RELATIONSHIP BETWEEN SERVICE QUALITY, CUSTOMER

SATISFACTION, BRAND IMAGE, AND TRUST WITH CUSTOMER

LOYALTY OF GOJEK TRANSPORTATION SERVICE IN
YOGYAKARTA CITY

A THESIS

Presented as Partial Fulfillment of the Requirements

to Obtain the Bachelor Degree in Management Department

By:
Fatih Setia Utami
Student Number: 08311389

DEPARTMENT OF MANAGEMENT
INTERNATIONAL PROGRAM
FACULTY OF ECONOMICS
UNIVERSITAS ISLAM INDONESIA
YOGYAKARTA
2017
ACKNOWLEDGEMENT
 In the name of Allah Almighty, The Most Gracious, I would like to express

my praise completely to Allah SWT because of His marvelous love and guidance

upon the accomplishment of this thesis. My honor also goes to Prophet Muhammad

SAW for his blessing and enlightening direction.

This thesis is written in order to fulfill the requirements to finish study as

Bachelor Degree in Management Department, Faculty of Economics, Universitas

Islam Indonesia. Realizing that all of this hard work could never be done alone, I

would like to deliver my sincere gratitude to:

1. Drs. Anas Hidayat, MBA., Ph.D as my Content Advisor, thank you for the

great discussion by your support, critiques, and advice during this thesis

period and my study.

2. Alfi Zakiya, S.Kom., S.Pd., as my language advisor thank you for your

guidance, patience, and time to make this thesis written in English properly.

3. My beloved parents: Maryanto, my father. Thank you for giving so much

learnings in my life. Farchatul Kifyati , my mother, who are not only the

sharpest mother ever but also the one whom I have to confess that she is the

real superwoman. Thank you for everything you’ve given to me.

4. My husband, Dimas Aulia Hendrawan, thank you for all the time you’ve

given to taking good care our little family. And my dearest son, Faeyza

Athariz Hendrawan, the greatest gift from Allah SWT, I love you to the

moon and back!

5. My brothers, Fakih, Falih, Farih. Thank you for being the best listener.

6. My Management IP 08 friends. Thank you for the most beautiful,

meaningful, and unforgetable moment that you’ve brought to my life guys.

7. International Program members, the seniors and the juniors, Accounting

2008 friends, thanks for the interesting friendship and giving the good

atmosphere in International Program area.

Finally, further comments, critiques, and suggestion will be appreciated in order

to make this thesis better. Hopefully, this thesis could open a wider knowledge

on the field of study.

Alhamdulillahirabbil’alamiin

Yogyakarta, August ,2017

Fatih Setia Utami

TABLE OF CONTENTS
Page of Title .............................................................................................................i

Approval Page..........................................................................................................ii

Legalization Page....................................................................................................iii

Declaration of Authenticity ....................................................................................iv

Acknowledgement ...................................................................................................v

Table of Contents ..................................................................................................vii

List of Table............................................................................................................xi

List of Figure .........................................................................................................xii

List of Appendices ...............................................................................................xiii

Abstract.................................................................................................................xiv

Abstrak...................................................................................................................xv

CHAPTER I: INTRODUCTION.........................................................................1

1.1. Background of Study ................................................................................1

1.2. Problem Statement....................................................................................7
1.3. Research Questions...................................................................................8
1.4. Research Objectives..................................................................................8
1.5. Limitation of The Study............................................................................8
1.6. Research Contribution..............................................................................9
1.7. Systematical of Writing.............................................................................9

CHAPTER II: LITERATURE REVIEW..........................................................11

2.1 Customer Loyalty.............................................................................................11

2.1.1.Understanding Customer Loyalty...........................................................11

2.2 Service Quality.................................................................................................14

 2.2.1 Understanding Service Quality...................................................................14

2.2.2 Element of Service Quality.........................................................................16

2.3 Customer Satisfaction......................................................................................18

2.3.1 Understanding Customer Satisfaction.......................................................18

2.3.2 Measuring Customer Satisfaction............................................................19

2.4 Brand Image.....................................................................................................20

2.5 Trust.................................................................................................................23

2.5.1 Understanding Trust...............................................................................23

2.5.2 Measuring Trust...................................................................................24

2.6 Hypothesis Formulation...................................................................................25

2.6.1 Relationship Quality Service, Customer Satisfaction, Brand Image,

and Trust With Customer Loyalty..............................................................26

2.7 Conceptual Framework....................................................................................27

CHAPTER III: RESEARCH METHOD...........................................................29

3.1 Type of Study...................................................................................................29

3.2 Research Subject..............................................................................................29

3.2.1 Population...................................................................................................29

3.2.2 Sampling Design........................................................................................31

3.3 Research Setting .............................................................................................31

3.4 Research Instrument........................................................................................31

3.4.1 Data Collection Method ...........................................................................31

 3.4.2 Validity......................................................................................................32
3.4.3 Reliability..................................................................................................32

3.5 Research Variable And Operational................................................................33

3.5.1 Independent Variable...............................................................................33

3.5.2 Dependent Variable.................................................................................37

3.6 Technique of Data Analysis.............................................................................37

3.7 Assumption of Multiple Regression.................................................................40

3.7.1 Linearity...............................................................................................41
3.7.2 Normality ............................................................................................42
3.7.3 Homoscedasticity.................................................................................43
3.7.4 Collinearity...........................................................................................45

3.8 Hypothesis Test................................................................................................47

3.8.1The goodness of Fit of The Regression Model......................................47

3.8.2 T test .................................... ................................................................49
3.8.3 F test.................................... .................................................................49
CHAPTER IV RESEARCH FINDINGS AND DISCUSSIONS......................50
4.1 Descriptive Statistic...................................................................................50
4.1.1Respondents’ Demographic Characteristics.........................................50
4.1.1.1 Repondents’ Gender.......................................................................50
4.1.1.2 Respondents’ age..........................................................................51
4.1.1.3 Respondents’ Provenance.............................................................52

4.1.1.4 Respondents’ Occupation..............................................................54

4.1.1.5 Respondents’ Intensity of Using Gojek Service...........................55

4.1.1.6 Data of Using The Other Gojek Service.......................................56

4.1.1.7 Respondents’ Income/ Salary Per Month......................................57
4.2 Result of Hypothesis Testing.....................................................................59
 4.2.1 Linear Regression Analysis..................................................................59
1.Normality Data.................................................................................59
2. Simple Linear Regression Equation................................................59
3. The Goodness of Fit of Regression Model ....................................60
4. F test and t test.................................................................................62
4.2.2 Multiple Regression Analysis..............................................................63
1. Regression assumption......................................................................63
a. Linearity........................................................................................63

b. Normality......................................................................................64

c. Homoscedacity..............................................................................65

d.Colinearity......................................................................................66

2. Multiple Regression Linear Analysis...............................................69

3. The Goodness of fit Regresssion Model..........................................71

4. Result of Hypothesis Testing...........................................................73

4.1 F test........................................................................................73
4.2 T test.........................................................................................74
4.3 Hypothesis H1..........................................................................75
4.4 Hypothesis H2..........................................................................75
4.5 Hypothesis H3..........................................................................76
4.6 Hypothesis H4..........................................................................76
4.7 Discussion and Implication......................................................76
CHAPTER V: CONCLUSIONS, LIMITATIONS, AND
RECOMMENDATIONS................................................................81
5.1 Conclusions ...............................................................................................81
5.2 Limitations.................................................................................................82
5.3 Recommendations......................................................................................82
REFERENCES.................................................................................................84
APPENDICES..................................................................................................89
LIST OF TABLE
Table 4.1. The Frequencies Distribution of the Respondents’ Gender .................50

Table 4.2. Frequency Distribution of Respondents’ Age.......................................51

Table 4.3. Respondents’ Provinces .......................................................................53

Table 4.4. Frequency Distribution of Respondents’ Occupation ........................54

Table 4.5. The table of Respondents’ Intensity in Using The Gojek Service .......55

Table 4.6. Respondents’ Data of Using The Other Gojek Services.......................56

Table 4.7. Respondents’ Income/ Salary ..............................................................58

Table 4.8. Simple Linear Regression Analysis......................................................60

Table 4.9. Coefficient Determination....................................................................61

Table 4.10. F test and t Test...................................................................................62

Table 4.11. The Result of multicolinierity test .....................................................68

Table 4.12. Results of Durbin Watson Test ..........................................................69

Table 4.13. Multiple Regression Analysis ............................................................70

Table 4.14. Determination Coefficient .................................................................72

Table 4.15. F test ...................................................................................................74

Table 4.16. t test ....................................................................................................74

LIST OF FIGURE
Figure 2.1. Conceptual Framework........................................................................28

Figure 4.1. Diagram of Respondents’ Gender......................................................51

Figure 4.2. Diagram of Respondents’ Age.............................................................52

Figure 4.3. Diagram of Respondents Province......................................................54

Figure 4.4. Diagram of Respondents’ Occupations .............................................56

Figure 4.5. Diagram of Respondents’ Intensity in Using The Gojek Service......56

Figure 4.7. Bar Chart Diagram of Respondents’ Income/ Salary ........................58

Figure 4.8. P-P Plot................................................................................................59

Figure 4.9. P-P Plot 2 ...........................................................................................63

Figure 4.10. Homoscedasticity Test.......................................................................66

LIST OF APPENDICES
Appendix 1.
Questionaire
....................................................................................................................................
90

Appenddix 2. Regression
Output
....................................................................................................................................
94
 RELATIONSHIP BETWEEN SERVICE QUALITY, CUSTOMER
SATISFACTION, BRAND IMAGE, AND TRUST WITH COSTUMER
LOYALTY OF GOJEK TRANSPORTATION SERVICE IN
YOGYAKARTA CITY

ABSTRACT

Customer satisfaction is important thing. Service quality is one of the factors

that influence customer satisfaction. Satisfied customers will consequently enhance
the brand image and trust in the brand. This research analyzed the influence of
service quality on customer satisfaction. Furthermore, this research also analyzed
the influence of customer satisfaction, brand image and trust on consumer loyalty.
This research used quantitative descriptive study that used primary data as
the source of data. The data was obtained by offline questionnaire using Likert scale
and had 100 respondents as sample. Data tool used in this research were Simple
Linear Regression and Multiple Linear Regression.
From the research it was found that, quality of service influenced positively
and significantly on customer satisfaction. Beside that, it also positively influence
and significantly influenced customer satisfaction, brand image and trust in
consumer loyalty.

Keywords: Service Quality, Customer Satisfaction, Brand Image, Trust And

Customer Loyalty
 RELATIONSHIP BETWEEN SERVICE QUALITY, CUSTOMER
SATISFACTION, BRAND IMAGE, AND TRUST WITH COSTUMER
LOYALTY OF GOJEK TRANSPORTATION SERVICE IN
YOGYAKARTA CITY

ABSTRAK

Kepuasan pelanggan adalah suatu hal yang sangat penting. Kualitas layanan
adalah salah satu faktor yang mempengaruhi kepuasan pelanggan. Pelanggan yang
puas terhadap suatu jasa akan selalu menggunakan jasa tersebut yang akibatnya
akan meningkatkan citra merek perusahaan dan kepercayaan pelanggan terhadap
merek tersebut. Penelitian ini menganalisis pengaruh kualitas layanan terhadap
kepuasan pelanggan. Lebih jauh, penelitian ini juga menganalisis pengaruh
kepuasan pelanggan, citra merek dan kepercayaan terhadap loyalitas konsumen.
Penelitian ini adalah penelitian deskriptif kuantitatif yang menggunakan
data primer sebagai sumber data. Data diperoleh dengan teknik kuesioner offline
menggunakan skala likert yang menghasilkan 100 responden sebagai sampel. Alat
analisis data yang digunakan dalam penelitian ini adalah Regresi Linear Sederhana
dan Regresi Linear Berganda.
Kualitas layanan ditemukan berpengaruh secara positif dan signifikan
terhadap kepuasan pelanggan. Sejalan dengan hal tersebut, ditemukan juga bahwa
secara positif dan signifikan kepuasan pelanggan, citra merek dan kepercayaan
berpengaruh terhadap loyalitas pelanggan.

Kata Kunci: Kualitas Pelayanan, Kepuasan Pelanggan, Citra Merek,

Kepercayaan Dan Loyalitas Pelanggan
 CHAPTER 1

INTRODUCTION

This chapter discusses about study background, problem statement, research

question, research objective and research contribution. It highlights the importance

of this study. User data and the reason of using gojek is also discussed in this

chapter. The given historical fact and future prediction are supported by the theory

and previous research findings, several potential problems identified and it was

derived in the reseach question. The last section of this chapter discusses the

research objective that needed to be accomplished and the research contribution to

the others.

1.1. Background of the Study

The customer loyalty has been the objective or goal for the marketer and the

company. Customer which continues to buy from the company, increases profit for

business in the future and can provide a positive word of mouth which are the main

keys of customer loyalty’s indicator (Hidayat et al., 2015). Customer loyalty is

critical to growth, corporate survival and an effort to build customer loyalty as well

as a foundation in developing sustainable competitive advantage (Han Cung, et al.,

2015). Thus it can be said that customer loyalty is an important object that must be

achieved so that companies can face strong competition. Achieving customer

loyalty will help the company to continue to exist and ensure the survival of the
company. The empirical findings of studies investigate the relationships between

service quality, customer satisfaction and their outcomes that had found broad

support (Loveman, 1998: 18; Anderson & Mittal, 2000). The relationships was

described in the Service-Profit Chain which was suggested by Heskett, Sasser and

Schlesinger (1997).

The loyalty index could provide input to managers in determining market

segmentation (Denio, 2006). Several factors influencing customer loyalty were

customer satisfaction, service quality and customer perception value (Juan & Yan,

2009). Strengthening the research results of Juan & Yang, Wan Lee (2010) and

Khatibi et al. (2002) proved that product quality and brand image was a strong

marketing antesedan that supports loyalty and success in high-tech

products/services markets. Supporting that opinion, Kursunluoylu (2011), Seok Lee

(2010), and Sophoasin (2010) proved that customer service is one of the tools used

to create customer satisfaction and loyalty. Thus, to increase customer loyalty, the

company could improve service quality (Liang Kheng et al., 2010) which will

further increase customer's trust and strengthen brand image (Tariq and Moussaoui,

2009). The opinions of several researchers above support the results of research by

Lorenzo et al. (2010) which stated that the provision of quality services leads to

competitive advantage, organizational growth, and increased profitability. The

results also showed that service quality was positively associated with customer

loyalty. Customer satisfaction has been considered to be based on the customer’s

experience on a particular service encounter (Cronin & Taylor, 1992). It was in line

with the fact that service quality was a determinant of customer satisfaction,
because service quality came from the outcome of the services of service providers

in organizations.

Another author stated in his theory that definitions of consumer satisfaction

was related to a specific transaction (the difference between predicted service and

perceived service) in contrast with ‘attitudes’, which were more enduring and less

situational-oriented (Lewis, 1993, p. 4-12). This was in line with the idea of Berry

et al. (1985, p. 106-107). Regarding the relationship between customer satisfaction

and service quality, Oliver (1993) firstly suggested that service quality would be

antecedent to customer satisfaction regardless of whether these constructs were

cumulative or transaction-specific. Some researchers had found empirical supports

for the view of the point mentioned above (Anderson & Sullivan, 1993; Fornell, et.

al., 1996; Spreng & Macky 1996); where customer satisfaction came as a result of

service quality. In relating customer satisfaction and service quality, researchers

had been more precise about the meaning and measurements of satisfaction and

service quality. Satisfaction and service quality had certain things in common, but

satisfaction generally was a broader concept, whereas service quality focused

specifically on dimensions of service (Wilson et al., 2008, p. 78). Although it was

stated that other factors such as price and product quality can affect customer

satisfaction, perceived service quality was a component of customer satisfaction

(Berry. et al. 1985, p. 106-107). This theory complied with the idea of Wilson et

al., (2008) and had been confirmed by the definition of customer satisfaction which

was presented by other researchers.

 Davies et al., (2003) said that the strong brand image would make customer

satisfied. Brand image had something to do with product appearance. Thus, the

customers were satisfied with it, they would become loyal to the brand even though

they had other optional brands (Vinhas & Faridah, 2008). Earls et al. (2004) also

said that strong brand image could make customers loyal. But whenever customers’

need change and the brand could not respond these changing needs, according to

Steel (2004), customers tended to switch another brand (in Miller & Muir, 2004).

Building a positive brand image could influence repeat purchase because a strong

brand image could lead to customer loyalty. Furthermore, Schultz (2005) stated that

customers who were loyal to certain brand would recommend it to other people and

not easily influenced by competitors to make purchases (in Sondoh et al., 2007).

Based on the framework above, there was a relation between brand image and

customer loyalty. According to Earls et al., (1998), a strong brand image would

make customer loyalty stronger (in Miller & Muir, 2004).

Customer satisfaction had a positive effect on an organization’s

profitability. According to Hoyer and MacInnis (2001), satisfied customers form

the foundation of any successful business as customer satisfaction led to repeat

purchase, brand loyalty, and positive word of mouth. However, Bowen and Chen

(2001) said that having satisfied customers was not enough, there had to be

extremely satisfied customers. This was because customer satisfaction must lead to

customer loyalty. Bansal and Gupta (2001) stated that building customer loyalty

was not a choice any longer with businesses, it was the only way of building

sustainable competitive advantage. Building loyalty with key customers had

become a core marketing objective shared by key players in all industries catering

to business customers. Loyalty was vulnerable because even if the customers were

satisfied with the services rendered by the financial institutions, there was always

an element of defect if they thought they could get better value for money in other

institutions. Satisfaction was essential but not enough to gauge loyalty among

customers. In other words, we could have customer satisfaction without loyalty but

it was difficult to comprehend having loyalty without satisfaction. Therefore, all

organizations should try and satisfy their customers and to ensure their loyalty.

The organization that involved in cause-related marketing activities

generated profits in terms of finance and also in terms of customer positive word of

mouth, positive customer internal relations and better value from customer. On

contrary, it generated financial resources for charitable organizations (Lingen,

2008). Basically, the corporate image of the organization influenced positively the

customer trust on the organization. Customers trusted more on highly reputed

organization, while marketing organization needed more emphasis to correspond

organizational distinctiveness more than the product features (Keh & Xie, 2009).

When the association wanted to get competitive benefit and first-class name

in the society, one fundamental instrument which smooths the progress of the

organization was cause-related marketing. That was the practice through which

people remember the organization in their minds and the institute which was

involved in cause related marketing that built up strong trust environment between

the customer and organization. At the same time the business that was attached to

great cause (long-term and short-term goals) got good name in the society and made
itself more trust worthy in their customer eyes (Papasolomou, Demetriou &

Crowther, 2006). There was no disbelief that cause related marketing to be

favorable for the profit and nonprofit association which the intention was to serve

up the humanity in the superior way, although at the same instance it could generate

an enormous trouble for those firms who desired to bring it into play just for its own

purpose (Svensson & Wood, 2011).

The object of this research was transportation service. Transportation

service was needed by the people on daily life. Along with the development of

technology and internet users that was rapidly increasing then people who do no

longer need to spend more time and effort to find public transportation, Gojek

offered online reservation through applications on smartphones that allowed users

to perform transactions. An online application-based taxi service is growing very

rapidly in Yogyakarta City. Until April 2017 there are several operators, among

others Go-car, Grab and Uber-X with the number of cars that reached more than

5000. Service quality has also been shown to be an important thing of customer

satisfaction from a theoretical view point (Heskett, Sasser & Schlesinger, 1997).

PT Gojek Indonesia as a pioneer company that initiated services online

motorcycle transportation was able to seize only proven market share within three

months of Go-Jek service that had been downloaded as many as 100,000 time. In

January 2015, the app was downloaded 32,360 times, in February 81,843 times and

March 131,795 times (Harry, 2016). The reason that society today prefer to use Go-

Jek as an online motorcycle taxi was a public perception that judge that online
motorcycle taxi is more practical than conventional motorcycle taxi, its users can

booked via the internet without having to jump in the field to find a motorcycle taxi.

The official service of the project officially expanded to Jogja on Monday

November 16, 2015. Vice President of Operations Division PT Gojek Indonesia

chose Jogja because it considers Yogyakarta as the number two tourist destination

after Bali (Admasari, 2015). Gojek continues to grow both in terms of users and

service providers. Customers feel safe because Go-Jek has been integrated below

institutional shade. Thus, the customers trust was greater, as well as standardized

tariffs so customers do not have to bargain with drivers. Another basic thing that

distinguishes between Gojek as an online motorcycle taxi with other online

motorcycle taxi was the brand labeling (branding) "Go-Jek". The Go-Jek brand was

going to be a differentiator with its competitors. Brand identity indicated the

advantages of a product that was not owned by another product. This research

focused on customer loyalty of Gojek with various factors influencing it, including

service quality, customer satisfaction, brand image and trust.

1.2. Problem Statement

Consumers now had a large selection of transportation tools. Gojek was the

main choice because it had many advantages compared to other types of transport

in the perimeter. The question was how can gojek be the main choice? Specific

consumers were loyal to Gojek. The next question is, what are the variables that

cause consumers to be loyal to Gojek? This research examined several related

variables, including service quality, customer satisfaction, brand image, and trust.
1.3. Research Questions
Based on the definition of the above problem, this research focused on

consumer loyalty. Therefore, the research questions were as follow:

1. Does service quality influence possitively on customer satisfaction?

2. Does consumer satisfaction influence possitively on consumer loyalty?

3. Does brand image influence possitively on consumer loyalty?

4. Does trust influence possitively on consumer loyalty?

1.4. Research Objectives

The specific objectives of this research were:

1. To analyse the influence of service quality on customer satisfaction.

2. To analyse the influence of consumer satisfaction on consumer loyalty.

3. To analyse the influence of the brand image on consumer loyalty.

4. To analyse the influence of trust on consumer loyalty.

1.5 Limitations of the Study

This research was addressed through the sample. The sampling frame was

the resident of Yogyakarta City, therefore it may not be generalizable to other

residents. The sample was people from the age of 17 to 55 years old and had ever

used the services of Gojek. Furthermore, this study did not explain the fact that the

population surveyed may be more familiar with the services of other online

motorcycle bikes.
1.6. Research Contributions

This research may give benefits to the following party:

1. Researcher

The results of this research can be used to measure consumer loyalty based

on various variables such as service quality, customer satisfaction, brand

image, and trust. Furthermore, these findings can be further analyzed using

a variety of other service products.

2. Practitioners of marketing

This research can help the decision maker to develop a systematic approach

to consumer loyalty.

1.7 Systematical Of writing

Chapter 1. Introduction

This chapter describes the background of the research, research question,

objective and contribution of the research, as well as the systematic writing.

Chapter 2 Review of Related Literature

This chapter is a description of the basic theory underlying service quality,

customer satisfaction, brand image, trust and its influence on customer loyalty,

previous research studies, hypotheses and conceptual framework.

Chapter 3 Research Method

In this chapter the research methodology described the study design,

population and sample. The instrument used to collect the data, including

methods implemented to maintain the validity and reliability of the instrument.

Chapter 4 Data Analysis and Discussions

This chapter describes the analysis of data followed by a discussion of the

research findings. The findings are relate to the research questions that guided

the research.

Chapter 5 Conclusions and Recommendations

This chapter consists of the conclusions and suggestions for Gojek

transportations and future research.

 CHAPTER II

REVIEW OF LITERATURE

2.1 Costumer Loyalty

2.1.1 Understanding Costumer Loyalty

Denio (2006) proved that the loyalty index can provide input to managers in

determining market segmentation. Customer loyalty is a function of customer or

customer satisfaction. According to Schanaars in Tjiptono (2007), basically the

purpose of business is to create satisfied customers. The creation of customer

satisfaction can provide several benefits, including some relationships between the

company and its customers become harmonious, provide a good basis for repeat

purchase and the creation of customer loyalty, and form a recommendation by word

of mouth. According to Tjiptono (2007), loyalty is a commitment of customers to

the store, brand or supplier based on positive attitude reflected in the form of

consistent repeat purchases.

Jennie in Nuraini (2009) stated that loyal customers are those who are very

satisfied with certain products and have enthusiasm to introduce them to anyone

they know. Further, at a later stage, such loyal customers will expand their "loyalty"

to other products made by the same manufacturer. In the end, they are costumers

who are loyal to certain producers forever. Griffin (2002) said that loyal customers

are people who make repetition of purchasing regularly, buy between product lines

and services, refer to others, and shows immunity to the pull of competitors.
 Customer loyalty and willingness to buy a product continuously at the same

retailer and can occur if the costumer is satisfied with the performance of the

company. According to Schnaars in Tjiptono (2007), basically the purpose of a

business is to create a satisfied customer. The creation of customer satisfaction can

provide benefits among others, the relationship between the company and the

customer becomes harmonious, providing a good foundation for repeat purchase

and the creation of customer loyalty and it will make a word of mouth

recommendation that benefits the company.

The process of a person to be loyal to the product of a company turned out

through several stages. This process lasts long enough, with emphasis and attention

being certainly different for each stage because each stage has different needs. By

paying attention to each stage and meeting the needs in each of these stages, the

company has a greater opportunity to form potential buyers to be loyal customers

of the company. Thus, it can be said that customer loyalty is an important object

that must be achieved so that customer loyalty will help the company to continue

to exist and ensure the survival of the company. By the end, customer satisfaction

came as a result of service quality. According to Grifin (2005), the concept of

customer loyalty can be divided into six levels, namely:

1. Suspect

Covering everyone who may buy the services of a company is referred to as

a suspect believing that they will buy, even if they do not know about the

company and the services it offers.

2. Prospect

It is people who have a need for certain services, and have the ability to buy

it. These prospects, even if they make a purchase, they already know the

existence of the company and the services offered because someone has

recommended the service.

3. Customers

In this level, costumers have made a transaction with the company, but do not

have a positive feeling towards the company, loyalty at this stage is not

visible.

4. Clients

At this stage, costumers have bought back the company, they have positive

feelings towards the company. Also, at this stage, loyalty to the company is

already visible.

5. Advocates

At this stage, the client actively supports the company by providing

recommendations to others to want to buy at the company.

6. Partners

At this stage there has been a strong and mutual beneficial relationship

between the company and the costumer and also the customer will reject the

products / services of other companies.

2.2 Service Quality

2.2.1 Understanding Quality

Quality is a form of assessment of the product to be purchased, whether it

meets the costumers expectance (Angipora, 2002: 162). Another definition of

quality is a dynamic condition associated with products, services, people, processes,

and environments that meet or exceed expectations (Tjiptono, 2006). According to

Cannon et al., (2008), quality is the ability of the product to satisfy the needs or

desires of the customer.

Quality is often regarded as a relative measure of the good of a product or

service consisting of design quality and conformity quality. Quality design is a

function of product specification, while the quality of conformity is a measure of

how far a product is able to meet the requirements or specifications of quality that

have been set (Tjiptono, 2006: 51). The quality of each product produced is one of

the elements that should receive genuine attention from the company, if the

company wants to win a competition in business (Angipora, 2002). Demands on

the quality of a product has become a must that must be met by the company. If not,

costumers will switch to other competing products that are considered to have better

product quality.

Therefore, the quality issues of the resulting product should be the

commitment of the management, which should then be clearly stated in other

management functions to be implemented. Thus, it become the focus of the overall

operational of the company as a whole. Beside that, the company should not be

complacent if it has produced a product that has adequate and competing quality,
but rather a good product quality as the result of continuous improvement that never

ends. This is in line with the ever-changing needs and desires of costumers who

never break. Moreover, the demand for the quality of products at this time and the

future has become a worldwide demand from the international world with the

provision of International Standard Organization (ISO) which benefits the

company's products that can be accepted internationally or only locally (Angipora,

2002: 174) .

Quality is often regarded as a relative measure of the good of a product or

service consisting of design quality and conformity quality. Quality design is a

function of product specification, while the quality of conformity is a measure of

how far a product is able to meet the requirements or specifications of quality that

have been set (Tjiptono, 2006: 51).

2.2.2.Quality of Service (Service Quality)

Quality of service is a statement of attitude, the relationship resulting from

the comparison between expectations (expectations) and performance (outcome)

(Usmara, 2003: 231). Another definition of service quality is the effort to meet the

needs and desires of customers and the accuracy of delivery to compensate for

customer expectations (Tjiptono, 2006: 59). Service quality has been defined as ‘the

delivery of excellent or superior service relative to customer expectation’

(Sophonsiri, 2010). Gronroos (1994) in Sophonsiri (2010) indicated that service

quality can be classified into two distinctive dimensions: technically and functional.
 According to Tjiptono (2006), the quality of service consists of the level of

excellence expected and control over the level of excellence to meet customer

desires. In other words, there are two major factors affecting service quality, ie

expected service and perceived service or service quality expected and service

quality received or perceived. If the services received or perceived in accordance

with the expected, the quality of services perceived good and satisfactory. If

services received or perceived exceed customer expectations, the quality of service

is perceived as the ideal quality. Conversely, if the service received lower than

expected, the quality of service is perceived as poor quality. Quality must start from

customer needs and end in customer perception. This means that a good quality

image is not based on the perspective or perception of the service provider but on

the customer's perspective or perception. Customers determine the quality or not of

a service. Thus, the quality of services depends on the ability of service providers

to meet customer expectations consistently.

2.2.3. Element of Service Quality

Paul and Donnelly (2007) in his book Marketing Management: Knowledge

and Skill Put forward in Quality Customer General Service used several attribute

following factors: Tangibles (of physical facilities, equipment, employees, and

means of communication); Reliability (the ability to provide that service promised

promptly, accurately, and satisfactorily); Responsiveness (is the desire of the staff

and employees to help customers and provide service with response); Assurances

(include knowledge, skills, modesty, the capabilities and trustworthiness of the staff
are free from danger, risk or doubt); Empathy(the range of ease in making

relationships, good communication, personal attention, and understanding needs

the customers).

SERVQUAL instrument (Berry et al., 1985) is a research-based set of general

expectations that customers have for their service providers. SERVQUAL consists

of five dimensions involving the core features of service provision. These

dimensions include reliability, tangibles, responsiveness, assurance and empathy.

Berry et al., (1985) proposed that meeting or exceeding customer expectations in

each of these key areas can improve customer satisfaction. According Sophonsiri

(2010), these dimensions are defined as:

(i) Reliability, the ability to perform the promised service dependably and

accurately.

(ii) Tangibles, the appearance of physical facilities, equipment, personnel and

communications materials.

(iii) Responsiveness, the willingness to help customers and to provide prompt

service.

(iv) Assurance, the knowledge and courtesy of employees and their ability to

convey trust and confidence.

(v) Empathy, the provision of caring, individualized attention to customers

2.3 Costumer Satisfaction

2.3.1. Understanding Costumer Satisfaction

Customer satisfaction is a feeling of pleasure or disappointment to someone

who emerged after comparing the performance (result) of the product with the

expected performance (Kotler and Keller, 2009: 177). Meanwhile, according to

Sumarwan (2003: 322), costumer satisfaction is the impact of the comparison

between customer expectations before making a transaction with costumers to

obtain purchased products. Tjiptono and Chandra (2005: 195) defined costumer

satisfaction as the costumer's response to the evaluation of the perceived difference

/ uncertainty between the previous expectations and the actual performance of the

perceived product after its use.

Buyer's satisfaction is known after making a purchase, depending on bid

performance in meeting buyer expectations. If performance is below expectations,

customers are not satisfied. If performance meets expectations, customers are

satisfied. If performance exceeds expectations, customers are very happy or happy

(Kotler and Keller, 2009: 177). While customer-centered companies seek to create

high customer satisfaction by lowering their prices or improving their services, the

results are likely to result in lower profits. Companies may be able to increase their

profitability in other ways rather than increasing satisfaction. The company has

many stakeholders, including employees, suppliers, suppliers, and shareholders.

The company spends more costs aimed at increasing customer satisfaction (Kotler

and Keller, 2009: 177).

 Customer loyalty is winning the confidence of the customer in favor of an

organization such as the relationship becomes a win-win situation for both the

organization as well as the customer. Customer loyalty is not a process that finishes

with the customer joining the loyalty program but actually a process that starts with

the customer joining the same. Customer loyalty is something more of what an

enterprise must get from the customer. As opposed to what the name suggests, it is

not just something that the customer has to build towards the enterprise (Ganiyu et

al., 2012)

2.3.2 Measuring Costumer Satisfaction

In general, satisfaction is a person’s feelings of pleasure or disappointment

that result from comparing a product’s perceived performance (or outcome) to

expectations. If the performance falls short of expectations, the customer is

dissatisfied. If it matches expectations, the customer is satisfied. If it exceeds

expectations, the customer is highly satisfied or delighted.Customer assessments of

product performance depend on many factors, especially the type of loyalty

relationship the customer has with the brand. Consumers often form more favorable

perceptions of a product with a brand they already feel positive about. Many

companies are systematically measuring how well they treat customers, identifying

the factors shaping satisfaction, and changing operations and marketing as a result.

A highly satisfied customer generally stays loyal longer, buys more as the company

introduces new and upgraded products, talks favorably to others about the company

and its products,pays less attention to competing brands and is less sensitive to
price, offers its ideas. High satisfaction or delight creates an emotional bond with

the brand or company, not just a rational preference (Kotler & Keller, 2012:128).

According to Kottler in Suwardi (2011), the key to retain customers is

consumer satisfaction. The indicators of consumer satisfaction can be seen from:

1. Re-purchase

Buy back, where the customer will return to the company to find goods /

services.

2. Creating word-of-mouth

In this case, customers will say good things about the company to others.

3. Creating brand image

Customers will pay less attention to brands and advertisements from

competitors' products.

4. Creating purchasing decisions at the same company

Purchasing other products from the same company.

2.4 Brand Image

Brand image has different definitions as expressed by some experts. This is

due to different views of the result of image brand image built in Costumers’ minds.

Here are some brand image notions from several sources:

1. Brand image is a set of Costumer beliefs about a particular brand (Kotler and

Armstrong, 2001)

2. Brand image is a collection of perceptions about an interrelated brand that exist

in the human mind (Ouwersoot and Tudorica, 2001)

3. Brand image can be defined as a perception about brand as reflected by the brand

association held in Costumer’s memory. This means that brand image is a brand

perception represented by a brand association existing in costumer’s memory

(Keller, 1998) as follow:

4. Brand association is anything linked in memory to a brand. This sense indicates

that brand association is something that relates to the brand in costumer’s

memory. (Aaker, 1991)

From the definition of the brand image above, it can be concluded that the

brand image is a collection of impressions that exist in the minds of costumers about

a brand. People who love images (image) believed that a strong brand image can

create a preference in the absence of other differences (Kotler, 2006 p.94).

According to Shimp (2003), there are three parts of image measurement on 16

brands. The first part is attribute. Attributes are the characteristics or aspects of the

advertised brand. Attributes are also divided into two parts: things that are not

related to the product (eg. price, packaging, user, and image of use), and things

related to the product (eg. color, size, design).

Brand image influences product appearance; thus, customers are satisfied and

will become loyal to the brand even though they have other optional. There is a

strong connection between brand image with customer loyalty. The sustained

loyality can be a customer trust. Satisfaction is essential to create loyalty among

customers. In other words, we are impossible to have customer satisfaction without

loyalty.
 The second part of the brand image measurement according to Shimp (2003)

is benefit. Benefit is divided into three parts: functional, symbolic, and experience.

Jones and Taylor (2007) in Rai and Sri Vastava (2012) asserted that basically there

is much literature that follows up and proposes the dimensions of loyalty. They

classify loyalty into three parts: behavioral loyalty, attitude loyalty, and cognitive

loyalty. Functional, ie benefits that seek to provide solutions to the problem of

consumption or potential problems that can be experienced by costumers, assuming

that a brand has specific benefits that can solve the problem. Symbolic, that is

directed to the costumer's desire in an effort to improve themselves, valued as a

member of a group, affiliation, and a sense of ownership. Experience, the Costumer

is a representation of their desire for a product that can provide a sense of fun,

diversity, and cognitive stimulation. Finally, the third part of Shimp's brand image

measurement is the overall evaluation, the subjective value or interest that the

customer adds to the consumption result.

According to Keller (1993), there are 3 factors that can form a brand image as

follow:

1. Strength of the association /strength of brand association

Depending on how information enters the Costumer's memory and how it

persists as part of the brand image.

2. The advantage of a brand association

The success of a marketing process often depends on the process of creating a

profitable brand association, where Costumers can trust their attributes to satisfy

the needs and wants of the Costumer.

3. Uniqueness of brand association

A brand must have a competitive advantage that is the reason for Costumers to

choose a particular brand. The uniqueness of a brand association can be based

on product attributes, product functionality or image that costumers enjoyed.

2.5 Trust

2.5.1 Understanding Trust

Trust is a complex and multi-dimensional phenomenon consisting of a blend

of belief in strong bonds, weak ties, and institutions.Thus, it can be said that trust is

a process (Khodyakov, 2007). Trust is a key variable for successful marketing

relationships (Morgan and Hunt, 1994). According to Peppers and Rogers (2004:

43), trust is the one-party belief in the reliability, durability, and integrity of others

in relationships and beliefs is the best act of interest and will produce positive results

for the trustworthy. This variable has a strong impact on the effectiveness and

efficiency of relationship marketing. Trust is a situation that occurs when a partner

is confident and honesty to his partner.

Trust involves a person's willingness to behave in a certain way because his

or her partner's beliefs will give him what he receives and a hope that someone has,

trustworthy words or other promises (Barnes, 2003: 148). Sheth (2004) stated that

trust is the willingness to rely on the ability, integrity and motivation of others to

act to serve needs and interests as agreed implicitly or explicitly. Trust has three

aspects of service provider characteristics:

1. Ability.

The costumer will assess whether the provider is competent enough to perform

its obligations and serve the costumer.

2. Integrity.

Both costumers will assess whether the company has integrity, where costumers

can believe in the work of the company.

3. Motivation.

Costumers’ trust (is a service provider who has the motivation to not perform

actions that are not in accordance with costumer expectations. Persons who are

trusted shall keep the other party, not their own needs and wishes.

Trust is the basic element of building a quality relationship model. Trust is a

partner's belief in a relationship. One will do the best for what his partner wants.

Trust is the key to quality relationships because it encourages marketers to work

more by emphasizing the work of the same relationship with their partners, refuse

short-term alternatives of interest with long-term strengthening with good

relationships with costumers, see high-risk activities more wisely because they

believe it will not be done opportunistically (Sheth, 2004).

2.5.2 Measuring Trust

The benefit of trust is essential for a successful relationship. The benefit

relationships built on trust are important and simple thing is cooperation. Trust can

ease the feelings of uncertainty and risk. Thus, actions are done to increase the
cooperation between members' relationships. With a low level of trust, members

learn to deliver results that outnumber more;

1. Commitment

Commitment is a component that can build relationships and is easily lost,

which will be formed only with those who trust each other.

2. Relationship duration

Trust drives out the relationships of the members of the work to generate

relationships and to resist the temptation not to give priority to short-term

outcomes and / or actions opportunistically. The seller's trust is qualitative

with the possibility that it will enter the business in the future, thereby

contributing to increase the duration of the relationship.

3. Quality

A party that believes is more likely to receive and use information from a

trusted party, and on the benefits of generating greater ciper than that

information.

Finally, the existence of trust in disputes or conflicts can be solved

efficiently and peacefully. In the absence of confidence, a perceived dispute is a

sign of future difficulties and usually leads to ending relationships.

2.6. Hypothesis Formulation

The previous section has discussed about the research and writing s from

the prior researchers about the associated topics in this research. This section will

discuss about the formulation of hypothesis based on prior findings and theory.
Since there are two main objectives in this research, the hypothesis formulation is

about the realationship of customer loyalty, service quality, customer satisfaction,

brand image and trust.

2.6.1 The Relationship Of Service Quality, Customer Satisfaction, Brand

Image, and Trust on Customer Loyalty

Denio (2006) confirmed that managers can get input from the size of the

loyalty index. To improve customer loyalty, the company can improve service

quality (Kheng et al, 2010; Juan & Yan, 2009; Kursunluoylu, 2011; Seok Lee, 2010;

Sophonsiri, 2010 and Lorenzo et al. 2010; Markovic et al., 2010). Product / service

quality is a strong marketing antecedent that creates customer loyalty. The results

also show that customer satisfaction is positively influenced by service quality

(Lorenzo et al, 2010; Tariq and Moussaoui (2009). Therefore, the hypothesis is as

follow:

H1: Service quality influence positively on customer satisfaction.

Customer satisfaction strongly influence customer loyalty (Donio, 2006;

Min & Wan, 2009, Kheng et al, 2010; Seok Lee, 2010; Kursunluoglu, 2011;

Markovic et al., 2010). Increased customer satisfaction will automatically increased

customer loyalty (Tariq and Moussaoui, 2009). Customer satisfaction is the basis

for any company to operate and make profit. Service quality creates customer

satisfaction, and the customers define quality. The customer satisfaction surveys

have to take place on the continuous basis, and the results have to be put into

practice in the company's strategies and operations (Matson, 2009). Satisfaction is

closely related to customer loyalty (Ganiyu et al, 2012). Therefore, the hypothesis

is as follow:

H2: Customer satisfaction influence positively on customer loyalty.

Brand image is a collection of perceptions about an interrelated brand that

exist in the human mind. Brand image is the antecedent of customer loyalty (Tariq

and Moussaoui, 2009; Alhaddad, 2015; Halim et al, 2014; Upamannyu & Mathur,

2012; Upamannyu & Mathur, 2012). Therefore, the hypothesis is as follow:

H3: Brand image influence influence positively on consumer loyalty.

According to Peppers and Rogers (2004: 43), trust is the one-party belief in

the reliability, durability, and integrity of others in relationships and beliefs is the

best act of interest and will produce positive results for the trustworthy. This

variable has a strong impact on the effectiveness and efficiency of relationship

marketing. Trust is a situation that occurs when a partner is confident and honesty

to his partner. Trust is antecedant customer loyalty (Tariq and Moussaoui, 2009;

Alhaddad, 2015; Halim et al, 2014; Upamannyu & Mathur, 2012). Therefore, the

hypothesis is as follow:

H4: Trust influence positively on consumer loyalty.

2.7 Conceptual Framework

The conceptual framework guide in this research can be seen in Figure 2.1.

Based on the literature review in the previous section, the researcher had generated

four hypotheses associated with the model. These hypothesis focussed on service

quality, customer satisfaction, brand image, and trust to answer relationship

between service quality, customer satisfaction, brand image, and trust with

customer loyalty. The focus was done to answer two questions, namely: (1) service

quality influence customer satisfaction, and subsequently (2) customer satisfaction,

together with brand image and trust influence the loyalty customer.

Figure 2.1. Conceptual Framework

Brand
Image
H2
H1 H3
Service Customer Customer
Quality Satifaction Loyalty

H4

Trust
 CHAPTER III

RESEARCH METHOD

This chapter discusses the type of research, research subject includes the

research population and research sample, research settings, research instrument,

research variables and operational and the techniques of data analysis.

3.1. Type of Study

This research can be classified as a causal study. It established a definitive

cause and analysed its influence relationship (Sekaran, 2000). This research

analyzed the influence of service quality on customer satisfaction continued to

customer loyalty. This research also analyzed the influence of service quality,

customer satisfaction, brand image, and trust on costumer loyalty toward

transportation toward service of Gojek in Yogyakarta city.

3.2. Research Subject

3.2.1 Population
This research was done to know the influence of service quality, customer

satisfaction, brand image, and trust on costumer loyalty toward transportation

service of Gojek in Yogyakarta city. Thus, the population used in this research was

Gojek customers or all the people of Yogyakarta city who had ever used Gojek

service.
3.2.2 Sampling Design
Questionnaires were distributed randomly using a sampling design of non

probability sampling and purposive sampling methode. Non probability sampling

describes the elements in the population do not have the opportunity that has been

selected to be chosen as the subject (Sekaran, 2000). The category of non-

probability sampling used in this research was purposive sampling to obtain

information from specific target groups. The sampling was confined to specific

types of people who can provide the desired information, and who conform to some

criteria set by the researcher.

Sample criteria used in this research were people living in Yogyakarta City

aged 17-55 years (the age range for the productive period of people, which also

indicated the ability to purchase, assess the transport service), male and women sex

and had ever used the services of Gojek. Survey was conducted for consumer

opinion about service quality, customer satisfaction, brand image, trust, and

costomer loyalty. Survey was conducted from July-August 2017. The number of

samples were 100 respondents.

3.3 Research Setting

Since the subject and sample were Gojek users, the setting of this research was

conducted via offline. The reason of choosing Gojek was because Gojek is the first

online transportation service in Indonesia which exist and recognized among the

Indonesian society, especially the people of Yogyakarta.

 3.4 Research Instrument
3.4.1 Data Collection Method
Primary data used in this research was collected directly from the research

objects by using measurement as the source of information. The research instrument

used to collect the data in this research was offline questionaire which takes the

objects directly to the field.

The questionnaire used in this research was a structured questionnaire because

the alternative answer was provided by the researcher. The question model in this

research was a closed question because the researcher provided an alternative

answer to choose one as the most appropriate answer. The questionnaire consisted

of three parts. The first section was the verification section that was used to identify

whether the selected subject was a user of Gojek or not. The second part of the

questionnaire was the demographic data of respondents (consisting of gender, age,

department and amount of allowance per month) while the third part was the data

that consisted of service quality, customer satisfaction, brand image, trust and

costumers loyalty.

The main part of questionaire used Seven-point Likert scale to measure

respondent evaluation toward each question item represented each construct. Label

and description of the Seven-point Likert scale were: 1 represented “Strongly

Disagree”, 2 represented “Disagree”, 3 represented “Somewhat Disagree”, 1

represent “Either Disagree”, 5 represented “Somewhat Agree”, 6 represented

“Strongly Disagree”, and 7 represented “Very Disagree”. The original method

implemented by Eid (2011) used 5 point Likert scale.

3.4.2 Validity

Validity test is the ability of scale to measure the intended concept (Sekaran,

2000). The function of validity test is to measure and analyze whether each item of

instrument could explain the variable observed or not (Effendi & Singarimbun,

1989). A measurement tool is claimed valid if it is able to measure what it wants to

measure. The high and low level of validity of measurement tool respectively show

how far the collected data do not stray from the path of description in the direction

of the tested variables (Effendi & Singarimbun, 1989).

To examine the validity of the questonaire, 100 questionaire were distributed to

Gojek Customer. The data was evaluated by using SPSS software to instantly

generate the corrected items correlation of each construction. If the total correlation

of corrected item is greather than critical value of validity coefficient (0.361), it

means that the item can be considered as valid, but if the validity coefficient of one

item is less than the critical value for validity coefficient (0.361), it means that the

item is invalid or failed.

3.4.3 Reliability

Reliability test is the index by which a measurement tool can be trusted

(Sekaran, 2000). If a measurement tool can give relatively consistent and stable

result for two or more users, it is indicated that the measurement tool is reliable

(Effendi & Singarimbun, 1989). Reliability test is designed to reveal to consistency

of measurement tools and it could give a relatively consistent result if there is a re-

measurement in the same subject. A reliable and appropiate measurement tool tends
to direct respondents to answer the question in the way expected. A reliable

measurement tool will provide a reliable result that is also relevant to the variable

used and the reality of condition examined, and result of any measurement

conducted in subsequent periods that will always be the same (Sekaran, 2000).

To examine the validity of the questionaire, 100 questionaires were distruted

directly to Gojek customers. The data was evaluated by using SPSS software to

instantly generate the alpha coefficient from Cronbach of each contruct. If the value

of alpha Cronbach (ɑ) is ≥ 0.6, the measurement tool of the research is claimed

reliable to be used.

3.5 Research Variables and Operational

3.5.1 Independent Variable

1. Service Quality

Halim et al. (2014) defined quality service as the assessment of customer

to company consistency in fulfilling consumer expectations to requirements

of product. The indicators used measure trust in this research that were

adapted form Yuen and Thei (2015). These indicators were:

a. Consistency of customer

b. Error-free documentation

c. On time

d. Frequency of transportation services

e. Accuracy of invoice

f. Effectiveness of driver
 g. Promptness of customer service

h. CSR and concern for human savety

i. Variety of service offering

j. Pricing of shipping service

k. Savety and security riding

2. Customer Satisfaction

Upamanyu & Sankal (2014) argued that customer satisfaction is a

comparison of expectations versus perceptions of experience. Customer

satisfaction (CS) is a measure of the degree to which a product or

service meets the customer's expectations. Customer satisfaction is a

measurement or an indicator of the degree to which customers or users

of an organizations’ product or services are pleased with those products

or services. Customer satisfaction differs depending on the situation

and the product or service. A customer may be satisfied with a product

or service, an experience, a purchase decision, a salesperson, store,

service provider, or an attribute or any of these three. The indicators

that used measure trust in this research were adapted form Upamannyu

& Sankpal (2014). These indicators were:

a. make me delighted

b. give me pleasure; thus, increases my frequency of use

c. make me feel good

 d. prevent me for looking another cheap brand

3. Brand Image

Upamanyu & Sankal (2014) argued that brand image is the perception

of the customer which is perceived by customers while buying

commodity and service but brand image cannot be treated as a

benchmark or guarantee for giving satisfaction to the customer.

Consumers develop various associations with the brand. Based on these

associations, they form brand image. An image is formed from the

brand on the basis of subjective perceptions of association’s bundle that

the consumers had about the brand. There were several indicators used

to measure brand image. Upamanyu & Mathur (2012) defined brand

image as the current view of the customers about a brand. It can be

defined as a unique bundle of associations within the minds of target

customers. It signifies what the brand presently stands for. It is a set of

beliefs held about a specific brand. In short, it is nothing but the

consumers perception about the product.

Brand image includes the strength of brand association, brand

associations advantage, and uniqueness of brand associations.

The indicators used measure trust in this research that were adapted

form Upamannyu & Sankpal (2014). These indicators were:

a. Provides a solution to my expectations, good impression,

dependeable for use, and make me comfort.

b. Feel delighted, fell good, and frequency of use.

 c. Looking cheap and desirable lifestyle gives me pleasure.

d. Perceived by others, the performance is promising and feeling

accepted by my social group.

4. Trust

Halim et al (2014) defined trust as a positive expectation of consumers

to producer for his ability to produce product that satisfy customer. It

includes : integrity, competence, consistency, and transparency. The

indicators used measured trust in this research which were adapted from

Hidayat et. Al. (2010). These indicators were:

a. Customer feel safe with the service

b. Customer believe with the company’s promise

c. Customer believe that the company’s quality service and consistent

d. Employees respect customer

e. Meet their obligations to customer

f. Believe with company’s service

3.5.2 Dependent Variable

1. Customer Loyalty

Halim et al (2014) defined loyalty customers as a customers positivity

for pleasure of product with specific brand so it can push using it, where
 there are a loyal customer characteristic.The indicators used measure of

customer loyalty in this research which were adapted form Hidayat

(2010). These indicators were:

a. The customer always decide that the company is the first choice.

b. From the experience that customer had, the company was the first

in the mind of customer.

c. Always recommend this service to friends, neighbours, and

relatives.

3.6 Techniques of Data Analysis

This research used regression analysis. Regression analysis is one of the most

commonly used statistical methods in practice. The purposes of regression analysis

are three-folds (Yan & Su, 2009):

1. Establish a casual relationship between response variable y and regressors

x1,x2,··· ,xn.

2. Predict y based on a set of values of x1,x2,··· ,xn.

3. Screen variables x1,x2,··· ,xn to identify which variables are more important

than others to explain the response variable y. Thus, the causal relationship can

be determined more efficiently and accurately.

There are two model of analysis used in this research. They are simple linear

regression and double linear regression. Before applying regression analysis, the

data used must be suitable with assumption test as follow in regression:

1) Single Regression model

 The term “regression” and the methods for investigating the

relationships between two variables may take a look to about 100 years ago.

It was firstly introduced by Francis Galton (1908), the renowned British

biologist, when he was engaged in the study of heredity (Yan & Su, 2009).

The simple linear regression is for modeling the linear relationship between

two variables. One of them is the dependent variable y and another is the

independent variable x. The simple linear regression model is typically

stated as follows:

y = β0 +β1x+ε,

where y is the dependent variable, β0 is the y intercept, β1 is the slope of

the simple linear regression line, x is the independent variable, and ε is the

random error. The dependent variable is also called response variable, and

the independent variable is called explanatory or predictor variable. An

explanatory variable explains causal changes in the response variables (Yan

& Su, 2009).

The regression service quality to satifaction customer (H1). The

general form of the single regression model is as follows:

Y = a + bX + e

where y is satisfaction customer, x is the service quality, a is constant, b is

the coefficient of the variable and e is error.

2) Multiple Linear Regression Model

 The general purpose of multiple linear regression is to seek for the

linear relationship between a dependent variable and several independent

variables. Multiple regression allows researchers to examine the effect of

more than one independent variables on response at the same time. For some

research questions, regression can be used to examine how much a

particular set of independent variables can explain sufficiently the outcome.

In other cases, multiple regression is used to examine the effect of outcome

while accounting for more than one factor that could influence the outcome

(Yan & Su, 2009).

Therefore, it can be concluded that the multiple linear regression

assumes that the response variable is a linear function of the model

parameters and there are more than one independent variables in the model.

A general form of a multiple linear regression model is as follow:

y = β0 +β1x1 +β2x2 +···+βkxk + ε

where ε is the random error. Regressors x1,x2,··· ,xk may contain regressors

and their higher order terms. In the classical setting, it is assumed that the

error term ε has the normal distribution with a mean of 0 and a constant

variance of σ2.

In this research, the regression was used to analyze brand image,

customer loyalty and trust on customer loyalty (H2). The general form of

the multiple linear regression model is as follows:

Y = a + b1X1 + b2X2 + b3X3 + e,

 whereas y is customer loyalty, x1 is the brand image, x2 is customer

satisfaction, x3 is trust, a is contant, b is the coefficient of the variable, and

e is error.

3.7 Assumption in Multiple Regression

The fact that we can make bold statements on causality from a regression

hinges on the classical linear model. If its assumptions are violated, then we

must re-specify our analysis and begin the new regression. The impact if this

assumption is as follows:

1. The worst impact is a bias in the F (then the model cant be trusted)

2. A second disastrous impact is a bias in the betas (the coefficient

estimates are unreliable)

3. Compared to the above, biases in the standard errors and T are not so

harmful. These biases only affect the reliability of our confidence about

the variability of an estimate, not the reliability about the value of the

estimate itself. (Gupta, 2000)

The assumptions of MR that are identified as primary concern in the

research include linearity, independence of errors, homoscedasticity,

normality (Yan & Su, 2009) and Ballance (2013) add collinerity in the

regression assumtion. Checking the assumptions carry significant benefits for

the researcher, reduce error, and increase reliability and validity of inferences.

Consideration of the issues surrounding the assumptions, multiple regression

should improve the insights for researchers as they build theories (Jaccard et

al in Ballance, 2013). In line with Yan & Su (2009) and Ballance (2013), this
research used linearity test, independence of errors, collinearity,

homoscedasticity and normality as the regression assumption test.

3.7.1 Linearity

Some researchers argued that linearity is the most important

assumption, as it directly relates to the bias of the results of the whole

analysis. Linearity defines the dependent variable as a linear function of

the predictor (independent) variables. Multiple regression can accurately

estimate the relationship between dependent and independent variables

when the relationship is linear in nature. The chance of non-linear

relationships is high in the social sciences, it is essential to examine

analyses for linearity. If linearity is violated, all the estimates of the

regression including regression coefficients, standard errors, and tests of

statistical significance may be biased. If the relationship between the

dependent and independent variables is not linear, the results of the

regression analysis will under- or over- estimate the true relationship and

increase the risk of Type I and Type II errors. When bias occurs it is likely

that it does not reproduce the true population values. Violation of this

assumption threatens the meaning of the parameters estimated in the

analysis (Ballance, 2013).

The linearity assumption is referred to as the assumed linear

relationship between the mean response E (yiIxi) and the predictors xi. This

assumption leads to E(εi) = 0 (Yan & Su, 2009).

 One method of preventing non-linearity is to use theory of previous

research to inform the current analysis to assist in choosing the appropriate

variables. One of the methods used to evaluate the existance of the

linearity problem is by seeing the residual plot showing the standardized

residuals vs. the predicted values and are very useful in detecting violations

in linearity. The residuals magnify the departures from linearity. If there is

no departure from linearity, a random scatter about the horizontal line will

be seen. Any systematic pattern or clustering of the residuals suggests

violation (Ballance, 2013).

3.7.2 Normality

Multiple regression assumed that variables have normal

distributions. This means that errors are normally distributed and a plot of

the values of the residuals will approximate a normal curve. The assumption

is based on the shape of normal distribution and gives the researcher

knowledge about what values to expect. Once the sampling distribution of

the mean is known, it is possible to make predictions for a new sample.

When scores on variables are skewed, correlations with other measure will

be attenuated, and when the range of scores in the sample is restricted

relative to the population correlations with scores, other variables will be

attenuated. Non-normally distributed variables can distort relationships and

significant tests. Outliers can influence both Type I and Type II errors and

the overall accuracy of results. The researcher can test this assumption
 through several pieces of information: visual inspection of data plots, skew,

curtosis, and P-Plots (Ballance, 2013).

Data cleaning can also be important in checking this assumption

through the identification of outliers. Statistical software has tools

designed for testing this assumption. Skewness and curtosis can be checked

in the statistic tables, and values that are close to zero that indicated normal

distribution. Normality can further be checked through histograms of the

standardized residuals. Histograms are bar graphs of the residuals with a

superimposed normal curve that show distribution. The normal curve is

fitted to the data using the observed mean and standard deviation as

estimated, and computing the corresponding chi square (Ballance, 2013).

3.7.3 Homoscedasticity

The assumption of homoscedasticity refers to equal variance of

errors across all levels of the independent variables. This means that

researchers assume that errors are spread out consistently between the

variables. This is an evidence when the variance around the regression line

is the same for all values of the predictor variable. When heteroscedasticity

is marked, it can lead to distortion of the findings and weaken the overall

analysis and statistical power of the analysis, which result in an increased

possibility of Type I error, erratic and untrustworthy F-test results, and

erroneous conclusions (Ballance, 2013). The homoscedasticity assumption

 requires that all random errors have the same constant variance (Yan & Su,

2009).

In related to the above, incorrect estimates of the variance lead to

the statistical and inferential problems that may hinder theory

development. However, it is good to note that the regression is fairly robust

to violation of this assumption. Homoscedasticity can be checked by visual

examination of a plot of the standardized residuals by the regression

standardized predicted value. Specifically, statistical software scatterplots

of residuals with independent variables are the method for examining this

assumption. Ideally, residuals are randomly scattered around zero (the

horizontal line) providing even distribution. Heteroscedasticity is indicated

when the scatter is not even; fan and butterfly shapes are common patterns

of violations. When the deviation is substantial, more formal tests for

heteroscedasticity should be performed, such as collapsing the predictive

variables into equal categories and comparing the variance of the residuals.

The rule of thumb for this method is that the ratio of high to low variance

which is less than ten is not problematic. Bartlett’s and Hartley’s tests have

been identified in the research as flexible and powerful tests to assess

homoscedasicity (Ballance, 2013).

3.7.4 Collinearity

Collinearity (also called multicollinearity) refers to the assumption

that the independent variables are uncorrelated. The researcher is able to

interpret regression coefficients as the effects of the independent variables

on the dependent variables when collinearity is low. This means that we

can make inferences about the causes and effects of variables reliably.

Multicollinearity occurs when several independent variables correlate at

high levels with one another, or when one independent variable is a near

linear combination of other independent variables (Ballance, 2103).

The more variables overlapped (correlate) the lesser the ability of

researchers to separate the effects of variables. In MR the independent

variables are allowed to be correlated to some degree. The regression is

designed to allow this, and provide the proportions of the overlapping

variance. Ideally, independent variables are more highly correlated with

the dependent variables compared to other independent variables. If this

assumption is not satisfied, autocorrelation is present (Ballance, 2103).

Multicollinearity can result in misleading and unusual results,

inflated standard errors, reduced power of the regression coefficients that

create a need for larger sample sizes. Interpretations and conclusions are

based on the size of the regression coefficients, their standard errors, or

associated t-tests that may be misleading because of the confounding

effects of collinearity. The result is that the researcher can underestimate

the relevance of a predictor, the hypothesis testing of interaction effects is

hampered, and the power for detecting the moderation relationship is

reduced because of the intercorrelation of the predictor variables (Ballance,

2103).
 One way to prevent multicollinearity is to combine overlapping

variables in the analysis and avoid including multiple measures of the same

construct in a regression. Statistical software packages include collinearity

diagnostics that measure the degree to which each variable is independent

of other independent variables. The effect of a given level of collinearity

can be evaluated in conjunction with the other factors of sample size, R2,

and magnitude of the coefficients. Widely used procedures examine the

correlation matrix of the predictor variables, computing the coefficients of

determination, R2, and measures the eigen values of the data matrix

including variance inflation factors (VIF). Tolerance measures the

influence of one independent variable on all other independent variables.

Tolerance levels for correlations range from zero (no independence) to one

(completely independent) (Ballance, 2103).

The VIF is an index of the amount that the variance of each

regression coefficient is increased over that with uncorrelated independent

variables. When a predictor variable has a strong linear association with

other predictor variables, the associated VIF is large and it is the evidence

of multicollinearity. The rule of thumb for a large VIF value is ten. Small

values for tolerance and large VIF values show the presence of

multicollinearity (Ballance, 2103).

3.8 Hyphotesis Test

3.8.1 The Goodness of Fit of the Regression Model

 A measurement of statistical model that explains the observed data

is the coefficient of determination, the square of the Pearson correlation

coefficient (r) between y and y accent. This measurement describes the

percentage of the total variance that can be explained from the predictor

criterion covariance. In its general form, the correlation coefficient is given

by (Eye & Schuster, 1998):

If we replace x with y accent in the above formula we obtain the multiple

correlation R. The reasons for making a distinction between r and R are as

follows:

- r is a measurement between a random variable y and its prediction from

a regression model.

- r lies in the interval -1 < r <_ 1 while the multiple correlation R cannot

be negative; that is, it lies in the interval 0 < R ≤ 1.

- R is always well defined, regardless of whether the independent vari-

able is assumed to be random or fixed. In contrast, calculating the

correlation between a random variable (Y) and a fixed predictor vari-

able (X) that is, a variable that is not considered random, makes no

sense (Eye & Schuster, 1998).

The square of the multiple correlation (R2) is the coefficient of deter-

mination. If the predictor is a random variable the slope parameter can be

expressed as follow (Eye & Schuster, 1998):

 where 𝑠𝑥2 and 𝑠𝑦2 are the variances of the predictor and the criterion,

where s x sy respectively. If the predictor (X) is a random variable it makes

sense to calculate the correlation between X and Y. Then, R2 is identical

to r2 as Y is just a linear transformation of X, and correlations are invariant

against linear transformations (Eye & Schuster, 1998).

Alternatively, R2 can be expressed as the ratio of variance explained

by regression (SSR) and the total variance (SSTO) or

where SSE is the variance of the residuals. The right-hand term in suggests

that R2 is a measure of proportionate reduction in error. In other words, R2

is a measure of the proportionate reduction in the variability of Y that can

be accomplished by using predictor (X) (Eye & Schuster, 1998).

3.8.2 T test

T test describes the reliability of individual coefficients. If the value

in “Sig.” is less than 0.05, it can be assumed that the estimation can be

asserted as true with a 95% level of confidence (Yan & Su, 2009).

3.8.3 F test

F test is done to test the simultaneous significant effect between the

independent and dependent variable. This method tests the hipothesis to

determine the level of mistake (α) and search for statistical value (p-value)

and comparing both score with the determined significant level (Yan & Su,

2009).
 CHAPTER IV

RESEARCH FINDING AND DISCUSSION

This chapter discusses the result of the data analysis including the

demographic characteristics of the respondents, the validity and reliability test and

the analysis result to answer research question.

4.1 Descriptive Statistic

4.2.1 Respondents’ Demographic Characteristics

4.2.1.1.Respondents’ Gender

Based on the gender, there were two categories, male and female.

The data of the analysis result based on the characteristics of the gender is

shown in the following Table 4.1 below:

Table 4.1 The Frequencies Distribution of the Respondents’ Gender

Gender Frequency Percentage

Male 64 64%

Female 36 36%

Total 100 100%

Source: Primary data processed, 2017

 Table 4.1 shows that from the total of 100 respondents in this

research, the male respondents were 64 people about 64 percent, whereas

the female respondents were 36 people about 36 percent. From the data

above and as shown in the figure 4.1, it can be concluded that male

respondents were dominant in this research.

Figure 4.1 Diagram of Respondents’ Gender

respondents' gender

36,00%

64,00%

male female

Source: Primary data processed, 2017

4.3.1.2 Respondents’ Age

Based on the age, there are four groups : below 18 years old, 19 -

24 years old, 25 - 30 years old, and above 30 years old. The data of the

respondents’ age is shown in Table 4.2 below.

Table 4.2 Frequency Distribution of Respondents’ Age

Age frequency Percentage

≤ 18 years 20 20 %
 19- 24 years 23 23 %

25 - 30 years 25 25 %

> 30 years 32 32 %

total 100 100 %

Source: Primary data processed, 2017

Table 4.2 shows that the respondents aged >30 years old dominated

this research with 32 respondents of the total 100 respondents or about 32

percents from the total respondents. Respondents aged below and 18 years

old are the minor participants. The comparison of this proportion can be

seen in Figure 4.2 below.

Figure 4.2. Diagram of Respondents’ Age

Respondents' Age
35

30

25

20

15

10

0
≤ 18 years 19- 24 years 25 - 30 years > 30 years

Age

Source: Primary data processed, 2017

4.3.1.3 Respondents Provenace

 The respondents provenances were grouped based on the provinces

in which they were from. They stay in Yogyakarta but they are from various

provenances. As Indonesia has 34 provinces, it was also grouped into 34

classes. The data of the respondents’ provenance is shown in Table 4.3

below.

Table 4.3 Respondents’ Provinces

Provinces Frequency Percentage

West Java 10 10%

DKI Jakarta 7 7%

D.I.Yogyakarta 16 16%

East Java 10 10%

Central Java 14 14 %

Bali 12 12%

North Sumatra 6 6%

South Kalimantan 7 7%

South Sumatra 6 6%

East Kalimantan 6 6%

Riau 4 4%

Aceh 2 2%

Total 100 100%

Source: Primary data processed, 2017

From Table 4.3 above, it can be seen that the respondents came

from 21 different provinces. They stayed in D.I Yogyakarta. The

 respondents were mostly from D.I Yogyakarta consisting of 16 percent from

the total respondents. Central Java was 14 percent and Bali was 12 percent

and also West and East Java were each 10 percent. DKI Jakarta consisted of

7 percent then North and South Sumatra also East Kalimantan were each 6

percent. The least respondents come from Aceh with 2 percent of the

respondents. The comparison of that proportion can be seen in Figure 4.3

below:

Figure 4.3 Diagram of Respondents Province

Respondents' Province
Riau

South Sumatra

North Sumatra

Central Java

D.I.Yogyakarta

West Java
0 2 4 6 8 10 12 14 16 18

frequency

Source: Primary data processed, 2017

4.2.1.4 Respondents’ Occupation

The respondents’ occupations were divided into four groups:

Student, Public Servant /Private, Entrepreneur and Others. The data of the

respondents’ occupation is shown in Table 4.4 below.

Table 4.4 Frequency Distribution of Respondents’ Occupation

 Occupation Frequency Percentage

Student 25 25 %

Public servant/ Private 38 38 %

enterpreuner 31 31 %

Others 6 6%

total 100 100%

Source: Primary data processed, 2017

Table 4.4 shows that most of the respondents’ occupation were public

servant/ private from 38 percent of the total respondents. Then, it was followed by

enterpreneur and students with 31 precent and 25 percent of the total respondents.

The least respondents’ occupation is other which was 6 percent of total

respondents.

Figure 4.4. Diagram of Respondents’ Occupations

Respondents' Occupations
40

35

30

25

20

15

10

0
Student Public servant/ Private enterpreuner Others

Frequency

Source: Primary data processed, 2017

4.2.1.5 Respondents’ Intensity in Using The Gojek Service

The respondents’ intensity in using the Gojek service were divided in

the conditions of rarely, often, very often and other.

Table 4.5. Respondents’ Intensity in Using The Gojek Service

Intensity Frequency Percentage

rarely 25 25 %

often 36 36 %

Very often 30 30 %

Other 9 9%

total 100 100%

Source: Primary data processed, 2017

From Table 4.5, it can be seen that the highest intensity was “often” of 36

or 36 percent then it was followed by “very often” of 30 or 30 percent and “rarely”

of 25 or 25 percent.

Figure 4.5. Diagram of Respondents’ Intensity in Using The Gojek Service

 Respondents Intensity in
using The Gojek Service
200
100
0
rarely often Very often Other total

frequency

Source: Primary data processed, 2017

4.3.1.6 Respondents’ Data of Using the Other Gojek Services (Go Food, Go

Glam, etc)

The respondents’ data of using the Gojek services can be seen from Table

4.6 below.

Table 4.6. Respondents’ Data of Using the other Gojek Services

Data Frequency Percentage

Yes 59 59 %

No 41 41 %

Total 100 100%

Source: Primary data processed, 2017

The table shows that the respondents who use the other Gojek service were

59 or 59 percent and the respondents who only use the Gojek Service are 41 or

41 percent.

Figure 4.6. Diagram of Respondents’ Intensity in Using The Other Gojek

Service
 Respondents Intensity
in Using The Other
Gojek Service

yes
41%
59% no

Source: Primary data processed, 2017

4.3.1.7 Respondents’ Income/Salary Per month

The respondents’ income/salary per month were in 4 categories.

They were 500,000-1,500,000 rupiahs, the second was 1,500,000-2,500,000

rupiahs, the third was 2,500,000-3,500,000 rupiahs, and other. Below is the

table of respondents income/salary.

Table 4.7. Respondents’ Income/salary

Income/salary Frequency Percentage

Rp.500,000-1,500,000 10 10%

Rp.1,500,000-2,500,000 27 27%

Rp.2,500,000-3,500,000 33 33%

other 30 30%

total 100 100%

Source: Primary data processed, 2017

 The frequency of respondents’ income/salary in the category of 500,000-

1,500,000 rupiahs was 10 or 10 percent. The second category of 1,500,000-

2,500,000 rupiahs was 27 or 27 percent. The third category of 2,500,000-3,500,000

rupiahs was 33 or 33 percent, and other was 30 or 30 percent.

Figure 4.7. Bar Chart Diagram of Respondents’ Income/Salary

120
100 Respondents' Income/
80 Salary
60
40
20
0

frequency

Source: Primary data processed, 2017

4.2 Result of Hypotesis Testing

4.2.1 Linear Regression Analysis

1. Normality Data

Below is the result of data normality test for the first hypothesis.
 Figur 4.8. P-P Plot

Source: Primary data processed, 2017

Figure 4.8 shows the residual points above and below

diagonal line. Therefore, it can be concluded that the data in this

research was suitable with the data of normality test.

2. Simple Linear Regression Equation

a. Simple Linier Regression Equation Model

This simple regression linear was applied to prove the

influence of service quality on customer satisfaction. The

following are the results of simple regression analysis in this

research:

Table 4.8. Simple Linear Regression Analysis

 Unstandardized Coefficients
Variable
B Std. Error
(Constant) -1,191 0,203
Service Quality 1,214 0,035

Sources : Primary data processed, 2017

From the output above, the following equation can be made:

Customer Satisfaction = - 0.191 + 1.214*service quality

The above equation can be explained as follow:

a) The constanta of -0.191 showed negative relationship and it

means that if the values of service quality reputed constant

(valued 0), customer satisfaction will decrease 0.191.

b) Regression Coefficient of variable service quality was 1.214

that showed positive relationship on customer satisfaction.

The increase of 1.0 on service quality, will increase 0.214 on

customer satisfaction.

b. The Goodness of Fit of Regression Model

1. Determination Coefficient

Goodness of fit from regression model can be seen

through determination coefficient. Determination

Coefficient is one of the measurements that measures how

good is the statistic model can explain the observed data

which is described in percentage of total varians and it is

explained from the covarian predictor criteria. For the next

adjusted R-square measures varians proportion in dependent

variable which is explained by the variation of indenpendent

variable while R-square measures the proportion of

dependent variation that is explained by the variation of

independent variabel (Eye & Chuster, 2009).

The dependent variable dissemination can be

predicted around its mean measured by Std Error of

Estimate. If Std Error is more than 10%, it means high. The

following is the analysis result of determination result in this

research:

Table 4.9. Coefficient Determination

Adjusted Std. Error of
Model R R Square
R Square the Estimate
1 0,963 0,927 0,927 0,25034
Source: Primary data processed, 2017

a. R value was 0.963 means that the correlation between

customer quality and customer satisfaction was 0.963.

It shows tight connection.

b. Determination coefficient (R2) gained was 0.927

means that the up and down variable of the customer

satisfaction (Y) variable was influenced by service

quality of 92.7%, while the rest was 7.3% which was

influenced by other factors or other variables which

was not observed.

 c. Adjusted R Square was 0.927 which shows that the

influence of customer quality variable on customer

satisfaction had the dependent variable of 92.7% and

the rest was influenced by other factors.

d. Standard Error Estimate (SSE) was 0.25034. Thus, the

standard error in this research was very small.

2. F test dan t test

F test was done to measure the significance of simultaneous

influence between the independent and dependent variable.

Thus, the hypothesis would be accepted if p-value < α, with the

significance of 0.05 (p-value < 0.05). T test described the

reliability of individual coefficients. If the value in “Sig.” is less

than 0.05, it can be assumed that the estimation can be asserted

as true with 95% level of confidence. Below were the results of

F-test and t test in this research.

Table 4.10. F Test and t Test

Test tools’ type F or t value Sig.

F test 1198.616 0.000
T test 34.621 0.000
Source: Primary data processed, 2017

From above table, it can be seen that p-value was 0.000 <

0.05. It can be concluded that service quality variable influenced

customer satisfaction.
4.2.2 Multiple Regression Analysis

1. Regression Assumtion (Classical Assumtion Test)

a. Linearity

Focusing on Ballance (2013) who stated that the accuracy of

double regression can be predicted from the linear relationship

(linear in nature) between dependent dan independent variable.

Linearity test was applied in this research so that the research

reproduced the true population values and also the regression

estimation (regression coefficient, standar errors dan tests of

statistical significance) resulted were not biased. The followings are

the results of liniearity test in this research:

Figure 4.9. P-P Plot 2

Source : Primary data processed, 2017

Residual plot in figure 4.9 shows residual points existed

between diagonal line. Thus, the model in this research was suitable

with linearity test.

b. Normality
 Multiple regression assumed that variables had normal

distributions. This means that errors were normally distributed, and

that a plot of the values of the residuals will approximately have a

normal curve. The assumption was based on the shape of normal

distribution and gave the researcher knowledge about what values to

expect. Once the sampling distribution of the mean was known, it is

possible to make predictions for a new sample. When scores on

variables were skewed, correlations with other measures would be

attenuated, and when the range of scores in the sample is restricted

relative to the population correlations with scores on other variables

would be attenuated (Ballance, 2013). Besides being used as linearity

test, PP Plot alco could be used as the tool of normality test for the

data together with that (Coxt & Wermut, 1994). It was stated that P-P

plot of the residual suggest was that the residual probably normally

distributed (Gupta, 2000). The researcher tested this assumption

through normal P-Plots. The figure above shows that the data in this

research’ was regression model. This research was normally

distributted because the points on normal PP Plot lied between the

diagonal lines.

c. Homoscedasicity

The assumption of homoscedasticity refers to equal variance of

errors across all levels of the independent variables. This means that
researchers assume that errors are spread out consistently between the

variables. This is an evidence when the variance around the regression

line is the same for all values of the predictor variable. When

heteroscedasticity is marked, it can lead to distortion of the findings

and weaken the overall analysis and statistical power of the analysis,

which result in an increased possibility of Type I error, erratic and

untrustworthy F-test results, and erroneous conclusions.

Homoscedasticity can be checked by visual examination of a plot of

the standardized residuals by the regression standardized predicted

value. Ideally, residuals are randomly scattered around zero (the

horizontal line) providing even distribution. Heteroscedasticity is

indicated when the scatter is not even; fan and butterfly shapes are

common patterns of violations (Ballance, 2013).

The diagram below shows the results of homoscedasticity test in

this research.

Figure 4.10. Homoscedasticity Test

 Source: Primary data processed, 2017

Based on the diagram above, it can be concluded that the research

model was suitable with the assumption of homoscedasticity test

because the points were spread around 0 (zero).

d. Colinearity

Collinearity (also called multicollinearity) refers to the

assumption that the independent variables are uncorrelated.

Multicollinearity occurs when several independent variables

correlate at high levels with one another, or when one independent

variable is a near linear combination of other independent variables.

Multicollinearity can result in misleading and unusual results,

inflated standard errors, reduced power of the regression coefficients

that create a need for larger sample sizes. Interpretations and

conclusions based on the size of the regression coefficients, their

standard errors, or associated t-tests may be misleading because of

the confounding effects of collinearity. The result is that the

researcher can underestimate the relevance of a predictor, the

hypothesis testing of interaction effects is hampered, and the power

for detecting the moderation relationship is reduced because of the

intercorrelation of the predictor variables (Ballance, 2103).

Widely used procedures examine the correlation matrix of the

predictor variables, computing the coefficients of determination, R2,

and measures of the eigenvalues of the data matrix including

variance inflation factors (VIF). Tolerance measures the influence of

one independent variable on all other independent variables.

Tolerance levels for correlations range from zero (no independence)

to one (completely independent). The VIF is an index of the amount

that the variance of each regression coefficient increased over

uncorrelated independent variables. When a predictor variable has a

strong linear association with other predictor variables, the

associated VIF is large and become the evidence of multicollinearity.

The rule of thumb for a large VIF value is ten. Small values for

tolerance and large VIF values show the presence of

multicollinearity (Ballance, 2103).

The following is the result of multicolinierity test in this

research:

Table 4.11. The Result of Multicolinierity Test

Variable Tolerance VIF notes
Customer Service 0,247 4,047 No multicolinierity
Brand Image 0,816 1,225 No multicoliniearity
Trust 0,241 4,146 No multicolinerity
Source : Primary data processed, 2017

Table 4.11 shows that in this research, multicolinierity was

not detected. This is shown from the value of tolerance which was

less than 1 and the value of VIF which was under 10. Since

multicolinierity was not detected in this research, the research was

free from autocorrelation which was in line with Ballance (2013)

 who said that a regression model which can be detected by

multicolinierity will be free from autocorrelation.

The Durbin-Watson statistic is used to test for

autocorrelation and lack of independence of residuals, which are

commonly seen in time series data (Yan & Su, 2009). The Durbin-

Watson (DW) test is the most widely used test for autocorrelation in

regression model. The limitation of this test is critical value that

depends on X-matrix, which leads to “indeterminate range”, only

applicable for first order autocorrelation and not suitable for

dynamic model. Good regresion Model required no problem with

the existed autocorrelation problem. The effects of autocorrelation

is that sample variant cannot describe the population’s variant.

1<DW<3 was used to detect the existence of autocorrelation

(Sufren dan Natanael, 2013).

Below is the result of Durbin Watson test in this research:

Table 4.12. Results of Durbin Watson Test

Model Durbin Watson
Model Regresi 1 1.531
Source: Primary data processed, 2017

The value of Durbin Watson in this regression model was

1.721 where the value was more than 1 and less than 3. Therefore, it

can be concluded that the regression model used in the research fullfiled

the requirements of 1<DW<3 as suggested by Sufren & Natael (2013).

 Thus, the regression model in this research did not contain

autocorrelation.

2. Multiple Regression Linear Analysis

This Double regression linear was applied to prove : (1) the

influence of customer satisfaction on customer loyalty (2) the influence

of brand image on customer loyalty and (3) the influence of trust on

customer loyalty. The following are the results of multiple regression

analysis in this research.

Table 4.13. Multiple Regression Analysis

Unstandardized Coefficients
Variable
B Std. Error
(Constant) -0.160 0.214
Customer Satisfaction 0.174 0.046
Brand Image 0.072 0.037
Trust 0.775 0.055
Source: Primary data processed, 2017

From the output above, the regression equation was as follow:

Customer Loyalty = - 0.160 + 0.174 x Customer Satisfaction + 0.072

x Brand Image + 0.775 x Trust.

The above equation can be explained as follow:

a) The constanta -0.160 showed negative relationship and it means

that if the values of service quality, customer service, brand image

and trust reputed constant (valued 0), customer loyalty would

decrease 0.160.
 b) Regression coefficient of variable customer satisfaction (X1) was

0.174 that showed positive relationship on customer loyalty. The

increase of 1.0 on customer service will increase 0.174 on customer

loyalty.

c) Regression coefficient of brand image variable (X2) was 0.072 that

showed positive relationship on the customer loyalty. The increase

of 1.0 brand image will increase 0.084 on customer loyalty.

d) Regression Coefficient of variable trust (X3) was 0.775 that

showed the positive relationship on the customer loyalty. The

increase of 1.0 on trust will increase 0.775 on customer loyalty.

e) From the four of the variables, the most dominant influence on the

customer loyalty was trust because it gained the highest beta value

of 0.716.

3. The Goodness of Fit of Regression Model

One measure of how well a statistical model explains the observed

data is the coefficient of determination, the square of the Pearson

correlation coefficient r between y and y accent. This measure describes

the percentage of the total variance that can be explained from the predictor

criterion covariance. Adjusted R-square measures the proportion of the

variance in the dependent variable (wage) that is explained by variations

in the independent variables. R-square measures the proportion of the

variation in the dependent variable (wage) that is explained by variations

in the independent variables and Std Error of Estimation that measures the

dispersion of the dependent variables estimate around its mean. Compare

this to the mean of the “Predicted" values of the dependent variable. If the

Std. Error is more than 10% of the mean, the predicted value of the

dependent variable is high.

The goodness of fit regression model can be seen from the

determination coefficient. The following were the result of determination

coefficient analysis.

Table 4.14. Coefficient Determination

R Adjusted R Std. Error of the

R
Square Square Estimate
0.967 0.936 0.934 0.20753
Sources: Primary data processed, 2017

a. R value was 0.967, means that the colleration among customer

service, brand image, and trust on customer loyalty was 0.967. It

means that there was a really tight connection.

b. Determination Coefficient (R2) gained 0.936 which means that the

up and down variable of the customer loyalty variable (Y) was

influenced by the independent variables. Those were customer

service, brand image, and trust of 93.6%, while the rest was 6.4%

which was influenced by other factors or variables. Other factors

could be promotion, price, competition condition, and others.

c. Adjusted R Square was gained 0.934. It shows that the contribution

influence of independent variable of service quality, customer

 satisfaction, brand image, and trust on the dependent variable of

customer loyalty was 93.4% and the rest was influenced by other

factors.

d. The Mistake of Standard Estimation (SSE) was used to measure the

accuracy of the real variable value. The bigger the mistake of

standard estimation, the lower the accuracy in explaining the

variable value. The less mistake of standard, the higher the accuracy

in explaining the variable value. Based on the data process the least

value of the standard error of estimation was 0.20753. The lesser

standard mistake, the higher the accuracy in explaining the variable

value.

4. Result of Hypotesis Testing

The hypotesis testing is conducted by examining the path coefficient in

the model which are presented in regression weight output. If coefficient is

significant, the hypothesis of the research is accepted. The criterion for the

hypotesis testing is wether the p value for specified path is less than 0.05 or

not. The result of the regression using IBM SPSS 21.0 software presented

in the F test and T test bellow:

4.1 F test

F test was done to test the significant simultaneous influence

between the independent and dependent variable. This method tested

the hipothesis to determine the level of mistake (α) and search statistical

value (p-value), then compare both scores with the determined

 significant level. Finally, the hipothesis made by the researcher would

be accepted if p-value < α, with the significant value of 0.05 (p-value <

0.05).

Table 4.15. F Test

Model F Sig.
Regression Model 1 446.343 0.000
Source: Primary data processed, 2017

The last column shows the goodness of the fit model. From

the table above, it can be seen that the score of p-value was 0.000 <

0.05. It can be concluded that the variable of service quality, customer

service, brand image, dan trust simultaneously influence the customer

loyalty.

4.2 T test

T test described the reliability of individual coefficients. If the value

in “Sig.” is less than 0.05, it can be assumed that the estimation can be

asserted as true with 95% level of confidence. Below are the results of

t test in this research:

Table 4.16. t Test

variable t Sig.
Customer Satisfaction 3.761 0.000
Brand Image 1.925 0.037
Trust 14.180 0.000
Source: Primary data processed, 2017
 Significant value for the variable of customer service was

0.000<0.05, significant value for variable of brand image was

0.037<0.05 and variable of trust had significant value of

0.000<0.05 which means that each variable influenced customer

loyalty.

4.3 Hypotesis H1

The hypothesis H1 stated service quality influence positively in

customer satisfaction. It is tested by regression coefficient, t test and F

test shows that both constructs have a positive and significant influence.

The value of beta regression in Table 4.8 showed a positive value, and

table 4.10 showed the test value of F was equal to 1198.616 with a

significance of 0.000 whereas the value is smaller than the value of

significant level (0.05). The result of t test has a value of 34.621 with a

significance value of 0.000 whereas the value was less than the value of

significant level (0.05). Significance of 0.000 indicated that the confident

level was 100%. Thus, the hypotesis H1 was accepted.

4.4 Hypotesys H2

The hypothesis H2 stated that customer satisfaction influence

positively on customer loyalty. It is tested by regression coefficient, t test

and F test showed that both constructs have a positive and significant

influence. The value of beta regression in Table 4.13 showed a positive

value, table 4.16 showed the value of t test which was equal to 3.761 with
 a significance of 0.000 whereas the value is smaller than the value of

significant level (0.05). The 0.000 significance value indicated that the

confident level was 100%. Thus, the hypothesis H2 was accepted.

4.5 Hypotesys H3

The hypothesis H3 stated that brand image influence positively on

customer loyalty. It is tested by regression coefficient, t test and F test

showed that both constructs had a positive and significant influence. The

value of beta regression in table 4.13 showed a positive value, table 4.16

showed the value of t test which was equal to 1.925 with the significance

of 0.037 whereas the value was smaller than the value of significant level

(0.05). The significance value of 0.037 indicated that the confident level

was 96.3%. Thus, the hypothesis H2 was accepted.

4.6 Hypotesys H4

The hypothesis H4 stated that trust influence positively on customer

loyalty. It is tested by regression coefficient, t test and F test showed that

both constructs have a positive and significant influence. The value of

beta regression in Table 4.13 showed a positive value, Table 4.16 showed

the value of t test which was equal to 14.180 with the significance of

0.000 whereas value is smaller than the value of significant level (0.05).

The 0.000 significance value indicated that the confident level was 100%.

Thus, the hypothesis H4 was accepted.

4.7 Discussion and Implication

The influence of service quality toward customer satisfaction were

examined using IBM SPSS 24.0. It also examined the influence of

customer satisfaction, brand image and trust toward customer loyalty. T

test and F test showed that service quality influenced customer

satisfaction, while, customer satisfaction, brand image and trust

influenced customer loyalty.

At Gojek transportation, service quality has a positive influence on

customer satisfaction. This was indicated by the value of regression

coefficient, t test and F test which showed positive value. F test value

was equal to 1198.616 with the significance of 0.000 whereas the value

was smaller than the value of significant level (0.05) while the t test value

had a value of 34.621 with a significant value of 0.000 whereas the value

was smaller than the value of significant level (0.05). It could be

interpreted that the higher the quality of service, the higher the customer

satisfaction toward Gojek.

These results were in line with the opinion of Juan & Yan (2009) and

Denio (2006) which stated that one of the factors that influence customer

satisfaction was service quality. These results were also consistent with

Lewis's (1993), Berry et al. (1985), Anderson & Sullivan (1993), Fornell

et al. (1996), Spreng & Macky (1996), Oliver (1993), Lorenzo et al.

(2010), Tariq and Moussaoui (2009) and Heskett, Sasser & Schlesinger
( 1997) which stated that service quality was an antecedant that was

related to customer satisfaction. These results were also consistent to

improve customer loyalty, the company can improve service quality

(Liang Kheng et al, 2010; Juan & Yan, 2009; Kursunluoylu, 2011; Seok

Lee, 2010; Sophoasin, 2010 and Lorenzo et al. 2010; Markovic et al.,

2010). The hypothesis H1 which stated service quality influence

positively on customer satisfaction was accepted.

Furthermore, customer satisfaction influence positively on customer

loyalty. The value of beta regression showed positive value and t test

value was 3.761 with the significance of equal to 0.000 whereas the value

was smaller than level of significant level (0.05). These results indicated

that customer satisfaction had influence on customer loyalty. This was in

line with the opinion of Jennie (1997) in Nuraini (2009), Donio (2006),

Min & Wan (2009), Kheng et al. (2010), Seok Lee (2010), Kursunluoglu,

(2011), Markovic et al (2010), Tariq and Moussaoui (2009) and Ganiyu

et al. (2012) which stated that loyal customers were customers who were

satisfied with certain products. The hypothesis H2 which stated that

customer satisfaction influence positively on customer loyalty was

accepted.

Brand image influence positively on customer loyalty. The value of

beta regression showed positive value and t test value was 3.761 with the

significance of equal to 0.000 whereas the value was smaller than level

of significant level (0.05). These results indicated that customer

satisfaction had an influence on customer loyalty. This was indicated by

a positive value of regression coefficient, t test value of 1.925 with the

significant of 0.037 whereas the value was smaller than the value of

significant level (0.05).

This result was in line with Juan & Yang, Wan Lee (2010) and

Khatibi et al. (2002), Earls et al. (2004) which stated that brand image

was an antecedent of customer loyalty. The results of this research were

also in line with Earls et al. (1998) in Miller & Muir (2004) that a strong

brand image would make customer loyalty stronger. The results of this

research were also in line with Tariq and Moussaoui (2009), Alhaddad

(2015), Halim et al., (2014), Upamannyu & Mathur (2012) that trust was

antecedant to customer loyalty. The hypothesis H3 which stated that

brand image influence positively on customer satisfaction was accepted.

Trust influence positively on customer loyalty. The value of beta

regression showed a positive value, t test value was 14.180 with a

significance of 0.000 whereas the value was less than the value of

significant level (0.05). The results of this research supported the opinion

of Morgan & Hunt (1994) and the opinion of Peppers and Rogers (2004)

which stated that trust was a key variable for successful marketing

relationships.

Thus, as suggested by Liang Kheng et al. (2010) and Tariq and

Moussaoui (2009), to improve customer loyalty, the company could

improve the quality of service to increase customer confidence and

stronger brand image. Because a strong brand image would make

customers more satisfied (Davies et al., 2003).

Brand image had something to do with the appearance of the

product. Thus, customers were satisfied with the product, they would be

loyal to the brand even if they have another brand (Vinhas & Faridah,

2008). But whenever customer needed changes and brands could not

respond to these needs, according to Steel (2004), customers tended to

switch to other brands (Miller & Muir, 2004). Building a positive brand

image could influence the repeat purchases because a strong brand image

could lead to customer loyalty. Furthermore, Schultz (2005) stated that

customers who were loyal to a particular brand would recommend this to

others and were not easily influenced by competitors to make purchases

(Sondoh et al., 2007).

 CHAPTER V

CONCLUSIONS, LIMITATIONS AND RECOMENDATIONS

This chapter clarifies the discussion of the finding in Chapter 4. Based on the

research finding of the analysis, there are several conclusions regarding the effect

of service quality to the customer satisfaction dan effect of customer satisfaction,

brand image dan trust to customer loyalty to Gojek. Several limitations of this study

and recomendations for further study and marketing practicioner are discussed in

the last part of this chapter.

5.1 Conclusions

This study analyzed the influence of service quality on customer satisfaction

and analyzed the influence of customer satisfaction, brand image and trust on

customer loyalty. In addition, a quantitative approach was used to analyze the

influence of service quality on customer satisfaction and the influence of

customer satisfaction, brand image and trust on customer loyalty by using

questionnaires that were distributed offline with a sample of 100 respondents.

From the result of the analysis, it can be concluded that:

1. Service quality influence positively and significantly on customer

satisfaction.

2. Customer satisfaction influence positively and significantly on

customer loyalty.
 3. Brand image influence positively and significantly on customer

loyalty.

4. Trust influence positively and significantly on customer loyalty.

5.2 Limitations

Limitation in this study occurs from several factors. The limitations of the study

are as follows:

1. Respondents were Gojek customer located in Yogyakarta. Although basically

they come from different places, this research had not been able to describe the

loyalty of customer Gojek in Indonesia.

2. Drivers were only Gojek drivers in the area of Yogyakarta. Thus, this research

cannot describe the quality of Gojek service (especially drivers) in a wider

scope (Indonesia).

3. This research focused only on online motorcycle services and ignore other

services offered by Gojek.

5.3 Recomendations

This research proved that service quality had an influence on customer

satisfaction of the customer of Gojek. In line with this, customer satisfaction, brand

image and trust influence on customer loyalty.

To obtain customer satisfaction, Gojek should improve their service quality.

Thus, the customer will be more loyal. On the other hand, Gojek should also

improve their brand image and customer trust through marketing strategy.
 For future study, it is suggested to add several important factors that may

influence customer determination and customer loyalty. The research model in this

research is suggested to be modified and developed to become a better research

construct model.
 REFERENCES

Aaker, David. (1991). Managing Brand Equity; Capitalizing on the Value of Brand
Name. Nrw York: Free Press.

Anderson, E. W., dan Mittal. (2000). Strengthening the Satisfaction-Profit Chain.

Journal of Service Research (3), 170-120.

Anderson, Eugene and Sullivan. (1993). The Antecedents and Consequences of

Customer Satisfaction, Marketing Science 12, 125-143.

Angipora. (2002). Dasar-Dasar Pemasaran. Jakarta: PT. Raja Grafindo Persada.

Bansal, S. & Gupta, G. (2001). Building Customer Loyalty Business-to-Business

Commerce. In J. N. Sheth, A. Parvatiyar & G. Shainesh, eds., Customer
Relationship Management. New Delhi: Tata McGraw-Hill.

Barnes, James G. (2003). Secrets of Customer Relationship Management.

Yogyakarta: Andi.

Bowen, John and Siang-Lih. (2001). The relationship between customer loyalty
and customer satisfaction. International Journal Contemporary Hospitality
Management. International Journal of Contemporary Hospitality
Management, 13 (5), 213-217.

Cannon & Tom. (2008). Basic Marketing. 6 th ed. New York: McGraw-Hill.
International Edition.

Chao, Wu & Yen. (2015). The Influence of Service Quality, Brand Image, and
Customer Satisfaction on Customer Loyalty for Private Karaoke Roomsin
Taiwan. The Journal of Global Business Management, 11 (1), 122-130.

Cronin, Joseph and Taylor. (1992). Measuring Service Quality : Reexamination and
Extension. Journal of Marketing 8(3), 12-19.

Davies, G., Chun, R., da Silva, V., Roper, S. (2003). Corporate Reputation and
Competitive-ness, London: Routledge.

Donio. (2006). Customer Satisfaction and Loyalty in a Digital Environment: an

Empirical Test. Journal of Consumer Marketing, 23 (7), 445-457.

Earl and Jones. (2004). Marketing 2nd edition. Mc Graw Hill Inc : United States
of America
Fornell, C., Johnson, M. D., Anderson, E. W., Cha, J., & Everitt Bryant, B. (1996).
Growing the trust relationship. Journal of Marketing, 60(4), 7-18.

Ganiyyu, Uche, Elizabeth. (2012). Is Customer Satisfaction an Indicator of

Customer Loyalty?. Australian Journal of Business and Management
Research , 2 (7), 14-20.

Griffin and Lemmon. (2002). Book-to-Market Equity, Distress Risk, and Stock
Returns. Journal of Finance. 57 (5), 2317–2336.

Halim, Heskett, Sasser and Schlesinger. (1997). The Service Profit Chain, Free
Press, New York, NY. European Journal of Business and Management, 6
(29), 112-121.

Hidayat, Saifullah & Ishak. (2015). Determinants of Satisfaction, Trust, and

Loyalty of Indonesian E-Commerce Customers. Pertanika Journal, 10(S1),
151-166.

Hoyer, W. D., & MacInnis, D. J. 2001. Consumer Behaviour. 2nd ed., Boston:
Houghton Mifflin Company

Jagdish N. Sheth, Banwari Mittal. (2004). Customer Behavior: A Managerial

Perspective.. South-Western : Thomson.

Johnson, Herman & Huber. (2006). The Evolution of Loyalty Intentions. Journal
of Marketing, 70 (1), 122–132.

Juan & Yan. (2010). Dimensions And Influencing Factors Of Customer Loyalty In
The Intermittent Service Industry. Journal Frontiers of Business Research
in China, 3(1), 63-78.

Keh, H. T., & Xie, Y. (2009). Corporate Reputation And Customer Behavioral
Intentions: The Roles Of Trust, Identification And Commitment. Industrial
Marketing Management, 38(1), 732–742.

Kotler dan Keller. (2012). Marketing Managemen 14E. New York. Prentist Hall.
Keller. (1998). Strategic Brand Management: Building, Measuring, and Managing
Brand Equity, Prentice Hall, New Jersey.

Khatibi, Ismail, Thyagarajan. What Drives Customer Loyalty: An Analysis From

The Telecomunication Industry. Journal of Targetting, 11 (1), 33-34.

Kodyakof. (2006). Trust as a Process:A Three-Dimensional Approach. Sociology.

41 (7). Sage Journal 41(1), 77-83.
Kotler and Amstrong. (2001). Prinsip-prinsip Pemasaran. Edisi keduabelas. Jilid
1. Jakarta: Erlangga.
Kotler dan Keller. (2009). Manajemen Pemasaran. Jilid I. Edisi ke 13 Jakarta:
Erlangga.

Kursunluoglu. (2011). Customer Service Effects on Customer Satisfaction and

Customer Loyalty: A Field Research in Shopping Centers in Izmir City –
Turkey. International Journal of Business and Social Science, 2(17), 331-
341.

Lee. (2010). Critical Factors That Establish Customer Loyalty To Mobile

Communications Service Providers: Empirical Evidence From The United
States. Proceedings of the Academy of Marketing Studies 15 (2), 77-85.

Lewis, B. R. (1993). Service Quality: Recent Developments In Financial Services.

International Journal of Bank Marketing, 2 (6), 19-26.

Lorenzo & Lane. (2010). How Service Quality Affects Customer Loyalty.
Proceedings of the Academy of Information and Management Sciences, 14
(1).

Loveman, G.W. (1998). Employee Satisfaction, Customer Loyaly, And Financial

Performance: An Empirical Examination Of The Service Profit Chain In
Retail Banking. Journal of Service Research, 1 (1), 18-31.

Marcovic, Raspor, Segaric. (2010). Customer Satisfaction And Customer Loyalty

Measurement In Hotel Settings: An Empirical Analysis. Tourism &
Hospitality Management, Conference Proceedings.

Miller & Muir. (2004). The Business of Brands. John Wiley & Sons, Ltd.

Morgan and D. Hunt. 1994. The Commitment-Trust Theory of Relationship

Marketing, Journal of Marketing, 58 (3), 20-38.

Nuraini. 2009. Analisis Pengaruh Kualitas Produk, Kualitas Pelayanan, Desain.

Produk, Harga dan Kepercayaan Terhadap Loyalitas Pelanggan. Skripsi.
Unicom. Retrieved From:
http://elib.unikom.ac.id/files/disk1/570/jbptunikompp-gdl-lukhipurna.pdf

Oliver, R. (1993). Satisfaction as Behavioral Perspective on the customer. NY:

McGraw-Hill.

Ouwersloot, Hans and Tudorica Anamaria. (2001). Brand Personality Creation

Through Advertising. Maxx Working Paper 01.
Papasolomou I. and Idowu, S. O. (2006). Are the Corporate Social Responsibility
Matters Based on Good Intention or False Pretences? An Empirical Study of
the Motivations Behind the Issuing of CSR Reports by UK Companies.
Corporate Governance Journal, 7 (2). 11-19.

Peppers, Don dan Rogers, Martha. (2004). Managing Customer Relationships: A

Strategic Framework. USA: John Wiley and Sons, Inc.

Peter and James. (2007). Marketing Management. New York Americas: The mc
Grow Hill Companies.

Rai & Sri vastava. (2012). Customer Loyalty Attributes:A Perspective. Nmims
Management Review XXII.

Schultz, D. E. (2005). The loyalty Paradox. Journal Marketing Management, 14(5),

10–11.

Sekaran. (2000). Research Methods for Business: A Skill Building Approach.

Jakarta: Salemba Empat.

Seok Lee. (2010). Factors Influencing Customer Loyalty of Mobile Phone Service:
Empirical Evidence from Koreans. JIBC, 15 (2), 12-21.

Shimp. (2003). Periklanan Promosi & Aspek Tambahan Komunikasi. Pemasaran

Terpadu. Jilid I ( edisi 5). Jakarta: Erlangga.

Silvia, Da., Vinhas, Rui., Alwi, Syed., & Faridah, Sharifah. (2008). Online
Corporate Brand Image, Satisfaction and Loyalty. Journal of Brand
Management 4 (119-144).

Sondoh. (2007). The Effect Of Brand Image On Overall Satisfaction And Loyalty
Intention In The Context Of Color Cosmetic. Journal Management Kedah
Campus, Malaysi, 2(1), 34-44.

Spreng & Macky. (1996). A Reexamination of the Determinants of Consumer

Satisfaction. Journal of Marketing 60 (2), 13-32.

Steel, R.P dan Ovalle, N.K. (2004). A Review and Meta Analysis on the Relatioship
Behavioral Intention and Employee Turnover. Journal of Applied
Psychology, 69 (2), 673-686.

Sumarwan. (2003). Perilaku Konsumen Teori dan Penerapannya dalam.

Pemasaran, Cetakan Pertama, Penerbit Ghalia lndonesia.

Suwardi. (2011). Menuju Kepuasan Pelanggan Melalui Penciptaan Kualitas

Pelayanan. Jurnal Pengembangan Humaniora, 11 (1), 9-18.
Svensson, G. and Wood, G. (2005), The serendipity of leadership effectiveness in
management and business practices. Journal Management Decision, 43 (7),
1001-9.

Tariq and Moussaoui. (2009). The Main Antecedent of Customer Loyalty in

Moroccan Banking Sector. International Journal of Business and
Management Science, 2(2), 101-115.

Tjiptono and Gregorius. (2005). Managemen Kualitas Jasa. Yogyakarta: Penerbit

Andi.

Tjiptono. (2006). Manajemen Jasa. Edisi Pertama. Yogyakarta : Andi.

Todd (2007). Trust Measures and Indicator for Customers and Investors. The
Conference Board of Canada: Building Customers Trust A Roundtable
Discussion.

Upamanyu & Mathur. (2012). Effect Of Brand Trust, Brand Affect And Brand
Image On Customer Brand Loyalty And Consumer Brand Extension Attitude
In FMCG Sector, 3 (2), 21-33.

Upammanyu & Sankal. (2014). Effect Of Brand Image On Customer Satisfaction

& Loyalty Intention And The Role Of Customer Satisfaction Between Brand
Image And Loyalty Intention. Journal of Social Science Research, 3 (2), 44-
52.

Usmara. (2003). Strategi Baru Manajemen Pemasaran. Yogyakarta: Amoro Book.

Wilson A., Zeithaml V.A., Bitner M.J., Gremler D.D. (2008). Services Marketing,
McGraw-Hill Education.

Yani. (2017, May 12). Gojek Gandeng 3 Bank. Koranjakarta.com. Rertieved from
http://www.koran-jakarta.com/gojek-gandeng-tiga-bank/.

Yuen & Thai. (2015). Service Quality And Customer Satisfaction In Liner
Shipping. International Journal of Quality and Service Sciences, 7 (3), 170-
183.

Zeithaml, V. A., Bitner, M. J., &Gremler, D. D. (2006). Service marketing:

Integrating customer focus across the firm (4th ed.). New York, NY:
McGraw-Hill/Irwin. p. 106-107.
APPENDICES
Appendix 1

QUESTIONNAIRE

Responder,
I beg your willingness to take the time to fill out this questionnaire. The honest
answers you provide are very useful for the research I'm doing, under the heading:
"Relationship between Service Quality, Customer Satisfaction, Brand Image, and
Trust With Customer Loyalty of Gojek Transportation Service In Yogyakarta City".

PART I

Directions: Please answer the following questions by put a mark (x) on the
answer you think is most correct.

1. What is your gender?

a. Man
b. Woman
2. You old are you?
a. ≤ 18 years
b. 19- 24 years
c. 25 - 30 years
d. > 30 years
3. Where do you come from?...................................................
4. What is your occupation?
a. Student
b. Public Servant / Private
c. Entrepreneur
d. Other____________
4. What is your intensity in using the Gojek Service
a. Rarely
b. often
c. very often
d. Other____________
5. Have you used other Gojek services (Go food, Go Glam, etc)?
a. Yes
b. No
6. How much is your income/salary per month?
a. 500,000 – 1,500,000 rupiahs
 b. 1,500,000 – 2,500,000 rupiahs
c. 2,500,000 – 3,500,000 rupiahs
d. Other_____________

PART II
Directions: Please answer the following questions by put a mark (x) on the
answer you think is most suitable.

Information:
SD : Strongly Disagree
D : Disagree
SLD : Slightly Disagree
N : Neither
SLA : Slightly Agree
A : Agree
SA : Strongly Agree

A. Customer Loyalty

No. Item SD D SLD N SLA A SA

1. I,m always decide that the company is the
first choice
2. The customer From the experience that
customer had, the company was the first
in the mind of customer
3. How likely are you to recommend this
service to friends, neighbors, and
relatives?

B. Service Quality

No. Item SD D SLD N SLA A SA

1. Gojek Consistency of customer
2. Gojek application error
3. Gojek always on time
4. Frequency of transportation services
5. Gojek have accuracy of invoice
6. Effectiveness of driver
7. Promptness of customer service
No. Item SD D SLD N SLA A SA
8. Gojek have variety of service offering
9. Gojek CSR and concern for human safety
10. Pricing of shipping service cheap
11. Gojek is safety and security riding

C. Customer Satisfaction

No. Item SD D SLD N SLA A SA

1. Gojek make me delighted
2. Gojek give me pleasure increases my
frequency of use
3. Gojek make feel good
4. Gojek prevent me for looking cheap
another brand

D. Brand Image

No. Item SD D SLD N SLA A SA

1. Provides a solution to my expectations
good impression
2. Dependable for use make me comfort
3. Feel delighted, fell good, feel frequency
of use
4. Looking cheap, desirable lifestyle, gives
me pleasure.
5. Perceived by others, perform it promises,
feel accepted my social group
E. Trust

No. Item SD D SLD N SLA A SA

1. I feel safe with the service
2. I believe with the company’s promise
3. I believe that the company’s quality
service and consistent
4. Gojek driver respect customer
5. I meet their obligations to customer
6. I believe with company’s service
Appendix 2

Regression Output
Regression 1
 Descriptive Statistics
Mean Std. Deviation N
CS 5,7734 ,92344 100
SQ 5,7380 ,73259 100

Correlations
CS SQ
Pearson Correlation CS 1,000 ,1003
SQ ,1003 1,000
Sig. (1-tailed) CS . ,000
SQ ,000 .
N CS 100 100
SQ 100 100

Variables Entered/Removeda
Model Variables Entered Variables Removed Method
1 SQb . Enter
a. Dependent Variable: CS
b. All requested variables entered.

Model Summaryb
Std. Change Statistics
R Adjuste Error of R F
Mo Squar dR the Square Chan Sig. F Durbin-
del R e Square Estimate Change ge df1 df2 Change Watson
1 ,1003 ,927 ,927 ,25034 ,927 1198, 1 94 ,000 1,776
a 616
a. Predictors: (Constant), SQ
b. Dependent Variable: CS

ANOVAa
 Model Sum of Squares df Mean Square F Sig.
1 Regression 75,119 1 75,119 1198,616 ,000b

Residual 5,891 94 ,063

Total 81,010 95

a. Dependent Variable: CS

b. Predictors: (Constant), SQ

Coefficientsa
Standardize
Unstandardized Collinearity
d Correlations
Coefficients Statistics
Model Coefficients t Sig.
Zero- Toleranc
B Std. Error Beta Partial Part VIF
order e

(Constan
-1,191 ,203 -5,875 ,000
1 t)

SQ 1,214 ,035 ,1003 34,621 ,000 ,1003 ,1003 ,1003 1,000 1,000

a. Dependent Variable: CS

Coefficient Correlationsa
Model SQ
1 Correlations SQ 1,000
Covariances SQ ,001
a. Dependent Variable: CS

Collinearity Diagnosticsa
Variance Proportions
Model Dimension Eigenvalue Condition Index (Constant) SQ
1 1 1,992 1,000 ,00 ,00
2 ,008 15,810 1,00 1,00
a. Dependent Variable: CS

Residuals Statisticsa
Minimum Maximum Mean Std. Deviation N
Predicted Value 3,2269 6,9775 5,7734 ,88923 100
Residual -,66382 ,65389 ,00000 ,24902 100
Std. Predicted Value -2,864 1,354 ,000 1,000 100
Std. Residual -2,652 2,612 ,000 ,995 100
a. Dependent Variable: CS

Charts
Regression 2

Descriptive Statistics
Mean Std. Deviation N
CL 5,7319 ,80544 100
CS 5,7734 ,92344 100
BI 5,8500 ,62912 100
TR 5,7579 ,79302 100

Correlations
CL CS BI TR
Pearson Correlation CL 1,000 ,884 ,459 ,1000
CS ,884 1,000 ,399 ,867
BI ,459 ,399 1,000 ,424
TR ,1000 ,867 ,424 1,000
Sig. (1-tailed) CL . ,000 ,000 ,000
CS ,000 . ,000 ,000
BI ,000 ,000 . ,000
TR ,000 ,000 ,000 .
N CL 100 100 100 100
CS 100 100 100 100
BI 100 100 100 100
TR 100 100 100 100

Variables Entered/Removeda
Variables
Model Variables Entered Removed Method
1 TR, BI, CSb . Enter
a. Dependent Variable: CL
b. All requested variables entered.

Model Summaryb
Std. Change Statistics
R Error of
Mod Squar Adjusted the R Square F Sig. F Durbin-
el R e R Square Estimate Change Change df1 df2 Change Watson
 1 ,1007 ,936 ,934 ,20753 ,936 446,343 3 92 ,000 1,531
a

a. Predictors: (Constant), TR, BI, CS

b. Dependent Variable: CL

ANOVAa
Sum of
Model Squares df Mean Square F Sig.
1 Regression 57,668 3 19,223 446,343 ,000b
Residual 3,1002 92 ,043
Total 61,630 95
a. Dependent Variable: CL
b. Predictors: (Constant), TR, BI, CS

Coefficientsa
Standardize
Unstandardized d Collinearity
Correlations
Coefficients Coefficient Statistics
Model t Sig.
s
Std. Zero- Toleranc
B Beta Partial Part VIF
Error order e

(Constan
-,160 ,214 -,748 ,456
t)

1 CS ,174 ,046 ,200 3,761 ,000 ,884 ,365 ,099 ,247 4,047
BI ,072 ,037 ,056 1,925 ,037 ,459 ,197 ,051 ,816 1,225
TR ,775 ,055 ,763 14,180 ,000 ,1000 ,828 ,375 ,241 4,146
a. Dependent Variable: CL

Coefficient Correlationsa
Model TR BI CS
TR 1,000 -,169 -,840
Correlations BI -,169 1,000 -,071
CS -,840 -,071 1,000
1
TR ,003 ,000 -,002
Covariances BI ,000 ,001 ,000
CS -,002 ,000 ,002
a. Dependent Variable: CL
 Collinearity Diagnosticsa
Variance Proportions
Eigenvalu Condition (Constant
Model Dimension e Index ) CS BI TR
1 1 3,975 1,000 ,00 ,00 ,00 ,00
2 ,017 15,357 ,16 ,13 ,11 ,04
3 ,006 26,507 ,78 ,00 ,88 ,00
4 ,003 37,990 ,05 ,87 ,00 ,100
a. Dependent Variable: CL

Residuals Statisticsa
Minimum Maximum Mean Std. Deviation N
Predicted Value 3,3336 6,9047 5,7319 ,77912 100
Residual -,91054 ,71268 ,00000 ,20422 100
Std. Predicted Value -3,078 1,505 ,000 1,000 100
Std. Residual -4,388 3,434 ,000 ,984 100
a. Dependent Variable: CL

Charts