Channel perormance enablers, channel conflict and channel member performance an investigation into vietnam airlines agencies

ABSTRACT This study investigates the relationships between channel performance enablers which include trust, commitment and cooperation, as well as channel conflict, and channel member p

UNIVERSITY OF ECONOMICS HO CHI MINH CITY

International School of Business

UNIVERSITY OF ECONOMICS HO CHI MINH CITY

International School of Business

MASTER OF BUSINESS (Honours)

SUPERVISOR: LUU TRONG TUAN PH.D

ACKNOWLEDGEMENTS

This thesis could not have been possible without the help of many people

First, I would like to take this opportunity to express my sincere gratitude to Dr Luu Trong Tuan for his clear and careful direction, his guidance and his corrections of my thesis I especially appreciate his continual and on-going enthusiasm in responding, reminding, and assisting in all aspects of this endeavor Secondly, I would like to express my sincere thanks to my company, Thien Hai Services Travel Trading Corporation, as well as my managers and supervisors for their patience, understanding and support of my thesis

Thirdly, I would like to thank my dear colleagues, friends and classmates for their valuable encouragement during the many months that I was working on this thesis Special thanks are also sent to those individuals working for the Vietnam Airlines’ agencies who have volunteered their valuable time in meeting, speaking with and responding to my thesis survey

Last, but not least, I would like to express my heartfelt thanks to my family for their support, their encouragement, and their faith while completing my thesis

ABSTRACT

This study investigates the relationships between channel performance enablers which include trust, commitment and cooperation, as well as channel conflict, and channel member performance in service industries, especially in the area of logistics as found in the airlines industry The objective of this study is to determine the influence that both channel performance enablers and channel conflict has on channel member performance Our research questions were: “Are there positive relationships between channel performance enablers and channel member performance?” “Is there a negative relationship between channel conflict and channel member performance?”

To test the hypotheses, we studied 208 respondents working in 18 various Vietnam Airlines’ agencies The data collection was based on the questionnaire adapted from previous studies and modified for the service industry as well as for the Vietnam context (Dwyer and Oh, 1987; Kumar et al, 1995; Lumineau and Oxley, 2007; Sezen and Yilmaz, 2007)

As expected, the results of the study show the existence of positive relationships between channel performance enablers and channel member performance In these relationships, cooperation has the strongest positive relationship with channel member performance (ß = 0.284, p<0.001) It shows that to achieve high channel member performance, agencies should increase the cooperation in the channel Next, commitment has the second strongest positive relationship with channel member performance (ß = 0.237, p<0.05) And last, trust has the least

strongest positive relationship with channel member performance (ß = 0.214, p<0.05)

However, the findings did not state the negative relationship that exists between channel conflict and channel member performance (ß = 0.099, p=0.124)

Past and previous research focused on the relationships between channel performance enablers and channel member performance primarily in the manufacturing sector This study here contributes to the practical theory between channel performance enablers and channel member performance in a different industry: the service industry

Key words: Trust, commitment, cooperation, channel conflict, channel member performance, Vietnam Airlines’ agencies

TABLE OF CONTENTS

ACKNOWLEDGEMENTS i

ABSTRACT ii

LIST OF FIGURES vii

LIST OF TABLES viii

CHAPTER 1 1

INTRODUCTION 1

1.1 Problem statement 1

1.2 Research Aims 3

1.3 Research scope 3

1.4 Research contribution 4

1.5 Research structure 4

CHAPTER 2 6

LITERATURE REVIEW 6

2.1 Channel member performance 6

2.2 Trust and channel member performance 8

2.3 Commitment and channel member performance 10

2.4 Co-operation and channel member performance 12

2.5 Channel conflict and channel member performance 14

2.6 Research Model 16

CHAPTER 3 18

RESEARCH METHODOLOGY 18

3.1 Research process 18

3.1.1 Step 1: Item generation step 19

3.1.2 Step 2: Pilot study 21

3.1.3 Step 3: Main study 21

3.2 Data collection 22

3.2.1 Sample size 22

3.2.2 Sampling method 23

3.3 Data source 23

3.4 Measurement of variables 24

3.5 Methods of data analysis 27

3.5.1 Reliability analysis measured by Cronbach’s Alpha 27

3.5.2 Validity measured by Exploratory Factor Analysis (EFA) 27

3.5.3 Multiple regression analysis 29

CHAPTER 4 32

DATA ANALYSIS 32

4.1 Preparation data 32

4.2 Descriptive statistics 32

4.3 Cronbach’s Alpha analysis 33

4.4 Exploratory Factor Analysis (EFA) 36

4.5 Regression analysis 40

4.5.1 Testing of the fit level of model 41

4.5.2 Results of regression analysis 47

CHAPTER 5 53

CONCLUSIONS 53

5.1 Conclusions 53

5.2 Manegerial implications 53

5.3 Limitations and further research 55

REFERENCES 59

APPENDIX 1 : QUESTIONNAIRE – ENGLISH VERSION 67

APPENDIX 2 : QUESTIONNAIRE – VIETNAMESE VERSION 71 APPENDIX 3 : DATA ANALYSIS RESULTS 75

LIST OF FIGURES

Figure 2.1: Conceptual Framework 16

Figure 3.1: Research Process 19

Figure 4.1: Scatter plot 43

Figure 4.2: Regression Standardized Residual 46

Figure 4.3: Normal P-P plot of Regression Standardized Residual 47

LIST OF TABLES

Table 2.1: Factors and items 25

Table 4.1: Descriptive statistic of samples 32

Table 4.2: Cronbach’s Alpha of four factors and channel member performance 34

Table 4.3: Means, standard deviations and reliabilities of channel performance enablers, channel conflict and channel member performance 34

Table 4.4: KMO and Bartlett’s Test for factors 37

Table 4.5: Total Variance Explained 38

Table 4.6: Rotated Component Matrix 39

Table 4.7: Casewise diagnostics (time 1) 41

Table 4.8: Casewise diagnostics (time 2) 42

Table 4.9: Model Summaryb 42

Table 4.10: Correlation matrix for variables 44

Table 4.11: Coefficients 45

Table 4.12: Model Summary 47

Table 4.13: Hypothesis testing results 51

Companies benefit from channel strategies in a variety of ways It allows them to understand customer needs as well as their purchasing patterns Companies with different kinds of products can satisfy much more of their customer needs as well

as attract new customers The large existing distributions can enable companies to focus on more precise target markets, improving their competitiveness (Kevin and John, 2002)

However, multiple channels can cause conflicts within their own systems In fact, channel conflict is the most serious concern for companies In a survey, Gilbert and Balcheldor (2000) state that channel conflict is the biggest issue companies face today in their sales strategy So, companies are attempting to redefine and redesign the structure of their channel distribution and make it more efficient while improving the channel member performance as well There is a need for empirical research to understand channel conflict and its influence on channel

member performance The first purpose of this paper is to establish the relationship between channel conflict and channel member performance

The trends that influence consumers’ preference for determined products and services have a strong impact on the issues related to channel management (Frazier, 1999) Those channels are combined in an integrated procedure from the channel members and the consumers (Rosenbloom, 2003) The relationships between channel members are very complicated Their goals and objectives are different Therefore, they tend to desire to achieve each side’s goals and ignore the others This leads them to be distant in their own relationship By improving and getting them closer to each other’s goals and objectives, companies have to build some channel performance enablers, including trust, commitment and cooperation Channel performance enablers are mentioned in many previous research projects to find the most effective way for manufacturers as well as distributors (Rousseau et al., 1998) They show that the channel performance enablers make the whole system operate more efficiently and more effectively They probe the primary aspects of channel member performance; such as, how trust, commitment, and cooperation impacts the system

However, they have studied some of these aspects separately They have individually studied the relationship by either conflict or trust or commitment or cooperation in the relationship between them in manufacturing (Nair, 2008; Lusch, 1976) This paper will simultaneously study how channel conflict and all channel performance enablers have influence on the channel member performance

in these complex relationships in one specific model In this study, we also

investigate the relationships between them in a service industry such as the airlines, an important logistic industry, instead of manufacturing like previous research projects In addition, this paper will focus on the special characteristics

of an emerging market like Vietnam

1.2 Research Aims

For understanding the influence of channel performance enablers including trust, commitment, and cooperation, and channel conflict on channel member performance between Vietnam Airlines’ and its agencies, this research aims to investigate:

1) The positive relationship between channel performance enablers and channel member performance

2) The negative influence of channel conflict on channel member performance

1.3 Research scope

Vietnam Airlines’ agencies are numerous and of various types According to Vietnam Airlines statistical nationwide figures, there are 222 agencies in the South, 64 agencies in the Central and 166 agencies in the North With over 450 company agencies, channel performance enablers and channel conflict between Vietnam Airlines and its agencies play an important role in its agencies’ contribution and participation as well as its success and performance However due to the limitations of data availability, research funds and time, the scope of this research is therefore limited to Vietnam Airlines’ agencies located in Ho Chi Minh City and will investigate the influence of channel performance enablers and

channel conflict on channel member performance This is considered appropriate because Ho Chi Minh City has the highest number of Vietnam airlines’ agencies

1.4 Research contribution

The paper helps to expand the amount of literature about the relationships that exist between channel performance enablers, channel conflict and channel member performance It identifies the factors that enable and improve channel member performance as well as the factors that disable and destroy channel member performance Based on these factors, managers and agencies can find the best way to increase channel member performance

It also probes the relationships between channel performance enablers, channel conflict and channel member performance in a new industry: services instead of manufacturing In an emerging market like Vietnam, the services industry is considered as a “targeted” industry to help develop the economy

Additionally, the findings will help to examine the relationships between channel performance enablers, channel conflict and channel member performance studied

in an emerging market within a specific culture: Southeast Asia

1.5 Research structure

This study is divided into five chapters as follows:

The first chapter is an “introduction” which includes the problem statement, research objectives, research scope, research contribution and the structure of this research

The second chapter is “literature reviews” which summarizes the concepts, the theories, the experiences and previous research about the elements that contribute

to the relationship between trust, commitment, cooperation, channel conflict and channel member performance The research model including hypotheses and conceptual framework is also given in this chapter

The third chapter pertains to the “research methodology” It introduces the research process, data collection and methods of data analysis

The fourth chapter explains the “data analysis” It reports the analysis results from the collected data obtained via the surveys

The last chapter is devoted to “conclusions” It discusses the result of the “data analysis” part and provides recommendations

CHAPTER 2

LITERATURE REVIEW

This chapter provides the definitions of channel member performance, channel performance enablers including trust, commitment, cooperation and channel conflict as well as discusses the relationships between them and channel member performance based on the previous research

2.1 Channel member performance

Channel member performance can be defined as “the degree to which the channel member engages in behavior that contributes to the fulfillment of the channel leader’s objectives” (Gaski and Nevin, 1985, p.136) Since the channel members can significantly influence a firm’s long-term success or failure, companies are continuously concerned about the performance levels of the organizations that comprise their marketing channels Moreover, the level of performance attained

by channel members can be critical if a firm is to achieve a differential advantage (Porter, 1985; Rosenbloom, 1973) However, all the previous research on marketing channels show that research on marketing channel member performance has not been their primary emphasis and little is known about what actually influences channel members to perform effectively in selling the products

of the firms they represent This may be partly due to the lack of a general agreement over the definition of performance (Lewis and Lambert , 1991) Terms such as effectiveness and productivity have been used and misused repeatedly, which has made the clear determination of what exactly performance is, very difficult

In this study, a managerial view is used to assess the performance of individual channel members It focuses on measuring the channel efficiency of channel members because the performance of the overall marketing channel is determined

to a large extent by the performance of individual channel members comprising the channel Channel member performance efficiency is measured by two indices: productivity and profitability Channel productivity refers to the efficiency with which output is generated by the inputs used or expended and generated from resources (Stern et al., 1998) Channel productivity is a measure of efficiency in using inputs (measured in terms of labor and capital) to generate outputs (measured in terms of revenues, gross margins, and value added) Channel profitability is a general measure of the financial performance of channel members in terms of return on investment, growth in profits, and liquidity leverage, among other indices In short, channel profitability indicates how efficiently marketing channel members utilize their financial resources and represent economic success or factors related to it (Rosenbloom, 1999)

A periodic assessment of channel member performance is crucial to ensure that the channel is progressing towards its goals and objectives According to the resource-based theory, a firm’s competitive advantage is a function of its valuable, rare and inimitable resource (Barney, 1996) For intermediaries, such resources are often intangible, embedded and knowledge-based such as experience and knowledge of the market, and include skills such as negotiation skills and skills in channel activities

2.2 Trust and channel member performance

Trust is a concept that is covered as a vagueness since “trust has never been a topic of mainstream sociology” (Luhmann, 1998) Despite an immense amount of research, a generally accepted definition of trust seems to be lacking in clarity (Lewis and Weigert, 1985) Recent discussions have brought huge attention to this area for many researchers The purpose is to bring more structure and order to this concept The concept of trust is very closely linked to risk and expectations Interestingly, trust is used as a substitute for risk At the same time, trust creates a risk for the truster As Baier (1986) clearly explains: “trust involves the belief that others will, so far as they can, look after our interests, that they will not take advantage of us and will not harm us Therefore, trust involves personal vulnerability caused by uncertainty about the future behavior of others Put more succinctly, we cannot be sure, but we believe, that they will be benign, or at least not malign, and act accordingly in a way which may possible put us at risk.” This implies that there is no defined certainty From all of this comes an acceptable postulation that the increasing distrust in our society is a direct result of the evolution to a post-modern society People today are faced with increasing uncertainty in many facets of their lives So trust is used to reduce this uncertainty Trust is never absolute but always conditional and contextual; it depends on the content on how it is to be defined

Most of the research that has been conducted on trust deals with interpersonal trust; that is, trust between people Interpersonal trust is sometimes approached at the aggregate level This term can be defined as something based on most

individuals It is a visible pattern in society as a whole This is contrary to being based on specific people or specific cases This is important because we are determining as the best we can, a clear explanation and understanding of trust Another perspective approaches trust from the most basic level: interpersonal interaction between two or more people So trust is defined as using a behavior that has a lot of forethought involved Therefore, trust is a strategy utilized to enhance a person’s advantage

Trust enables cooperative human endeavors (Fukuyama, 1996) and is vital to inter-organizational relationships (Gambetta, 1988; Rousseau et al., 1998; Blomqvist, 2002) People naturally trust people, more than technology (Friedman

et al., 2000) Trust is also seen as an expression of confidence in an organizational environment This leads to cooperative behavior among individuals and groups within and between organizations (Nandhakumar et al, 2006)

According to Bijlsma & Koopman (2003), trust is a key to organizational performance because it makes cooperation voluntary This form of cooperation becomes increasingly important when command and control styles of management are no longer effective Trust is a solution for specific problems of action in preference to alternatives, in spite of the possibility of being disappointed by the action of others (Luhmann, 1988) Trust is a complex concept because it is related to the confident expectation that persons involved in the action will act competently and dutifully (Smith, 2005)

Past research has shown the influence of trust on channel member performance (Rich, 1997; Robinson, 1996) A high level of trust in an agency increases the

chance that one will take a risk such as co-operation and sharing information Therefore, a high degree of trust may lead to increase channel member performance, leading to the following hypothesis:

H1: There is a positive relationship between trust and channel member

performance

2.3 Commitment and channel member performance

There are several variations for the definition of commitment suggested by researchers Attention was given to the concept of commitment after the study of Kantor in the 1970s This was followed in the 1980s when American managers started to study Japanese companies to figure out the reasons why they were so successful They surmised that this success was somehow attributed to the channel members’ loyalty The interest in the relationship between the mobility

of channel members and the concept of commitment then continued to gain widespread attention in the 1990s (Yilmaz and Dil, 2008)

Since the definition of commitment constantly changes, depending on who is using the term, it is certain that the ideas pertaining to commitment, such as

“whether the commitment is good or bad, constant or variable”, will change primarily due to the flexibility of the use of the term

Many theoretical studies about commitment have focused on the channel member’s commitment to the system From taking this approach, there have emerged two different trends of thoughts Morrow (1983) suggests that commitment should be separated from other commitment concepts such as work

commitment, business ethics, addiction to career He contends that commitment should be dealt with separately from its related concepts So it is good to comprehend the effects that this concept has on channel members’ performance Basic business tenets state that the survival of a channel is dependent on the members’ retention in the system It is argued that a system is as powerful as the level of the commitment of the channel members to that system Systems try to prevent channel members from quitting by using incentives such as commission increases, improved opportunities, better working conditions, promotions and other rewards (Bayram, 2005)

When a channel decides to make an unconditional commitment to the channel process, this action represents a long-term orientation toward the selected channel relationship (Narus and Anderson, 1986) The motivation behind this action, the engine that is powering this action, is for one of three reasons: enhancing the value of the channel’s market offering to its customers or to lower the channel’s total cost or it could even be for both reasons (Stern and El-Ansary, 1992) The most important outcome from doing all this is to improve channel member performance

Once this commitment to channel relationships is established, channel members automatically share mutual goals and objectives Their values are also shared among themselves It’s similar to an individual joining a team Once committed to play, then all else is in synch Once committed, they all work more closely to achieve both their individual as well as joint goals Therefore it can be said that this commitment can lead to increased coordination within the channel which will

enable all channel members to experience increased and improved performance (Anderson and Weitz, 1989) Unless there is instinctive and coordinated behavior, the channel may have to develop more extensive control systems to guarantee the desired level of performance All this results the increase of channel costs (O’Relly and Chatman, 1986) All of these actions impact performance for the entire channel as well as for each member who is part of the channel Accordingly, our second hypothesis states:

H2: There is a positive relationship between commitment and channel

member performance

2.4 Co-operation and channel member performance

Social relationships are based on the understanding that the participating parties will be involved with a long-term commitment Gilliland and Bello (2002) contend that in this type of relationship, social goals are more important than economic goals or objectives Heide (1994) adds that socially bonded parties

“forego present rewards on the basis of an expectation of long run equity.” So, in situations where the results reflect low economic performance, partners react to this by strengthening their bonds They also work hard for long-term success of the relationship In addition to the motivation to maintain their relationship, there

is another factor that uncovers the greater effectiveness of cooperation in the case

of past poor-yielding performance Its foundation is based on the beneficial disrupting effects of difficulty on some negative aspects of inter-firm relationships when everything is going well Selnes and Sallis (2003) describe a “dark side” of relationships, which in some cases involves the need for excessive care to

maintain a pleasant relationship but which can lead to counterproductive behaviors among the channel participants These behaviors include:

The intentional avoidance of announcing negative information in order to keep good relationships between members in channel;

The shortage of monitoring of the distributor activities, which causes the lack

of necessary information between the suppliers products and its performance with these products; and

Loss of creativity in case both partners fall into a group consensus rather than adopting independently individual action

These elements can result in a reduced quality of cooperation because the exchange of information between the parties is an essential element of cooperation In addition, cooperation is about solving problems together If the partners do not raise important issues, they are likely to remain unresolved

Gilliland (2004) contends that the pressure to obtain high economic performance can change a firm’s attitudes and behaviors A low performance jeopardizes the continuation of the relationship and acts as a deterrent to continue with a status quo behavior as if all is running smoothly Members will realize the seriousness

of the problems and issues facing the future of their relationship and must openly and frankly discuss the actual problems that need to be solved

No channel member can perform all of the functions necessary to move the product through the channel which forces all the channel members into a state of mutual dependence Because each firm depends on the other to carry out its functions efficiently and economically, cooperation among members is essential and a vital behavior (Lynn and Robert., 1996) Moreover, the major goal of

channel members is usually profit maximization Therefore, members need to perform their functions well in order to contribute to this goal which in turn will improve their performance Additionally, it is useful to view channel members as important organizations composed of interdependent institutions because they must cooperate to perform distribution tasks in terms of pursuing independent and collaborative goals Cooperation between members can help improve the competitiveness of the channel (Martha, Ellram and Lisa M, 1997) and the interaction between members can be described as a social exchange process in which each channel member interactively increase its commitments to the relationship Therefore, high levels of co-operation between an agency and Vietnam Airlines may lead to increase and improve channel member performance Our third hypothesis is:

H3: There is a positive relationship between co-operation and channel

member performance

2.5 Channel conflict and channel member performance

Conflicts are inherent, inevitable and are widely spread out in channel systems due to the constant interaction and interdependence of its members in a search for mutual objectives (Singh, 2006; Koza and Dant, 2006, 2007) In previous research, conflicts are beneficial for the reinforcement and the upgrade of joint business actions (Coughlan et al., 2006) On the other hand, conflicts are defined

as a situation that occurs when a channel members has the perception that the actions of another one show a behavior in preventing the accomplishment of its goals or the effective performance of it (Stern et al., 1998; Rosenbloom, 2003)

Channel conflict occurs when channel coalition perceives that another is engaged

in behaviors that impede the first entity from achieving its goals (Frazier, 1983) Channel conflict is probably the most serious concern for members that depend on

“channeling” for its success Members are attempting to re-construct their distribution systems and develop a more efficient process that will disrupt traditional channels All this is implemented resulting in conflict both internally among channel members and managers and externally with distribution partners Steinfield et al (2002) claims that the avoidance of channel conflict critical for click-and-mortar companies if they are to achieve the possible benefits of multiple marketing channels

According to Brown and Day (1981), there is a negative correlation between the conflict and the satisfaction of a member of a channel with the performance of another member The level of satisfaction of an intermediate member of a distribution channel varies with the level of existing conflict; concluding that without a conflict incidence, the satisfaction level is low With a level of natural conflict, the satisfaction level is higher than when there is none When a conflict level is perceivable and obvious, the satisfaction level diminishes significantly (Dant and Schul, 1992) Conflicting objectives by the various channel members can lead to internal conflict over customers, raise the potential for customers’ confusion and dissatisfaction In addition, conflict implies a level of tension, frustration, and disagreement in the relationship due to one member that obstructs the other member in reaching its goal (Geyskens et al., 1999) Although channel conflict can be functional (Coughlan et al., 2006, Anderson et al., 1990, Stern and

El-Ansary, 1998), high level of dysfunctional conflict will decrease the relationships between channel members

Therefore, high channel conflict between agencies and Vietnam Airlines may lead

to reduced channel member performance Our last hypothesis is:

H4: There is a negative relationship between channel confict and

channel member performance

2.6 Research Model

Channel

Performance

Enablers

Firgure 2.1: Conceptual framework

In summary, this chapter reviews some literature about the definittion of channel member enablers, channel conflict and channel member performance from previous studies In addition, it considers some arguments and gives the

H1 (+)

H4 (-) H3 (+) H2 (+)

theoretical hypotheses for this paper The next chapter will present research methodology for this study

CHAPTER 3

RESEARCH METHODOLOGY

This chapter depicts how the research will be conducted It describes the research process, data collection, data source, measurement of variables, methods of data analysis

3.1 Research process

According to Hair at el (2001), the research process includes three steps which are presented in Figure 3.1

Item generation

Questionnaire

In-depth Interviews

Final Draft

Quantitative pilot study

Questionnaire

Quantitative main study

Official Survey

Reliability Analysis

Deleted 2 items

Exploratory Factor Analysis

Regression Analysis

Conclusion & Limitation

Figure 3.1: Research process

3.1.1 Step 1: Item generation step

This step is based on the literature review chapter and gives the draft questionnaire with the measurement scales Then it would be followed by conducting interviews using an in-depth interview method (about six people)

These respondents are managers working at different Vietnam Airlines’ agencies including Hoang Giang Viet, Thien Hai, Skyworld, Viet Travel, Gia Tien, Nam Phan These six individuals would provide their ideas about the draft questionnaire After that, the questionnaire would be revised to develop the final draft questionnaire

This process is conducted as follows:

- Direct interviews were conducted with six senior managers who had ten to twenty years of experience in working with Vietnam Airlines as agencies

In airline agencies, the managers who have more than ten years of experience in working with Vietnam Airlines are qualified to judge the questionnaire relating to the relationships between them and Vietnam Airlines In addition, the questionnaire was also sent to an expert who studies management The purpose of this interview is to get some expert advice for any needed adjustments and also their judgment on whether the questionnaire is viable, accurate (or not) within the context of Vietnam

- All of the managers and experts supported the accuracy and compatibility

of the questionnaire The questionnaire was adjusted based on the ideas of these managers and experts in order to be suitable in the context of Vietnam

As a result of the above interviews, the total number of items used for designing the first draft of the questionnaire is 29 items The English version questionnaire

is adapted from previous research (Dwyer and Oh, 1987, Kumar et al., 1995; Lumineau and Oxley, 2007; Sezen and Yilmaz, 2007) The English questionnaire

was translated into Vietnamese for distribution An English teacher with twelve years of experience has helped to check the translation to assure the meaning equivalence of these two versions

3.1.2 Step 2: Pilot study

The purpose of the pilot survey is to preliminarily test the viability of the questionnaire and to see if there are any needed modifications or improvements

In this step, the process is interviewing the actual research subjects in the research study The purpose of these interviews is not to collect the data; its purpose is to collect the ideas of the experts about the accuracy, the clarity, and the understandability of the questionnaire Then, the final questionnaire will be accepted based on support ideas from these experts

3.1.3 Step 3: Main study

This step collects quantitative data in the form of a question survey It uses scales

of measurement already developed and validated in regards to trust (Dwyer and

Oh, 1987), commitment (Kumar at el., 1995), co-operation (Lumineau and Oxley, 2007), channel conflict (Lumineau and Oxley, 2007) and channel member performance (Sezen and Yilmaz, 2007)

A quantitative approach will be adopted because it is highly structured Furthermore, the measuring of trust, commitment, co-operation and channel conflict can determine which of these directly affects channel member performance Quantitative research is associated with exploring connections between variables Quantitative research is appropriate to use under a positivist

paradigm and natural science approach, which helps explain the links between theory and empirical research (Bryman and Bell, 2007)

The last step is to test the hypothesis of the study as well as explore the relationships between variables in the model utilizing multiple regression analysis The multiple regression analysis will be performed to evaluate the relationships between four independent variables consisting of “trust”,

“commitment”, “cooperation” and “channel conflict” “Channel member performance” is the dependent variable

3.2 Data collection

3.2.1 Sample size

This study applies the Exploratory Factor Analysis (EFA) to test the validity and the multiple regression analysis to test model and hypotheses So in each case, there needs to be a sufficient enough sample size to analyse

- For factor analysis, Bollen (1989) gives the way to identify sample “n” using this formula:

n > 5 * number of questions in questionnaires

This study includes 29 questions, so the minimum sample “n” for factor analysis

This study has 4 independent variables, so the minimum sample “n” for multiple regression is

The samples were selected using simple convenience sampling techniques (Nguyen, 2012) In other words, the samples were selected via the relationships between Vietnam Airlines’ agencies to collect samples from them

In brief, based on the theories and on previous research, the survey questionnaire includes 29 items divided into five factors: trust, commitment, cooperation,

channel conflict, channel member performance All of them are shown in Table 2.1

Table 2.1: Factors and items

Although times change and the future may be somewhat

uncertain, you know Vietnam Airlines will support your

company

5

Vietnam Airlines’ behavior is very consistent

6

Vietnam Airlines follows through on their promises

7 You can expect Vietnam Airlines to tell you the truth

Your agency is committed to the preservation of good working

relationships with Vietnam Airlines

Kumar et al (1995)

2

Your relationship with Vietnam Airlines could be described as

one of high commitment

3

In your agency, you expect to have a long term relationship with

Vietnam Airlines

C Cooperation

1

Your agency’s relationship with Vietnam Airlines is better

described as a "cooperative effort" rather than an "arm's length

negotiation"

Lumineau and Oxley

(2007)

2

Overall, your agency and Vietnam Airlines perform well together

in carrying out their respective tasks

3

Overall, the future of your relationship with Vietnam Airlines

promises to be beneficial to both of you

4

You believe Vietnam Airlines recognizes the effort your agency

puts into selling its products and supports you for it

2 Your discussions in areas of disagreement with Vietnam Airlines

create more problems than they solve

Vietnam Airlines’ policies make things difficult for you

E Channel Member Performance

1

The performance of the entire channel has improved as a result of

your relationship with Vietnam Airlines

Sezen and Yilmaz (2007)

2

Your agency encourages open and honest dialogue on all issues

3

The management of your agency motivates staff to set and

achieve high performance standards

4

The management of your agency actively seeks a win/win

solution on cross-functional issues

Your agency is committed to customer satisfaction

3.5 Methods of data analysis

3.5.1 Reliability analysis measured by Cronbach’s Alpha

To test the reliabilty of measuring scales, the Cronbach’s Alpha coefficient is used

to evaluate the data The Cronbach’s Alpha of each factor is equal or greater than 0.7 is reliable while any item has the corrected item – total correlation value is less than 0.3 will be deleted (Leech et al., 2005)

3.5.2 Validity measured by Exploratory Factor Analysis (EFA)

Factor analysis is an advanced statistical technique used to explore the underlying structure of relationships among a large number of variables (factors) by creating groups of variables (factors) having high inter-relationships with another These groups are assumed to represent dimensions within the data The application of factor analysis will help not only to summarize a large list of several original variables into a smaller set of explainable components with minimum loss of information but also to proceed with other multivariate techniques (such as regression analysis) in a much simpler and easier-to-explain pattern Statistical Package Software for Social Science (SPSS) program will be used to run the factor analysis and the analysis process will be based on the guideline of Hair et

al (2001) According to Hair, et al (2001), the value of the Bartlett Test of Sphericity and measure of Kaiser-Meyer-Olkin (KMO) index could be used to determine whether the data set is appropriate for factor analysis The Bartlett Test

of Sphericity is the test for the presence of correlations among these variables The result of the test with a significance level of less than 5% will confirm that there are significant correlations among at least some of the variables in the correlation matrix at 5% significance level If the correlations among the variables (factors) show them to be low or equal, the data set may not be appropriate for factor analysis since there is no indication of any structure for variables (factors) grouping The Kaiser-Meyer-Olkin (KMO) Measure of Sampling Adequacy (MSA) is used to quantify the degree of inter-correlations among the variables The KMO index of data set should not be less than 0.5

The Varimax Method, developed by Kaiser (1958), is selected for interpreting the components (groups) After the Varimax rotation, factors with high inter-correlation will cluster together under the same group Any item with factor loading, which is the correlation of the factor and its group, less than 0.5 will be dropped out of the groups Cross-loadings are not accepted Groups with Eigenvalue, which is Latent Root Criterion, greater than 1.00 are considered significant and able to explain an important amount of the data’s variability Groups with Eigenvalue less than 1.0 are insignificant and should be disregarded Total of cumulative variance explained by the number of groups selected should

be more than 50% (Gerbing and Anderson, 1988)

3.5.3 Multiple regression analysis

The multiple regression analysis will be used to: provide information on the accuracy about predictions; test how well the regression model fits the collected data; determine the variation in the dependent variables; explain the independent variables; and test the hypotheses on the regression equation According to Leech

et al., (2005), the multiple regression analysis process includes the following steps:

a) Testing the fit level of model

- Assumption 1: No significant outliers

Outlier: Standardized Residual is greater than ±3 SDs

- Assumption 2: independence of residuals (errors)

The Durbin – Watson value is greater than 0 and less than 4 is good

R is the multiple correlation coefficient R can range in value from 0

to 1, with higher values indicating the the predicted values are more closely correlated to the dependent variable (i.e the greater the value

of R, the better the independent variables are at predicting the dependent variable)

R2 is the coefficient of determination, present the proportion of variance in the dependent vaiable that can be explained by the independent variables

F test: tests whether the regression model is a good fit for the data

- Assumption 3: a linear relationship between the predictor variables and the dependent variable

- Assumption 4: homoscedasticity of residuals (equal error variances)

- Assumption 5: No multicollinearity: inspect the correlation coefficient

If the correlation between independent variables is less than 0.7, it has no multicollinearity

- Assumption 6: residuals (errors) are normally distributed

b) Testing regression coefficient

In the “Coefficients” table, if any dependent variable has Sig less than 0.05, it has significant correlation with dependent variable and otherwise, if any independent variable has Sig greater than 0.05, it has no significant correlation with dependent variable