationship of Service Quality and Behavioral Intentions

Customer satisfaction’s effect on the relationship of Service quality and Behavioral intentions has obtained attentions in various researches Osman et al (2013) showed the results of significant and positive partial mediating effect of customer satisfaction

on the relationship of service quality and customer loyalty in their study about rural tourism industry in Malaysia

Service quality which is developed in this research’s model is argued to affect Customer Satisfaction, in turn, customer satisfaction has influence on the behavioral intentions Cronin and Taylor (1992) ever showed that customer satisfaction positively affects the intention to use by customers

Sunny, I.R and John, A.G (2014) found that customer satisfaction moderates and

explains the influence of factors including service quality to behavioral intentions Accordingly, when customer perceives factors like service quality high, their satisfaction will be increased, which in turn will help to reinforce the behavioral intentions

Hence, the fourth hypothesis is made in order to inspect the moderating effect of Customer Satisfaction on the relationship between Service Quality and Behavioral Intentions

H3: Customer Satisfaction has moderating effect on the relationship between Service

Quality and Behavioral Intentions

All four hypotheses are illustrated through theoretical framework in the next part of the research

CHAPTER 3 THEORETICAL FRAMEWORK, METHODOLOGY

3.1 Theoretical framework

By the support of literatures and hypotheses development, the following framework has been developed

Figure 3.1 Conceptual research framework

The framework with four hypotheses will help to test the following:

(1) The effect of Service Quality to Behavioral intentions

(2) The mediating role of Customer Satisfaction to the relationship between Service Quality and Behavioral Intention

(3) The moderating role of Customer Satisfaction to the relationship between Service Quality and Behavioral Intention

3.2 Methodology

Measurement

A questionnaire was developed based on the measurements which have been used in some previous studies

Service Quality, as explained in the above section, is following the hierarchical model developed by Cronin and Brady (2001) Accordingly, Service Quality is evaluated by three main factors (Interaction, Physical environment and Outcome) Hence, there are

27 items representing all sub-dimensions For example, Attitude is measured by 3 items, i.e 1 You can count on ticket staff being friendly, 2 The attitude of the staff demonstrates their willingness to help me, 3 The attitude of the staff shows me that they understand my needs, or Design is also measured by 3 items, i.e 1 This BRT ticketing device’s layout never fails to impress me, 2 The ticket device layout serves

my purposes, 3 The BRT operator understands that the design of its ticketing facility is important to me, etc

Customer Satisfaction is measured by 5 items utilizing from a study of Wen et al (2005) Those items do include the evaluation over the overall satisfaction as well as expectation of passengers comparing to perceived service by those questions like (1) Overall, I was satisfied with this ticket service, (2) The perceived service was better than the expectation and (3) The perceived service was equivalent to my ideal service The construct of Behavioral Intentions is examined by using 5 items from study of Zeithaml et al (1996) These 5 items survey respondents’ intentions in recommending and positive word of mouth as well as repurchase behaviors by such questions as (1) I would say positive things about the ticket system to other people, (2) I would recommend the service to someone who seeks my advice, and (3) I would use BRT more for its ticket service in the future

Other than respondents’ demographic information which are measured by categorical scales, all items of three constructs are measured using five-point Likert scale, i.e strongly disagree (1), disagree (2), don’t know (3), agree (4) and strongly agree (5)

Table 3.1 Questionnaire sources

Sample and data collection

Survey questionnaire technique was used via direct interviews with passengers at some stations along BRT Line 1 in Hanoi City, and e-survey to all people those have experienced the BRT Line 1’s ticket system since October 1, 2018 Samples were convenience-based collected to reach the sufficient number of respondents

The number of samples were decided based on 5:1 standard of Hair et al (1998) Accordingly, to satisfy the requirements of quantitative analysis, each item needs at least 5 observations, meaning 5 answers Thus, the questionnaire has 37 items, resulting in at least 185 samples In conclusion, 205 samples were collected, in which

200 are accepted for final data of analysis because 5 written questionnaires had not been fully completed Among 200 samples, 167 were collected online and 33 were collected from direct interviews

Majority of respondents are students and full-time employees, respectively showing 37% and 37.5% of employment status Almost respondents are receiving less than 10 million income per month Majority of respondents is very young (39.5% in the age range of 18-24) Monthly pass refers to ticket which is valid for the one or few months, while single journey ticket means ticket which is valid for only one trip on BRT bus There are 118 among 200 respondents are using monthly pass, remaining are single journey ticket users More details of the result are illustrated in Table 3-2

Table 3.2 Demographic Profile of Respondents

Q’ty (N=200) Percentage (%) AGE

22 7.5

3 37.5

12

3 1.5

CHAPTER 4 RESULT ANALYSIS AND FINDINGS

4.1 Validity of measurement

Exploratory factor analysis

The reliability of measurements is evaluated based on statistics criteria on SPSS ver 22 such as EFA (exploratory factor analysis), Cronbach’s Alpha (minimum 0.6), KMO and Bartlett’s test, rotating factors with varimax, Eigenvalues

These tests have helped to reduce invalid items which are not representing measurements

EFA (Explanatory factor analysis) is necessary to identify underlying factors (latent variables) for a measurement set from a large number of measured variables In this research, EFA is applied for all samples (N=200) to determine the item-factor assignment In literature, according to Hair et al (1998), factor loading is criteria to guarantee the reality meaning of EFA

It is suggested about cut-off points, specifically, 0.32 as poor, 0.45 as fair, 0.55 as good, 0.63 as very good and 0.71 as excellent (Tabachnick and Fidell, 2007) However, in fact, there are no rules for the cut-off point This research aims to follow exploratory factor analysis, thus, the cut-off point is decided as at least 0.5 for the final scale

As explain in aforesaid part of conceptual framework, to test the effect of Service quality to Behavioral intentions, mediating and moderating effects by Customer satisfaction on the relationship between Service quality and Behavioral intentions, some single and multiple regressions will be conducted by using SPSS As a result, Varimax rotation is utilized since it is seen as the best method and most widely used for the research which focuses on identifying factors which are uncorrelated (Maskey, Fei and Nguyen, 2017)

The first run of EFA’s result implied that items of CS1 (I was interested in using this ticket system), BI1 (I would say positive things about the ticket system to other people) and BI5 (I would use BRT more for its ticket service in the coming time) are explaining in different components Hence, these 3 items were cut-off and continue another EFA with Varimax rotation

Accordingly, it is found that the remaining 34 items can be represented by 11 factors of Service Quality (identified as Interaction quality: Attitude – Behavior - Expertise, Physical environment quality: Ambient conditions – Design – Social Factor and Outcome Quality: Waiting time – Tangibles – Valence), Customer Satisfaction and Behavioral Intentions Table 4-1 is illustrating the result of Varimax rotation

Table 4.1 Rotated Component Matrix

The remaining items have adequately shown the convergence and differentiation of all items after EFA Table 4-2 (extracted from SPSS result) explained the results of KMO, Bartlett’s test of Sphericity

Table 4.2 KMO and Bartlett's Test

Kaiser-Meyer-Olkin Measure of Sampling

Bartlett's Test of Sphericity

Bartlett’s test of Sphericity is statistical test to see the presence of correlations between variables, and significant Bartlett’s test of Sphericity is needed According to the above result, Bartlett’s test of Sphericity is significant with p=0.000 (<0.05)

According to Girden (2001), those components with Eigenvalues less than 1.00 are not stable, factors for these components are not remained in the analysis The results of Eigenvalues for 11 components are valid (>1.0)

Cumulative percentage of variance is 72.898 (>50%), meaning that the component can explain 72.898% of the data variance

Reliability (Cronbach’s Alpha)

After cutoff in EFA, remaining items are utilized for reliability test (Cronbach’s Alpha) Cronbach’s Alpha measures internal consistency among a set of items in one group Alpha of 0.6-0.7 indicates acceptable level of reliability as a general accepted rule

(Ursachi et al., 2015) This research chooses to get all items with Cronbach’s alpha of

at least 0.6

Hence, the test for reliability (Cronbach’s Alpha) is implemented based on the remaining 34 items The Cronbach’s Alpha of Outcome Quality is 0.507 which is invalid comparing to the criteria of at least 0.6 As a result, all items of Outcome Quality are deleted since they are not ensuring sufficient reliability to be kept for the analysis Table 4-3 shows Cronbach’s alpha of all remaining items

Table 4.3 Exploratory Factor Analysis and Reliability Estimated Results

loadings Service Quality

(1) Interaction quality (Cronbach's alpha: 0.688)

Attitude (Cronbach's alpha: 0.768)

SQ2 The attitude of the staff demonstrates their willingness to help me 0.829 SQ3 The attitude of the staff shows me that they understand my needs 0.763

Behavior (Cronbach's alpha: 0.768)

SQ4 I can count on the staff taking actions to address my needs 0.826

SQ6 The behavior of the staff indicates to me that they understand my needs 0.756

Expertise (Cronbach's alpha: 0.767)

SQ9 The staff understands that I rely on their knowledge to meet my needs 0.782

(2) Physical environment quality (Cronbach's alpha: 0.762)

Ambient conditions (Cronbach's alpha: 0.849)

SQ10 When using the ticket system, you can rely on there being a good atmosphere 0.849 SQ11 The ticket system’s ambience is what I’m looking for in a ticketing system of public transport 0.774 SQ12 The BRT operator understands that its atmosphere in using ticket system is important to me 0.875

Design (Cronbach's alpha: 0.781)

SQ13 This BRT ticketing device’s layout never fails to impress me 0.825

SQ15 The BRT operator understands that the design of its ticketing facility is important to me 0.84

Social factor (Crobach's alpha: 0.730)

SQ16 I find that other passengers on the BRT consistently leave me with a good impression of its

SQ17 The other passengers do not affect its ability to provide me with good ticket service 0.847 SQ18 The BRT operator understands that other patrons affect my perception of its ticket service 0.753

Customer Satisfaction (Cronbach's alpha: 0.868)

CS5 The perceived service was equivalent to my ideal service 0.841

Behavioral Intentions (Cronbach's alpha: 0.874)

BI2 I would recommend the service to someone who seeks my advice 0.742 BI3 I would encourage friends and relatives to use the BRT for its ticket service 0.835 BI4 I would consider the BRT my first choice to use public transport service for its ticket service 0.76

4.2 Descriptive statistics and correlation

Basic features of data in the research is described by descriptive statistics which simply summarizes the samples and measurement Table 4-4 illustrates the correlation among all variables in the model Accordingly, Behavioral Intentions (BEHAVINT) is correlated with Customer Satisfaction (CUSTSAT) (r=0.354, p<0.01), Service Quality (SERVQUAL) (r=0.593, p<0.01), Interaction quality (INTERAC) (r=0.555, p<0.01) and Physical Environment (PHYSENV) (r=0.449, p<0.01) This explains and supports the total model with three constructs, i.e Service quality, Customer satisfaction and Behavioral intentions

In addition, the model of Service quality is confirmed by the results which shows that Service quality has positive correlation with all sub-dimensions (Interaction Quality (INTERAC) (r=0.840, p<0.01): Attitude (r=0.479, p<0.01), Behavior (r=0.634, p<0.01), Expertise (r=0.508, p<0.01) and Physical Environment Quality (PHYSENV) (r=0.850, p<0.01): Ambient Condition (r=0.674, p<0.01), Design (0.568, p<0.01), Social Factor (r=0.536, p<0.01)) under significant level of 0.01 This result has reflected the reliability of the measurements of Service Quality model

All sub-dimensions in Interaction Quality and Physical Environment Quality are also reflecting positive and significant correlation with Interaction Quality and Physical Environment Quality (p<0.01) This supports the hierarchical model in measuring Service Quality by sub-dimensions

Table 4.4 Descriptive Statistics and Correlations

10 Design 239** -0.01 568** 263** 245** 146* 0.125 691** 235** 1

11 Social Factor 305** 202** 536** 332** 235** 149* 277** 570** 321** 145* 1

** Correlation is significant at the 0.01 level (2-tailed)

* Correlation is significant at the 0.05 level (2-tailed)

These results allow the next analysis to evaluate the effect of independent variables (SERVQUAL, CUSTSAT) on dependent variable (BEHAVINT) by four (04) tests in item 4.3

4.3 Hypotheses tests and result analysis

Five (05) tests have been executed to inspect the four (04) hypotheses of this research as following:

H1a: Service quality has positive effect on Customer satisfaction

H1b: Service quality has positive effect on Behavioral intentions

H2: Customer satisfaction has positive effect on Behavioral intentions

H3: Customer satisfaction has moderating effect on the relationship between Service

quality and Behavioral intentions

Single and Multiple linear regressions were executed by SPSS ver 22, resulting in

summarizing information in the Table 4-5

Table 4.5 Linear regression results

Customer Satisfaction (CUSTSAT)

Service Quality (SERQUAL)

.593 (p=0.000)

.547 (p=0.000)

.238 (p=0.15)

.182 (p = 0.01)

Customer Satisfaction (CUSTSAT)

.254 (p=0.000)

-0.376 (p=0.243)

.354 (p=0.000)

.758 (p=0.048)

F Value

107.374 (p=0.000)

69.59 (p=0.000)

48.423 (p=0.000)

28.280 (p=0.000)

6.773 (p = 0.01)

Table 4-5 summarizes the linear regression results for Behavioral Intention and Customer

Satisfaction as Dependent variables based on below functions

Test 1 BEHAVINT f(SERQUAL)

Test 2 BEHAVINT f(SERQUAL, CUSTSAT)

Test 3 BEHAVINT f(SERQUAL, CUSTSAT, SERQUAL x CUSTSAT) Test 4

Test 5

CUSTSAT f(SERQUAL) BEHAVINT f(CUSTSAT) Test 1 is conducted to evaluate the relationship between Service Quality as independent variable and Behavioral Intentions as dependent variable The result of regression for Test 1 showed that Service Quality has positive effect on Behavioral Intentions of Customers (beta = 0.593) and this relationship is statistical adequacy (p<0.001), F value

=107.374 with p<0.001 Accordingly, H1b is supported by this test result

Test 2 is done to simultaneously evaluate the relationship of independent variables (Service Quality and Customer Satisfaction) and dependent variable (Behavioral Intentions) According to the result summarized in the table, both independent variables showed positive effect on Behavioral intentions, respectively, beta = 0.547 for Service Quality with p<0.001 and beta = 0.254 for Customer Satisfaction with p<0.001, F value = 69.59 (p<0.001)

Test 3 is implemented to analyze the moderating role of Customer Satisfaction in the relationship between Service Quality and Behavioral Intentions According to the result, the item of Customer Satisfaction x Service Quality showed beta value of 0.754 with p<0.05, or statistical adequacy, F=48.423 with p=0.000 This result, together with Test 2,

supports H3, confirming the moderating role of Customer Satisfaction in the relationship

between Service Quality and Behavioral Intentions

Test 4 is executed between Customer Satisfaction (Dependent variable) and Service

Quality (Independent variable) The result in Test 4 has supported H1a in the model,