INTRODUCTION
Background
The telecommunications service sector has gained significant economic importance, prompting researchers to focus their studies on this field Since the 1990s, the industry has emerged as a crucial driver of economic development in numerous countries, fueled by rapid technological advancements and a growing number of network operators competing vigorously.
As competition intensifies in the market, firms increasingly prioritize customer retention to maintain their market share, particularly in the telecommunications sector With rising costs associated with acquiring new customers, companies are focusing their strategic efforts on fostering long-term relationships with existing clients Once customers are connected to a telecommunications network, their ongoing loyalty becomes crucial for the company's success, highlighting the importance of effective customer retention strategies.
Customer retention is essential in the mobile service industry, as operators face annual subscriber losses exceeding 30% and significant customer acquisition costs To address this challenge, mobile operators must implement well-designed programs aimed at enhancing customer retention (Lee, 2001).
Numerous studies have been conducted on the critical issue of customer retention, leading to the development and estimation of various dynamic models that explain and measure the determinants and influencing factors of this important aspect of business.
Problem definition and research questions
The telecommunications industry is currently grappling with heightened national and international competition, a slowing growth rate, and a saturated market Consequently, an increasing number of service providers are competing for a limited pool of new customers In this challenging landscape, companies must allocate significant resources to enhance customer satisfaction and retention among their existing clientele.
Customer retention is crucial for a company's survival and future growth To sustain stable profits, especially when subscription levels have saturated, a defensive strategy focused on retaining existing customers becomes more vital than an aggressive approach aimed at attracting new ones.
Numerous studies indicate a connection between customer satisfaction and retention; however, satisfaction alone does not fully account for customer retention, as customers often face constraints when selecting suppliers These constraints, referred to as switching barriers, alongside customer satisfaction, significantly influence supplier choice (Fornell, 1992) Limited empirical research has explored the impact of different switching barriers on supplier satisfaction and customer retention It has been noted that customers remain loyal to a supplier either out of desire or necessity (Ping, 1993) High switching barriers compel customers to stay with suppliers, regardless of their satisfaction levels.
The research questions that are discussed in this thesis are as below:
Question 1: Can switching barriers be separated into negative and positive factors?
Question 2: How do positive switching barriers have impact to customer satisfaction and customer retention in mobile telecommunication service in Vietnam?
Question 3: How do negative switching barriers have impact to customer satisfaction and customer retention in mobile telecommunication service in Vietnam?
Research purpose
This thesis aims to develop a model for assessing customer retention in the Vietnamese mobile service market, identifying key influencers that can assist service providers in enhancing their customer retention rates.
The research model identifies customer satisfaction, positive switching barriers, and negative switching barriers as key determinants of customer retention Unlike traditional approaches, this study will not treat switching barriers as a singular factor influencing customer satisfaction or retention, nor will it explore their mediating role in the relationship between these two elements Instead, both the positive and negative effects of switching barriers will be analyzed within the same framework, allowing for a distinct examination of their individual impacts on customer satisfaction and retention.
Research limitation
This thesis serves as a foundational framework for future research in various service sectors, although it does have limitations Notably, it does not address other influential factors on customer retention, such as demographic characteristics and customer usage patterns of mobile services, which remain unexplored in this study.
This thesis is structured into five chapters, beginning with an overview of the significance of the research area and the introduction of key research questions Chapter Two reviews literature on customer satisfaction and retention, exploring the relationship between these concepts, the classification of switching barriers into positive and negative factors, and their impact on customer satisfaction and retention This chapter also presents the proposed model and hypotheses Chapter Three details the research methodology, including the pilot testing phase Chapter Four analyzes the sample characteristics, key constructs measurement, and the results of the path analysis for the proposed model Finally, Chapter Five discusses the implications of the study for management practices, theoretical contributions, and directions for future research.
Chapter 4: Data analysis and findings
Organization of the thesis
2.1.1 Defining customers and customer purchasing process
The term "customer" typically refers to end-users of a product or anyone receiving a service, encompassing both internal and external customers Internal customers are employees within an organization, while external customers include stakeholders such as clients, consumers, and constituents Identifying the specific types of customers surveyed is crucial for accurately analyzing customer satisfaction results.
In this study, the concerned customers are the individual consumers/users who subscribe mobile services of any operator in Vietnam
Researchers indicate that customers typically navigate a five-stage decision-making process when making purchases, which includes need recognition, information search, evaluation of alternatives, the purchase itself, and post-purchase evaluation (Kotler and Keller, 2006) In the service sector, Lovelock and Wirtz (2007) propose a three-stage model of service consumption that encompasses the pre-purchase service, the encounter stage, and the post-encounter stage.
The pre-purchase stage consists of three key components: first, need awareness and information search, where customers clarify their needs; second, exploring solutions and identifying potential suppliers and alternative service products; and finally, evaluating these alternatives to make an informed decision on the service purchase This stage is influenced by various factors.
LITERATURE REVIEW
Customer satisfaction
2.1.1 Defining customers and customer purchasing process
The term "customer" typically describes end-users of a product or anyone receiving a service, encompassing both internal and external customers Internal customers include staff or employees, while external customers consist of stakeholders such as clients, consumers, and constituents Identifying the specific types of customers surveyed is crucial for accurately analyzing customer satisfaction results.
In this study, the concerned customers are the individual consumers/users who subscribe mobile services of any operator in Vietnam
Researchers indicate that customers typically navigate a five-stage decision-making process during purchases, which includes need recognition, information search, evaluation of alternatives, purchase, and post-purchase evaluation (Kotler and Keller, 2006) In the service sector, Lovelock and Wirtz (2007) introduced a three-stage model of service consumption, comprising pre-purchase service, the encounter stage, and the post-encounter stage.
The pre-purchase stage consists of three key components: first, need awareness and information search, where consumers clarify their needs; second, the exploration of solutions, identifying potential suppliers and alternative service products; and finally, the evaluation of these alternatives to make an informed decision on the service purchase This stage is influenced by the consumer's search for specific service attributes, as well as their perceived risks and expectations.
The service encounter is a critical stage where customers make requests to their selected supplier, with payment processed at a later time This phase encompasses both service delivery by personnel and self-service options, marking a pivotal moment as customers have already begun to experience the service.
The post-encounter stage is crucial as it assesses the service performance and its impact on future customer intentions During this phase, customers experience satisfaction or dissatisfaction, which ultimately influences their decision to remain loyal to the brand.
In the mobile telecom market, purchasing and activating a SIM card instantly makes a customer a subscriber, highlighting the importance of the decision-making process in influencing customer satisfaction, repurchase intentions, and recommendations to others This ongoing value exchange occurs as customers utilize services and engage in the production and delivery processes Ultimately, a customer's loyalty to a network is determined by several factors, with the quality of service and their satisfaction level being among the most critical influences on their decision to stay or switch providers.
Customer satisfaction has become a focal point for both scholars and practitioners, as it plays a crucial role in business strategy and is essential for achieving success in today's competitive market.
1994) It is therefore important to understand this terminology in detail as conceptualized in this study Some of the definitions given by scholars for customer satisfaction are as follows:
Customer satisfaction is a psychological concept that reflects the sense of well-being and pleasure derived from receiving products or services that meet one's expectations and desires.
• Customer satisfaction is “as an attitude-like judgment followed by a purchase act or a series of consumer product interactions.” Youjae Yi, (1990 cited in
• Customer satisfaction is ‘‘a consumer’s post-purchase evaluation and affective response to the overall product or service experience.’’ (cited from
• Satisfaction is “merely the result of things not going wrong; satisfying the needs and desires of consumers.’’(cited from Besterfield 1994);
Customer satisfaction is defined as an experience-based evaluation by the customer regarding the extent to which their expectations about the specific features or overall performance of the services received from a provider have been met (Bruhn, 2003).
• ‘‘Satisfaction is a person’s feeling of pleasure or disappointment resulting from comparing a product’s performance (outcome) in relation to his or her expectation.’’ (cited from Kotler P & Kevin L K., 2006 p 144)
Customer satisfaction as a Process and an Outcome:
The debate surrounding customer satisfaction centers on whether it should be viewed as an outcome or a process While early definitions leaned towards satisfaction as a process, this perspective remains predominant among scholars today (Oliver, 1980; Parasuraman et al., 1991) According to the process view, customer satisfaction arises from the emotional response triggered by comparing perceived performance against established standards, such as expectations or desires (Khalifa & Liu, 2002).
According to Richard Oliver (1980), customer satisfaction occurs when the performance of a product or service meets expectations, leading to positive disconfirmation Conversely, dissatisfaction arises when performance falls short of expectations, resulting in negative disconfirmation When expectations surpass perceived performance, customers experience high satisfaction This perspective emphasizes satisfaction as a process, focusing on the factors that lead to satisfaction, particularly those that arise during the service delivery process (Vavra, 1997).
Recent studies define satisfaction as a post-purchase experience and an outcome of the service consumption process (Vavra, 1997) This perspective is grounded in motivation theories, suggesting that consumer behavior is aimed at achieving specific goals (Vroom, 1964) Consequently, satisfaction is seen as a key goal that consumers strive to attain.
In this study, we define customer satisfaction from a process perspective, emphasizing that in Vietnam's mobile telecom market, customers assess mobile services primarily during the service delivery process This evaluation is ongoing and extends beyond a mere outcome that customers seek to achieve.
Customer satisfaction as Cognitive and Affective responses:
The debate surrounding customer satisfaction centers on whether it is primarily cognitive or affective Many scholars view satisfaction as a process, yet its true nature remains ambiguous Some argue that satisfaction involves a cognitive assessment, where customers compare a provider's offerings against their expectations Conversely, others contend that satisfaction reflects an emotional state shaped by service delivery experiences that influence customer emotions Recent research suggests that satisfaction encompasses both cognitive and affective dimensions, highlighting the complexity of customer experiences (Oliver, R.L., 1993a; Gronroos, C.).
2001) They argue that “ satisfaction is naturally tied to cognitive judgments and to affective reactions elicited in consumption” (Mano and Oliver, 1993, p 451)
This study conceptualizes customer satisfaction as both cognitive and affective since we believe customers express their satisfaction with the service quality they consumed in both cognitive and emotional way
Customer satisfaction as Transactional or Cumulative:
The debate among scholars regarding customer satisfaction centers on whether it should be viewed as cumulative or transactional From a transactional standpoint, customer satisfaction is assessed based on a single post-purchase evaluation of a service encounter Conversely, the cumulative perspective defines customer satisfaction as an overall evaluation of a product or service, taking into account various purchase and consumption experiences over time This cumulative approach offers greater diagnostic and predictive value, as it reflects a series of interactions rather than just one transaction Consequently, this study adopts a cumulative framework for understanding customer satisfaction.
Customer retention
Competition in Vietnam's mobile telecom market is highly intense, having reached saturation for some time Consequently, the focus has shifted from acquiring new customers to retaining existing ones.
The pressure to reduce customer churn, together with the high cost of acquiring new customer, have forced mobile service providers to take a hard look at their customer retention strategies
Customer retention has become a pivotal focus in marketing, shifting from mere customer satisfaction to a strategic priority in today's competitive landscape Research indicates that understanding customers' willingness to engage and maintain relationships with service providers is essential for long-term loyalty As highlighted by Oliver (1999), fostering retention is now regarded as the foremost goal for businesses aiming to thrive in a challenging market.
To enhance customer loyalty and retention, companies must go beyond traditional metrics like satisfaction and defection, particularly in the competitive telecommunications sector Research indicates that once customers are connected to a telecommunications network, their long-term commitment to a specific operator is crucial for the company's success, often more so than in other industries (Gerpott et al., 2001, p 249).
Customer retention focuses on sustaining the relationship between a service provider and their customers This can be accomplished through two main strategies: encouraging repeat purchases or extending contracts over time Additionally, it involves fostering customer intent for future purchases and minimizing the likelihood of contract termination.
Business relationships can be maintained involuntarily due to mobility barriers, which vary based on each customer's perception and circumstances regarding switching suppliers or discontinuing a service category (Bliemel & Eggert, 1998) Conversely, a customer may choose to continue transactions out of a favorable attitude towards the provider and the services offered, aiming to sustain the relationship for mutual benefit This scenario is referred to as customer loyalty, where the relationship is pursued for shared advantages, while merely staying with a supplier without such intent is termed customer retention (Homburg & Bruhn, 1998) Consequently, while customer loyalty and retention are closely linked in terms of cause and effect, they represent distinct concepts.
As mobile service operators face peak subscriber levels, acquiring new customers has become increasingly challenging and expensive Consequently, the prevailing industry perspective emphasizes that the most effective marketing strategy moving forward is to focus on retaining existing customers by enhancing customer loyalty and value (Kim et al., 2004, p 146).
Building long-term business relationships leads to significant cost savings by reducing expenses associated with attracting new customers, such as advertising, personal selling, and account setup Additionally, these relationships minimize the costs of educating new clients about business procedures and streamline interactions during the customer learning process (Peppers and Rogers, 1995; Reichheld, 2003).
Customer retention offers a dual benefit for businesses, enhancing revenue through increased sales while simultaneously reducing costs through various savings strategies As a result, effective customer retention has emerged as a crucial element for achieving long-term business success (Rust and Zahorik, 1993).
While businesses acknowledge the importance of customer retention, a clear strategy for boosting loyalty remains elusive Many companies gauge customer satisfaction, assuming that high satisfaction scores will lead to customer loyalty However, satisfied customers may still switch to competitors, highlighting the need for more effective loyalty strategies.
Customer retention, the ability to foster loyalty among consumers for repeated purchases, is increasingly vital for businesses According to Kotler (2000), acquiring new customers is five times more expensive than retaining existing ones, highlighting the importance of maintaining a strong customer base.
Customer retention is crucial for a company's ongoing success, as loyal customers contribute significantly by increasing purchases, paying premium prices, and offering positive referrals through word of mouth Research shows that businesses, particularly in the telecommunications sector, face a monthly loss of 2-4% of their customers, which can translate to millions in lost revenue and profit Therefore, fostering customer loyalty is essential for sustainable growth.
Reichheld and Sasser's studies emphasize the importance of customer lifetime value and retention by addressing customer complaints, preventing churn, and understanding competitive migration Customers stay loyal primarily due to the value they derive from their suppliers Therefore, effective customer retention involves maintaining an ongoing, active relationship with clients (Cannie, 1994).
Understanding customer attrition is crucial for businesses, as an increase in lost customers directly impacts revenue Retaining existing customers is significantly more cost-effective than acquiring new ones, especially in today's highly competitive market Therefore, companies must prioritize strategies that encourage customer loyalty and minimize churn.
Both suppliers and customers can significantly benefit from retention in their relationship While suppliers gain from fostering a loyal customer base, customers also enjoy the advantages of long-term associations, creating a mutually beneficial dynamic.
Increasing customer retention by just 5% can lead to profit increases of up to 85%, as strong customer relationships reduce operating costs and enhance the effectiveness of marketing efforts When customers are familiar with a company and its employees, they tend to ask fewer questions and experience fewer issues, further lowering costs Additionally, satisfied customers contribute to profits through referrals, as they share positive experiences with friends and family, reinforcing their own choices and driving new business.
Switching barriers
Earlier studies of factors affecting customer retention usually pay their attention to customer satisfaction, the switching barriers and relationship between them (Dick & Basu, 1994; Gerpott, Rams, & Schindler, 2001; Lee & Cunningham, 2001)
Switching barriers serve as an adjustment factor in the relationship between customer satisfaction and retention, indicating that varying levels of customer retention can occur even with consistent customer satisfaction, depending on the strength of these barriers (Colgate & Lang, 2001; Jones et al., 2002; Lee & Cunningham, 2001; Kim, 2000) Kim's research in the Korean mobile service market identified key elements of switching barriers, including switching costs, interpersonal relationships, the attractiveness of alternatives, and service recovery He concluded that these barriers impact customer retention both directly and as a mediator between customer satisfaction and retention in the mobile service sector.
Figure 2.3 Affecting role of switching barriers, both mediating and direct to customer retention
The switching barrier refers to the challenges and burdens—financial, social, and psychological—that customers perceive when considering a transition to a new service provider.
Yang (2004) identified that switching barriers significantly mediate the relationship between customer value and loyalty, as well as between perceived satisfaction and loyalty Key factors influencing the connection between customer satisfaction and loyalty include market regulation, switching costs, brand equity, existing loyalty programs, and product differentiation from competitors Numerous authors, including Fornell (1992), Jones et al (2000), and Julander and Soderlund (2003), have recognized these factors as essential switching barriers.
(2003) confirmed that firms might retain their customers by creating switching barriers that should add value to their services
Figure 2.4 Mediating role of switching cost to both customer value – customer loyalty and perceived satisfaction – customer loyalty linkage
A switching barrier, as defined by Jones et al (2000), refers to factors that make it difficult or expensive for customers to change providers For companies with a significant customer base, it is crucial to comprehend the reasons behind customer retention and to identify strategies that can either encourage or discourage switching Additionally, new market entrants must understand why customers choose not to switch in order to develop effective strategies that can overcome these barriers and capture market share (Colgate and Lang, 2001).
Barriers to customer defection, such as fostering strong interpersonal relationships and establishing switching costs, serve as effective retention strategies These barriers not only promote customer loyalty but also enable companies to retain customers even after occasional service disappointments Without effective retention strategies, customers are likely to leave, as there would be no compelling reason to stay or significant obstacles to deter them from switching.
Switching costs refer to the expenses incurred by customers when changing suppliers, encompassing time, money, and psychological factors (Dick & Basu, 1994) These costs represent the perceived risks associated with switching, which include potential financial, social, and psychological losses that customers might face if they choose to transition to a different provider (Murray, 1991).
Julander and Soderberg (2003) highlight that switching barriers can have both positive and negative implications for customers According to Hirschman (1970), positive switching barriers reflect a customer's desire to maintain a relationship, while negative switching barriers indicate a sense of obligation to remain in that relationship.
Positive switching barriers arise from the strong interpersonal relationships that suppliers cultivate with their customers throughout the lifecycle of a product or service, as noted by researchers such as Berry and Parasuraman (1991) and Tumball and Willson.
Interpersonal relationships refer to the psychological and social connections between customers and suppliers, encompassing elements like care, trust, communication, and intimacy (Gremler, 1995) These relationships, developed through repeated interactions, can enhance the bond between suppliers and customers, ultimately fostering long-term partnerships Both companies and customers often seek to establish enduring relationships, highlighting the mutual desire for sustained engagement in business interactions.
Customers increasingly seek to establish and nurture valuable interpersonal relationships with companies (Gwinner, Gremler, & Bitner, 1998) By making relationship-specific investments, businesses can enhance customer dependence and raise switching barriers, ultimately fostering loyalty (Jones, Mothersbaugh, & Betty, 2000).
Modern relationship-specific activities provide customers with a range of benefits, including social perks like fellowship and personal recognition, economic advantages such as discounts and time savings, psychological support that helps reduce anxiety, and opportunities for personal customization Research by Gwinner et al (1998) indicates that customers are willing to invest in and nurture relationships with suppliers that offer superior value benefits, tailored to their financial situations.
Research indicates that switching costs significantly influence customer retention in the mobile service industry Deregulation and globalization have led to heightened competition among mobile operators, resulting in decreased traditional switching costs, which complicates customer retention efforts Conversely, the evolving networked environment has introduced new switching costs that mobile operators can leverage to enhance customer retention strategies.
Switching costs are categorized into three main types: loss cost, adaptation cost, and move-in cost Loss cost reflects the perceived loss of social status or performance associated with terminating a service contract with an existing provider Adaptation cost encompasses the perceived expenses related to adjusting to new suppliers, including search and learning costs Lastly, move-in cost pertains to the economic expenses incurred when transitioning to a new carrier, such as purchasing a new device and paying subscriber fees.
Switching barriers are influenced by customer perceptions of the availability of viable competing alternatives in the marketplace Research indicates that when suitable alternatives are scarce, the likelihood of customers ending their relationship with their current supplier diminishes Conversely, when customers recognize the existence of more appealing options, they are more inclined to make a switch.
Positive and negative switching barriers
As mentioned in previous section, Hirschman (1970) makes the distinction between
In relationships, the distinction between 'having to be' and 'wanting to be' is crucial, with the former representing a negative reason for staying and the latter a positive one According to Jones et al (2000), switching barriers can be categorized into positive and negative elements From both theoretical and managerial viewpoints, it is essential to differentiate between positive switching barriers, which reflect a customer's desire to remain in a relationship, and negative switching barriers, which indicate a customer's reluctance to leave a supplier Psychologically, the motivation behind continued use of a service or product—whether it stems from genuine desire or the high cost of leaving—significantly influences customer perception and satisfaction.
Julander (2003) categorized switching barriers into two types: negative and positive He emphasized that switching costs, which are the financial and logistical obstacles associated with changing suppliers, are a significant negative switching barrier High switching costs can effectively lock customers into their current suppliers Additionally, factors such as market monopoly and supplier power can further entrench customers to their suppliers Customer investments in the supplier relationship, including time, money, and effort, also serve as negative switching barriers, particularly when physical investments in equipment are involved Thus, substantial customer investments are classified as negative switching barriers that inhibit supplier switching.
Julander (2003) identifies positive switching barriers as the appeal of alternative options When customers perceive their chosen supplier as superior to other available alternatives, they are more likely to remain loyal, effectively becoming "locked in" to that supplier This loyalty is driven by a strong motivation to continue the relationship with their preferred supplier.
Positive interpersonal relationships serve as effective switching barriers, encouraging customers to remain loyal to suppliers These relationships are often integrated into the overall product or service experience Additionally, loyal customer discounts and established customer habits further reinforce these barriers, enhancing the attractiveness of the offered product or service package (Fornell, 1992).
Julander (2003) developed a model examining the interplay between customer satisfaction, switching barriers, and customer retention He found that customers facing negative switching barriers tend to experience lower satisfaction levels compared to those in less constrained situations However, these negative barriers lead to higher customer retention, as individuals feel compelled to remain with their current suppliers Conversely, customers with significant positive switching barriers report higher satisfaction, which indirectly enhances their intention to repurchase.
Proposed research model and hypotheses
This thesis examines the interrelationship between customer satisfaction, switching barriers—categorized into negative and positive factors—and customer retention To address the research questions outlined in Chapter 1, we focused on understanding these dynamics within the context of mobile service users in Vietnam, utilizing the model developed by Julander (2003) to assess its applicability and fit.
Figure 2.5 Research model used in this thesis to investigate relationship among
The following hypotheses were raised and tested with the selected research model
• H1: Negative switching barriers are negatively and directly associated with customer satisfaction
• H2: Negative switching barriers are positively and directly associated with customer retention
• H3: Positive switching barriers are positively and directly associated with customer satisfaction
• H4: Positive switching barriers are positively and indirectly associated with customer retention.
RESEARCH METHODOLOGY
Research purpose
The purpose of research serves as a comprehensive statement outlining the intended achievements of the study Based on this purpose, research can be categorized into three main types: exploratory, descriptive, and explanatory, as noted by Saunders et al (2000, 2007) and Cooper and Schindler.
Exploratory research is a study aimed at uncovering new insights and understanding phenomena from a fresh perspective, particularly in areas that are unclear or under-researched (Saunders et al., 2007) This type of research is essential when a researcher seeks to clarify a situation or problem where key variables may be undefined To achieve this, exploratory research employs various methods, including reviewing existing documentation, consulting experts, and conducting focus group interviews.
Descriptive research is a systematic study aimed at accurately depicting individuals, events, or situations, as defined by Saunders et al (2007) This type of research formalizes the investigation with structured methodologies to effectively describe and present factual information about a phenomenon, reflecting its true nature or perception.
Explanatory research is a type of study aimed at establishing the relationships between variables, focusing on how one variable influences another Its primary purpose is to explain the causes and effects associated with one or more variables, often referred to as causal studies This approach is particularly useful when the research seeks to answer the question of 'why' within a specific context.
This thesis is classified as explanatory research, as it utilizes a model previously developed and tested by other authors to explore the relationships between customer satisfaction, switching barriers, and customer retention.
Research Approach
Research can be conducted through two main approaches: deductive and inductive The deductive approach starts with a theory or hypothesis and involves designing a strategy to test its validity in a specific context, moving from general principles to specific instances In contrast, the inductive approach begins with observations of phenomena, leading to data collection and the development of a theory or generalization, thus moving from specific observations to broader conclusions.
This study employs a deductive approach to analyze the relationship between customer satisfaction, switching barriers, and customer retention within Vietnam's mobile service market By carefully selecting and applying established empirical theories and models, we aim to provide insights into these critical factors influencing consumer behavior in the telecommunications sector.
Sampling
When selecting an appropriate sample, researchers typically choose between two main methods: random (probability) sampling and non-random (non-probability) sampling Random sampling ensures that every member of the population has an equal chance of being selected, while non-probability sampling relies on criteria other than random selection, such as convenience or personal judgment (Zikmund, 2000).
The convenience sampling technique was used in this research since questionnaires were distributed over email and form available online to all people that we have relationship
A sample should be reliable and valid in order to enable us to generalize the findings from the sample to the population under investigation (Canava et al 2001)
To determine the appropriate sample size for exploratory factor analysis, Hair et al (1998) recommend a minimum of five samples for each item being analyzed Given that the research model includes 11 observed items, the minimum required sample size is 55.
To ensure reliable output in path analysis using Structural Equation Modeling (SEM), it is recommended to have a minimum of 15 samples per observed variable, as established by Hair et al (1998) Consequently, to satisfy the sampling size requirement, valid data from at least 165 respondents is necessary.
Data collection procedure
Scholars emphasize that survey strategies primarily utilize self-administered or interviewer-administered structured and unstructured interviews, along with questionnaires (Saunders et al., 2000; Cooper and Schindler, 2006) They agree that questionnaires are suitable for both descriptive and explanatory studies and should feature a well-organized layout, clear and relevant questions, comprehensive items, and a logical sequence This ensures respondents are willing to provide answers, leading to the decision to employ a questionnaire for the research.
This study utilized a self-administered, structured questionnaire to gather data from respondents across various age groups and relationship types The complete questionnaire was distributed via email to the author's contacts and made available online, encouraging additional feedback through an online form Data collection took place over four consecutive weeks in October and November 2011.
A total of 243 questionnaires were distributed, yielding 189 responses, of which 8 were deemed invalid This resulted in 181 valid questionnaires, achieving a response rate of 74.5%, which is considered satisfactory for further analysis Additionally, this number exceeds the minimum required sample size of 165.
Measurement
In our study, we evaluated all variables using multi-item scales based on a 5-point Likert scale, where 1 represents "strongly disagree" and 5 signifies "strongly agree." The constructs were developed following the foundational research of Julander (2003), who established specific items to effectively measure each component of his model.
Customer satisfaction included 3 observed variables, coded respectively CS1, CS2 and CS3:
• CS1: I am satisfied with the supplier
• CS2: The supplier meets all the requirements that I see reasonable
• CS3: The supplier satisfies my need
Customer retention included 2 observed variables, coded respectively CR1 and
• CR1: I intend to continue to be customer of this supplier
• CR2: Next time I shall need services of the supplier, I will buy it from him
The study identified seven observed variables related to switching barriers, building on Julander's (2003) classification of these barriers into positive and negative factors This thesis specifically focuses on negative switching barriers, incorporating five observed variables, designated as NSB1 to NSB5.
• NSB1: There are few other suppliers that are realistic alternatives for me
• NSB2: It is difficult for me to use other suppliers
• NSB3: It would be complicated for me to change supplier
• NSB4: I feel locked to this supplier
• NSB5: It take a lot of time to get information about other suppliers
Positive switching barriers are measured by 2 observed variables and coded PSB1 and PSB2:
• PSB1: I feel uncertain about whether other suppliers can give the same service as this one
• PSB2: If I were to choose another supplier I do not know what I will get
Pilot testing
Saunders et al (2000) and Cooper and Schindler (2006) emphasize the importance of pre-testing questionnaires to ensure reliability and validity in research A preliminary draft of the questionnaire, available in both English and Vietnamese, was distributed to focus group members to evaluate the clarity and relevance of the questions, particularly in the context of translation Subsequently, a sample of fifteen mobile service subscribers was pre-tested using a convenient sampling method, aligning with Fink's (2003b in Saunders et al 2007) recommendation of a minimum sample size of ten for effective pre-testing Participants were informed about the purpose of the questionnaire and assured of their anonymity prior to completing it.
The fifteen respondents initially received the questionnaire via email, followed by a phone call from the authors to confirm receipt and encourage thorough review A subsequent call was made to discuss each question in detail, allowing us to gauge their understanding and make minor adjustments to some Vietnamese questions Additionally, we engaged another 30 respondents to assess the reliability of the measurement scales The results of the Cronbach alpha test indicated a high reliability for the Customer Satisfaction measurement scale (0.947), Customer Retention measurement scale (0.818), Negative Switching Barrier measurement scale (0.786), and Positive Switching Barrier measurement scale (0.885) as shown in Table 3.1.
In our analysis, we identified that the variable NSB5 needed to be removed to improve the Cronbach's alpha value for the 'switching barrier' measurement scale, which increased to 0.901 upon its deletion After excluding NSB5 during the pilot stage, we conducted the Cronbach alpha test again for switching barriers, resulting in a satisfactory outcome with no further items needing removal (see Appendix 2).
Table 3.1 Cronbach’s alpha result – Pilot stage
Scale Mean if Item Deleted
Scale Variance if Item Deleted
Cronbach's Alpha if Item Deleted Customer satisfaction, Cronbach’s alpha = 0.947
Negative switching barrier, Cronbach’s alpha = 0.786
Table 3.1 Cronbach’s alpha result – Pilot stage
Scale Mean if Item Deleted
Scale Variance if Item Deleted
Cronbach's Alpha if Item Deleted Customer satisfaction, Cronbach’s alpha = 0.947
Positive switching barrier, Cronbach’s alpha = 0.885
(*):The value is negative due to a negative average covariance among items (only 1 item left if deleted)
In the pilot stage, Explanatory Factor Analysis was employed to assess the validity of all defined constructs, resulting in the extraction of a single factor from each construct, thereby confirming the measurement scale's validity (Appendix 2).
Data analysis method
The reliability of each construct in the model was initially assessed using Cronbach's alpha Subsequently, an Explanatory Factor Analysis (EFA) was conducted to validate all constructs These analyses are essential prerequisites before testing the model's hypotheses through Structural Equation Modeling (SEM) using AMOS 16.
Cronbach's alpha is the standard metric for assessing the reliability and internal consistency of predetermined scales in research A higher alpha score indicates greater reliability of the measurement tool.
• Alpha is 0.9 or higher: high reliability
• Alpha is from 0.8 to 0.89: good reliability
• Alpha is from 0.7 to 0.79: acceptable reliability
• Alpha is from 0.65 to 0.69: marginal reliability
Factor analysis is defined as a collection of procedures aimed at analyzing the relationships among a set of observed variables within a group (Cureton & D'Agostino, 1983) Its primary purpose is to explain the inter-correlations among multiple variables by proposing a smaller number of common factors (Cureton & D'Agostino, 1983) According to Bryman and Cramer (1990), factor analysis encompasses various related statistical techniques that are essential for assessing the validity of a construct, or latent variable, as measured by a set of observed variables.
Structural Equation Modeling (SEM) is a very general, very powerful multivariate analysis technique that includes specialized versions of a number of other analysis methods as special cases
Major applications of SEM include:
Causal modeling, also known as path analysis, is a statistical technique that hypothesizes and tests causal relationships among variables using a system of linear equations This approach can incorporate both manifest and latent variables, allowing for a comprehensive understanding of the underlying causal structures.
• Confirmatory factor analysis, an extension of factor analysis in which specific hypotheses about the structure of the factor loadings and inter- correlations are tested;
• Second order factor analysis, a variation of factor analysis in which the correlation matrix of the common factors is itself factor analyzed to provide second order factors;
• Regression models, an extension of linear regression analysis in which regression weights may be constrained to be equal to each other, or to specified numerical values;
• Covariance structure models, which hypothesize that a covariance matrix has a particular form For example, you can test the hypothesis that a set of variables all have equal variances with this procedure;
• Correlation structure models, which hypothesize that a correlation matrix has a particular form A classic example is the hypothesis that the correlation matrix has the structure of a circumplex (Guttman, 1954)
This thesis utilized Structural Equation Modeling (SEM) to evaluate the applicability of Julander's (2003) model in the context of Vietnam's customer and market conditions, while also testing the hypotheses outlined in Chapter 2.
DATA ANALYSIS AND FINDINGS
The questionnaire
There were 243 questionnaires distributed and we received 189 feedbacks from respondents, among of them 8 were then identified as invalid So in total, we had
181 valid returned questionnaires which contributed a response rate of 74.5%.
Descriptive result
There were some personal, general and informative questions in the questionnaires that their statistical results are briefly discusses and visualized here
From the result, it can be stated that 42.5% of respondents were male and 57.5% of them were female (Appendix 2) Regarding their age, majority of them were from
26 to 35 (60.8%) The results for the other age groups are shown in the following Table 4.1
Table 4.1 Age statistics of respondents
The study revealed that a significant majority of respondents, 85.1%, held a bachelor's degree (Table 4.2) This finding is influenced by the convenience sampling method employed, as the participants were primarily individuals with direct or indirect connections to the author, which may not accurately represent the overall mobile user population.
Table 4.2 Education Background statistics of respondents
Valid High School or less 12 6.6 6.6 6.6
The majority of respondents are employed staff, comprising 75.7%, while a smaller percentage includes students at 6.6% and managers at 13.3%, as shown in Table 4.3 The variation among these groups can largely be attributed to the sampling method utilized in this thesis.
Table 4.3 Occupation statistics of respondents
The survey reveals that a significant portion of respondents, specifically 44.8%, earn between 5-10 million VND monthly, while 30.9% earn between 10-20 million VND This income distribution is largely attributed to the fact that most respondents are employees based in Ho Chi Minh City.
Table 4.4 Monthly income statistics of respondents
Valid Lower 5 million VND/month 30 16.6 16.6 16.6
Table 4.5 Living city/town Statistics of respondents
Valid Ho Chi Minh city 143 79.0 79.0 79.0
In Vietnam, but not in cities 16 8.8 8.8 100.0
The survey included respondents from nearly all major mobile service providers in Vietnam, with significant representation from leading operators Notably, Vinaphone accounted for 35.4% of the subscribers, followed by Mobifone at 23.2% and Viettel at 18.8%, as illustrated in Table 4.6.
Table 4.6 Mobile network statistics of respondents
Accessing reliability and validity of collected data
We utilized Cronbach’s alpha to evaluate the reliability of data gathered from 181 valid respondents, deeming measurement scales and data valid if the alpha value is above 0.6 (Tho et al., 2009) Furthermore, any observed variable with a higher alpha value when excluded should be eliminated from the model.
Table 4.7 Cronbach’s alpha value – Main research stage
Scale Mean if Item Deleted
Scale Variance if Item Deleted
Cronbach's Alpha if Item Deleted Customer satisfaction, Cronbach’s alpha = 0.934
Negative switching barrier, Cronbach’s alpha = 0.961
Positive switching barrier, Cronbach’s alpha = 0.961
(*):The value is negative due to a negative average covariance among items (only
The Cronbach’s alpha values for all constructs exceeded 0.6, and each item demonstrated a corrected item-total correlation greater than 0.5, confirming the reliability of the measurement scales and the collected data (Trong et al., 2008) Consequently, there was no need to remove or modify any observed variables from the questionnaire, which was developed following the pilot stage.
In the next phase, we conducted Explanatory Factor Analysis (EFA) to eliminate unsuitable items and validate the collected data prior to model construction and hypothesis testing The EFA results are deemed acceptable when they pass the KMO (Kaiser-Meyer-Olkin) and Bartlett tests; specifically, the KMO value should exceed 0.5, ideally approaching 1, while the Bartlett significance value must be less than 0.05.
Table 4.8 Table KMO and Bartlett's Test
Kaiser-Meyer-Olkin Measure of Sampling Adequacy .804
Bartlett's Test of Sphericity Approx Chi-Square 2182.671 df 55.000
We had KMO value 0.804 and Barlett significant value 0 (Table 4.8) which proved our data was good
In Exploratory Factor Analysis (EFA), a measurement item is considered valid if its factor loading on a single specific factor exceeds 0.5 Additionally, all extracted factors from the dataset are deemed acceptable only when their cumulative variance extraction surpasses 50% (Trong & Ngoc, 2008).
The exploratory factor analysis (EFA) identified four predefined factors, which collectively accounted for 91.014% of the variance among all observed variables, significantly exceeding the 50% threshold (Table 4.9).
Table 4.9 Total Variance Explained by EFA
Initial Eigenvalues Extraction Sums of Squared
Rotation Sums of Squared Loadings
Extraction Method: Principal Component Analysis
The Rotated Component Matrix (Table 4.10) reveals that all pre-designed items exhibit factor loadings exceeding 0.5 for their respective factors and below 0.5 for all other factors This analysis confirms the validity of our measurement scales, allowing us to retain all observed variables in our study.
Table 4.10 Rotated Component Matrix a by EFA
Extraction Method: Principal Component Analysis
Rotation Method: Varimax with Kaiser Normalization a Rotation converged in 6 iterations.
Accessing model fit
In this phase, we applied the gathered data to the model introduced in Chapter 2 and executed path analysis using the AMOS application, resulting in the path diagram shown in Figure 4.1.
Figure 4.1 Path diagram with regression weights calculated by AMOS
The AMOS output revealed a CMIN/DF value of 1.018, indicating a good model fit, as values between 1 and 2 suggest an acceptable fit according to Carmines and McIver (1981).
Table 4.11 CMIN value calculated by AMOS
Model NPAR CMIN DF P CMIN/DF
Browne and Cudeck (1993) demonstrated through their practical experience that a Root Mean Square Error of Approximation (RMSEA) value of approximately 0.05 or lower signifies a close fit of the model This finding supports the validity of our proposed model.
Table 4.12 RMSEA value calculated by AMOS
Model RMSEA LO 90 HI 90 PCLOSE
Besides these 2 commonly used value to measure model fit shown above, there’re also other defined parameters to access model fit as displayed in Appendix 2
So in conclusion, we had our proposed model fit with data collected over 181 valid respondents.
Testing hypotheses and answering research questions
To recall, in Chapter 1 we had the following researching questions:
• Question 1: Can switching barriers be separated into negative and positive factors?
• Question 2: How do positive switching barriers have impact to customer satisfaction and customer retention in mobile telecommunication service in Vietnam?
• Question 3: How do negative switching barriers have impact to customer satisfaction and customer retention in mobile telecommunication service in Vietnam?
To address the research questions, a comprehensive literature review was conducted in Chapter 2, leading to the selection of a model developed by Julander (2003) This model served as the foundation for testing our hypotheses and seeking answers to the identified research questions.
• H1: Negative switching barriers are negatively and directly associated with customer satisfaction
• H2: Negative switching barriers are positively and directly associated with customer retention
• H3: Positive switching barriers are positively and directly associated with customer satisfaction
• H4: Positive switching barriers are positively and indirectly associated with customer retention
While research question 1 was answered through literature review, question 2 will be answered by testing H1 and H2 Meantime, H3 and H4 will help to answer research question 3
The path diagram illustrates that negative switching barriers (neg_bar) adversely impact customer satisfaction while positively influencing customer retention Conversely, positive switching barriers (pos_bar) enhance customer satisfaction, leading to a favorable indirect effect on customer retention.
Table 4.13 shows the standardized regression weights of the model:
Table 4.13 Standardized Regression Weights of path model calculated by AMOS
Satisfaction = -0.414*Neg_bar + 0.498*Pos_bar Retention = 0.688*Neg_bar + 0.758*Satisfaction
The non-standardized regression model revealed that all regression weights exhibited a standard error of less than 1 and a significant value below 0.001, confirming that all estimates are statistically significant.
Table 4.14 Non-standardized Regression Weights of path model
Estimate S.E C.R P Label Satisfaction < - Neg_bar -.221 032 -6.911 ***
NSB1 < - Neg_bar 1.000 NSB2 < - Neg_bar 884 033 26.576 ***
PSB1 < - Pos_bar 1.000 PSB2 < - Pos_bar 867 048 18.011 ***
With all results analyzed above, we can confirm all the hypotheses (H1, H2, H3, H4) are accepted
More results of output from testing this model by collected data are presented in Appendix 2.