Master’s Degree Marketing
2013
Rabobank Midden en Noord Drenthe University of Groningen Bedrijfskunde & Economische Wetenschappen[
IS THE RELATION BETWEEN
RELATIONSHIP MARKETING AND
RELATIONSHIP QUALITY MODERATED
BY MEMBERSHIP?
]
Groningen, February 2013 Author: E.A.A. Latumalea Studentnr: 1428284
Supervisor: prof. dr. J. C. Hoekstra Co-Reader: Drs. J. Berger
Preface and Acknowledgements
In front of you lies not only my thesis but a part of my lives story. I’m proud to have been able to finish this thesis, although the proces was long and filled with drawbacks. I would like
to take the opportunity to thank my mother and farther for their unwavering support and
giving me just that one little push I needed when all seemed lost. Even though cancer, hearth
diseases and parkinson has ript through our lives you two have always been a beacon of hope
trust and especially perseverance.
I also like to thank my supervisor prof. Dr. J.C. Hoekstra for putting up with me during
this daunting period in my life. I may not have been the most perfect way of graduating but I
must say it has given me a lot of experience and understanding and wit hit I can close several
black pages in my life.
“It is good to have an end to journey toward; but it is the journey that
matters!”
Abstract
In this thesis the effects of antecedents, in literature mostly cited as relationship
marketing constructs, on the constructs of relationship quality is analyzed en discussed. The
relationship quality constructs are respectively defined as commitment, trust and relationship
satisfaction. Also the moderating effects of members versus customers of a financial
organization is tested for the relation between these antecedents and relationship quality
constructs. The outcome is that there exists no moderating effect for members relative to
customers. In multivariate context, without the moderating effects, there are significant direct
effects captured.
Keywords: Relationship Marketing, Relationship Quality, Antecedents, Membership
Index
PREFACE AND ACKNOWLEDGEMENTS ____________________________________ ABSTRACT ______________________________________________________________ II LIST OF TABLES AND FIGURES __________________________________________ V 1 INTRODUCTION ______________________________________________________ 1
1.1 Goal of this Thesis ___________________________________________________________ 3 1.2 Objective __________________________________________________________________ 4 1.3 Research Questions __________________________________________________________ 4 1.4 Cooperative Organizations and Moderation _______________________________________ 5 1.5 Procedure __________________________________________________________________ 6 1.6 Thesis Structure _____________________________________________________________ 6
2 LITERATURE STUDY AND HYPOTHESES _______________________________ 7
2.1 Relationship Marketing _______________________________________________________ 7 2.2 The Importance of Value in Relationships. ________________________________________ 8 2.3 Relationship Quality _________________________________________________________ 9
2.3.1 Commitment ____________________________________________________________________ 10 2.3.2 Trust __________________________________________________________________________ 11 2.3.3 Relationship Satisfaction __________________________________________________________ 12
3 CONCEPTUAL MODEL AND HYPOTHESES ____________________________ 14
3.1 Conceptual Model __________________________________________________________ 14 3.2 Customer-Focused Antecedents _______________________________________________ 15
3.2.1 Relational Benefits _______________________________________________________________ 15 3.2.2 Perceived Service Quality __________________________________________________________ 16 3.2.3 Dependence on Seller ____________________________________________________________ 16
3.3 Seller Specific Antecedents ___________________________________________________ 17
3.3.1 Relationship Investment ___________________________________________________________ 17 3.3.2 Seller Expertise _________________________________________________________________ 18
3.4 Dyadic Antecedents _________________________________________________________ 18
3.4.1 Communication _________________________________________________________________ 18 3.4.2 Similarity between Buyer and Seller _________________________________________________ 19 3.4.3 Relationship Duration ____________________________________________________________ 19 3.4.4 Interaction Frequency ____________________________________________________________ 20 3.4.5 Manifest Conflict ________________________________________________________________ 20
3.5 Moderator of Antecedents’ Influence on Relationship Quality ________________________ 21
4 METHODOLOGY ____________________________________________________ 22
4.4.1 Univariate Model ________________________________________________________________ 26 4.4.2 Multivariate Model ______________________________________________________________ 26 4.4.3 Moderation Effects in Multivariate Models ____________________________________________ 27
5 RESULTS OF EMPIRICAL STUDY _____________________________________ 29
5.1 Bivariate Outcomes _________________________________________________________ 29 5.2 Multivariate Outcomes ______________________________________________________ 31 5.3 Multivariate Outcomes with Moderation Effects __________________________________ 33
6 CONCLUSIONS ______________________________________________________ 35
6.1 Limitations________________________________________________________________ 37
List of Tables and Figures
TABLE 1: _____________________________________________________________________________23 TABLE 2: _____________________________________________________________________________24 TABLE 3: _____________________________________________________________________________29 Table 4:_______________________________________________________________________ 31 TABLE 5:______________________________________________________________________________34 TABLE 6:______________________________________________________________________________35 Table 7:_______________________________________________________________________ 361 Introduction
Briefly, marketing has progressed from a goods-dominant outlook, wherein tangible output and discrete transactions were central, to a service leading view. In the service leading view intangibility, exchange practices, and relationships are more important (Vargo and Lusch, 2004). Vargo and Lusch (2004) find it important to note that the service oriented view in economics should not be associated with the narrow and old-fashioned view which treated services as a by product (Rathmell, 1966). It should also not be treated as something to
support the goods sales. Lastly, services are not restricted to the services industries, like health care, government and education. Services are more broad than what was already known, and is redefined as the use of special attributes, like knowledge and expertise, which is
advantageous for its own unit or other units. In earlier days the study of marketing focused on the distribution and exchange of goods and industrial products and presented a basis in economics. The first marketing researchers focussed their attention on goods exchange. The main, goods-centered vision of marketing not only may hold back a complete appreciation for the role of services however also may partly block a widespread understanding of marketing in general. For instance, Gummesson (1995) states the following:
‘Customers do not buy goods or services: They buy offerings which render services which create value.… The traditional division between goods and services is long outdated. It is not a matter of redefining services and seeing them from a customer perspective; activities render services, things render services. The shift in focus to services is a shift from the means and the producer perspective to the utilization and the customer perspective.’
The concentration of marketing on primary competences essentially places marketing at the midpoint of business functions and disciplines. Prahalad and Hamel (1990, p. 82) propose, “core competence is communication, involvement, and a deep commitment to working across organizational boundaries.” Besides, they mention (p. 82) that primary
competences are “collective learning in the organization, especially [about] how to coordinate diverse production skills.” Channel intermediaries and network associates signify primary competences that are planned to achieve competitive gain by carrying out particular marketing functions. The firms can have long-term sustainability merely if they are harmonized with other channel and network associates.
dynamic desires. A service-centered leading understanding denotes that value is well-defined by and created together with the client instead of implanted in output. Haeckel (1999) detects successful businesses moving from practicing a ‘make-and-sell’ plan to a ‘sense-and-respond’ plan. In the service focused interpretation of marketing, firms are in a continuous course of making and testing hypotheses. The main goal is not output maximization as firms try to serve clients better and develop their performance.
Contemporary scientific literature has a particular interest in the way organizations
develop relationships with their customers (Woodruff and Gardial, 1996, Reinartz and Kumar, 2002, Venkatesan and Kumar, 2004). Since the origination of cooperative organizations, firms had a special relationship with their customers. These customers were members of the
organization and in turn in financial persepective responsible towards the cooperative organization. Over the past decades cooperative organizations, like all organization models, have been subject to changes. Relationship marketing (RM) has arisen as one of the leading concepts in business strategy environments, however RM surveys regularly produce mixed outcomes. Even though the major statement that RM has an positive effect on performance, various of the authors’ outcomes have noteworthy implications for research and practice.
RM, together with business practice and academic study, has experienced enormous development in the past years (Srinivasan and Moorman, 2005). Morgan and Hunt (1994) describe RM as marketing undertakings engaged towards establishing, developing, and sustaining successful relational connections. Most study and practice adopts that RM exertions cause durable customer relationships that enrich seller performance outcomes, comprising sales growth, share, and profits (Crosby, Evans and Cowles, 1990; Morgan and Hunt, 1994), nevertheless particular business directors have been dissatisfied in the success of their RM efforts (Colgate and Danaher, 2000). Investigators have also advocated that in certain circumstances, RM may have a negative impact on seller-focused outcomes (De Wulf, Odekerken-Schröder and Iacobucci, 2001; Hibbard et al., 2001).
quality, given by for example trust and satisfaction. They also state that to convert those opportunities into sales, managers are dependent on the characteristics similarity and expertise.
Generally, these discoveries signpost that the usefulness of RM work might show a discrepancy subject to the particular RM strategy and framework. This inconsistency with regard to performance suggests the need to integrate the empirical research and to understand better the RM strategies that are most effective for building strong relationships, the outcomes that are most affected by customer relationships, and the conditions in which RM is most effective for generating positive seller outcomes. Understanding the key drivers of RM usefulness can upsurge the firms’ yield on RM investments and offer investigators with insights into methods to form more comprehensive models of the effect of RM on
performance (Reinartz and Kumar, 2003). Most study have hypothesized the impacts of RM on outcomes as fully mediated by one or more of the relational concepts of trust, commitment and relationship satisfaction (Dwyer, Schurr and Oh, 1987; Crosby, Evans and Cowles, 1990; Morgan and Hunt, 1994). However, the aforementioned studies already are extensively
performed by for example Palmatier, Dant, Grewal & Evans (2006).
The role of members in service settings, has received limited attention.This thesis wants to make a signifying contribution to the settled works. This can be achieved by adding a new indicator to the study of Palmatier et al. (2006), which can divide the RM effects on
relationship quality in two groups. Therefore this thesis suggests to implement membership as an indicator. The prior knowledge and the addition of the moderating role of membership can contribute to a more reactive organization, which accommodates better to the wants of their members and/ or customers.
1.1 Goal of this Thesis
1.2 Objective
The prevailing literature suggests an extensive variety of antecedents for the various dimensions of relationship quality, and academics differ about which one best describes the characteristics of a relational interchange that influence customer-focused outcomes. For instance, Morgan and Hunt (1994) suggest that trust and commitment are key factors being explained by antecedents, while others advise that either trust or commitment solely is the critical significant dimension in the relationship quality (e.g., Anderson and Weitz, 1992; Gruen, Summers and Acito, 2000). In empirical context these different resulting constructs have been connected to several antecedents.
This research will focus on the antecedents of the various relationship quality dimensions, how they are perceived, how they differ between members and customers and finally how they are affected by antecedents. Alongside the operationalisation of the model a choice has been made to divide the study into two target groups namely, customers and members. This division examines the differences between these two segmentation groups. The relationship marketing model (RM strategies relationship quality) which is conceptualized in this model, can be applied across several different frameworks in which members or customers may have varying effects. Therefore, an objective of the present study is to identify and empirically test the influence of potential moderators on the linkages in the RM model. Can this knowledge contribute to a more responsive cooperative organization, that caters to the needs of members and perhaps recruit even more members? There will also be the possibility to see if there really are two groups or that there is little difference between the two groups.
1.3 Research Questions
This research, analytically reviews and analyzes the RM strategies, relationship quality and member/customer moderator to offer understanding of the following two research questions:
(1) Does membership of an organization enhances the effect of RM strategies on relationship quality?
1.4 Cooperative Organizations and Moderation
What is the main characteristic of a cooperative organization? A cooperative organization can not exist without its members. Without members the fundamental principal on whom a cooperative organization is based is undermined. Membership of a cooperative organization is defined as "a conglomerate of rights and duties to which the cooperative organization and the member commit towards each other." As a member you are part of a legal form. According to the Dutch legislature a cooperative organization has the same legal form as an association.
This means for cooperation the same rules apply as for example a sports club (Galle, 1999). Without cooperative members there is no cooperative organization. But why should members also be actively involved in the cooperative organization? The answer lies in the characteristics of the cooperative organization, such as voluntary membership, democratic decision-making and the tradition of member’s capital as a source of equity (Bijman, 2007). A high level of member engagement has the following advantages for a cooperative. Bijman, 2007 states that members are more loyal and therefore less likely to leave the cooperative in times of adversity. Members leaving the cooperative create problems, because scale factors for the efficiency of the cooperative are one of the main reasons that members join a cooperative organization. Also on the operational level, loyalty of members is necessary because it increases the reliability of the planning process. Secondly, members are less inclined to behave opportunistically (Annual Report, Rabobank Noord-Groningen, 2007). Groups always facilitate the possibility of free riding. Involvement prevents such behavior and reduces the need for expensive control mechanisms to be installed. Thirdly, members are more willingly to participate in the decision making process, they will more likely attend the general meeting or get involved in other forms of management. This promotes proper
1.5 Procedure
This thesis will construct a conceptual model that displays antecedents of customer relationships in financial cooperative setting. Following this conceptual model a survey will be held under customers and members of a financial cooperative organization to determine the way in which value of the different antecedents found in the literature are perceived by the two segmentation groups in the financial sector. Based on this data, this paper gives an insight into how members and customers differ with respect to their relationship with the financial cooperative organization.
1.6 Thesis Structure
Chapter 1 consists of an introduction into the background to this thesis. In Chapter 2, the literature will be addressed on customer relationships. Chapter 3 will start with the
2
Literature Study and Hypotheses
The key concepts in this study will be derived from the general relationship marketing theory and will therefore begin in section 2.1 with introduction of the key concepts in this field: relationship marketing, commitment, trust and relationship satisfaction.
2.1 Relationship Marketing
Relationship marketing was first drawn in the marketing literature by Berry (1983) as “attracting, maintaining and enhancing customer relationships”. The aim of relationship marketing is to form long-term mutually sustaining relations with suppliers, distributors and customers. By producing these long-term sustaining relations, the bussinesses preserve their long-term preference and business (e.g. Grönroos, 1994; Kotler, 2000). Relationship
marketing lets companies move from focusing on fixed transactions to building long term, profitable relationships that consumers perceive to be mutually beneficial (Berry, 1983, 1995). Relationship marketing centres itself on the notion that important client accounts need to be concentrated and continuous attention to these clients need to be set. Berry (1995) states that relationship marketing is not a precise strategy for all customers, some customers may be very profitable as transactional customers, but are not profitable as relationship customers. Consequently, multiple relationship strategies may be required at both the collective, and individual level (Grönroos, 1995). When customers have long term horizons and high switching costs (customers depend strongly on their present seller) relationship marketing works well, leading the supplier and customer to invest considerable assets and effort in keeping the relationship healthy. Service marketing and relationship marketing are closely related due to the interaction that frequently happens between the customer and the service provider. Services are generally high in experience and creating customer attitudes towards the service as well as the delivering organization, in general consumers will have more trust in word-of-mouth rather than advertising as a way of selecting the right provider. They also have a tendency to rely heavily on personnel, physical signals to evaluate quality and price, while staying loyal to service providing organizations that continuously fulfill their requirements (Hennig-Thurau et al., 2002).
to end their business (Berry, 1995). Marketing resources targeting current customers make it possible for relationship marketing to focus on two of these possibilities. Thy emphasize on this issue by reduction of the amount of customers which are planning to end their
relationship with the organization and through the up sell and cross sell of current
relationships. Whenever a customer tries to interact with a organization a service encounter occurs (Bitner, 1995).
2.2 The Importance of Value in Relationships.
The relational marketing perspective recommends that the customer should be
incorporated into the management process. This line of attack has its origins in the literature on industrial and service marketing. Efforts have been made to generate trustworthy
customers which can be very expensive for an organization. Research shows that relational marketing exertions improve the production of marketing efforts (Sheth and Parvativar, 1995). Early research has shown that this method is not only gainful but could also be a source to generate a maintainable competitive advantage (Reichfeld and Sasser, 1990). Empirical studies have exposed that holding a customer is ten times cheaper than attracting new customers (Heskett et al., 1990). It can be concluded that efforts must be made by companies to try to retain customers.
To start a relationship, there must be at least two parties that will gain certain advantages and benefits when developing and maintaining relationships (Gwinnner et al., 1998). The benefits a supplier derives from the relationship are coupled to the loyalty of the customer. A loyal customer will generate more revenue than a customer that discontinues the relationship. This loyalty leads to increasing the volume of business the customer does with the
organization. Maintaining en nurturing a good relationship with the customer will also provide more insight into the needs of the customer, this will result in an advantageous position for the organization to manoeuvre or adjust accordingly. Miscellaneous costs may be reduced in this way.
the fact that this way of getting recommendation is cheaper it also generates a level of trust difficult to copy by other means of marketing communication. References through word of mouth are more loyal than customers that enter into a business relation through other channels (Goodwin and Gremler, 1996). On the other hand one must take into account the fact that not all customers are alike. It is therefore essential to focus on generating value for the right customers (Jones and Sasser, 1995; Reichheld, 1996). From a customer standpoint, they will obtain economic benefits through the accumulated value of the relationship (Peterson, 1995). In addition, they receive a service that better fits their needs. It also creates the possibility to enjoy a special treatment or extra services from the company (Gwinner et al., 1998; Rust et al., 2000).
Another type of major benefits that may be relevant to the customer is the social benefits resulting from building a relationship with the provider. In summary, clients initiate and maintain a market relationship because they expect that there will be value gained due to their participation in this relation (Peterson, 1995). From this perspective, a strategic change is needed within organizations. Organizations should concentrate their management more on the value perceived by the customer. The goal should be to create a successful relationship
strategy to target their customers with (Ravald and Grönroos, 1996; Reichheld, 1996; Weistein and Johnson, 1999; Rust et al., 2000). The first step to implement a successful relationship strategy should be to find out exactly how customers form their opinions.
Secondly, organizations have to highlight on how these opinions can be measured in order to use it as a tool for management. Woodruff (1997), emphasizes the lack of operational tools of value management that are currently available for organizations.
2.3 Relationship Quality
The selection of relationship quality as a relationship outcome in this research is consistent with earlier research on relationship marketing (Kumar, Scheer and Steenkamp, 1995). Relationship quality has been reflected as a general valuation method of the power of a relationship (Garbarino and Jobnson, 1999; Smith, 1998). Earlier studies hypothesize
and Cowles (1990) and Dwyer, Schurr, and Oh (1987) define relationship satisfaction and trust being indicators of the higher-order concepts of relationship quality.
Hennig-Thurau and Klee (1997), Leuthesser (1997) and Dorsch, Swanson and Kelley (1998) claim to complement the dimensions of relationsbip quality with relationship satisfaction. Consequently, the aforementioned authors adopt that an enhanced quality relationship is supplemented by satisfaction, trust, and commitment. They favor the abstract relationship quality construct above its more detailed dimensions since, although these
numerous forms of attitude might be theoretically different, customers have difficulty making well distinctions between them and have a habit to lump them together (Crosby, Evans and Cowles, 1990). In the following paragraphs, the different dimensions of relationship quality are elaborated.
2.3.1 Commitment
Morgan and Hunt (1994) define relationship commitment as a conversation partner trusting that a good relationship with another is so vital such that maximum struggle has to be made to sustain the relation. In other words, the devoted party is certain of maintaining the relationship to make certain that it stands for a very long time. He builds the relationship on three different constructs in the literature, namely commitment in social exchange by Cook and Emerson (1978), marriage by Thompson and Spanier (1983) and organizations by Meyer and Allen (1984). The definition of Morgan and Hunt (1994) is the same as which is defined and developed by Moorman, Zaltman and Deshpande (1992): "Commitment to the
relationship is defined as an enduring desire to maintain a valued relationship."
The "valued relationship" resembles the thought that relationship commitment can be present merely in case the relationship is believed to be essential. In same fashion one can state that, their “enduring desire to maintain” is in line with the description of Morgan and Hunt (1994) that a committed party wants the relationship to stand forever and is keen to make efforts at retaining this relationship.
exchange." More explicitly, in the marriage research papers, McDonald (1981) has defined commitment differently, "Clearly, the major differentiation of these exchange relationship types, is the mutual social trust and the resultant commitment on the part of the individuals to establish and maintain exchange relationships."
Besides, commitment has been believed to be a serious concept in the literatures in which organizational, managerial, seller and buyer behavior is a key topic. According to Becker (1960) and Reichers (1985), among the dimensions of the summed behaviors, organizational commitment is one of the oldest and most studied constructs in organizational behavior theory. It is one type amongst relationship commitment which has been defined as very important to the organization in its internal relationships. In this perspective, commitment can be deliberated as vital to this research since it leads to such essential results in which turnover decreases (Porter el al., 1974), greater stimulus to the motivation of parties (Farrell and Rusbult, 1981), and as a last point it rises the organizational and social responsibility
behaviors (Williams and Anderson, 1991). Besides, relationship commitment arises from the aforementioned variables, it also constitutes from elements which can directly be manipulated by the organization, like employing and training practices, job equity, and managerial and directional support (Eisenberger et al., 1990; Caldwell, et al., 1990).
Berry and Parasuraman (1991) define for the services relationship marketing dimension that "Relationships are built on the foundation of mutual commitment." In same fashion, studies have argued that consumers follow a path in which they are keen to stick to one particular brand. The key point of argue was why consumers in general are more loyal to competitive brands and won’t deviate from their choice. Initially, loyalty was viewed as simply repeat buying but it was difficult to agree on the ‘true’ incentives behind commitment.
2.3.2 Trust
Since a period of a decennia, some of the studies have revealed that trust can have a significant and positive effect on the behaviors and attitudes of a company’s customers and/or members and channel associates. To give an illustration, some authors have submitted that trust significantly has a negative impact on opportunism (Rindfleisch and Moorman, 2003); has a positive effect on customer loyalty (Agustin and Singh, 2005); a positive effect on service usage (Maltz and Kohli, 1996); gives rise to more commitment (Jap and Ganesan, 2000); and in addition trust positively influences cooperative and interactive exchange relationships (Cannon and Perreault, 1999; Hibbard, Kumar and Stern, 2001; Jap and
to recognize particular activities that organizations can carry out to develop trust. To give another example, organization have been able to boost op trust by maintaining a good conversation with customers or members (Anderson and Weitz, 1989; Doney and Cannon, 1997), have them satisfied (Ganesan, 1994), and keeping and nurturing mutually supporting relationships with them (Kumar, Scheer and Steenkamp, 1995b). Several other authors having studied relationship marketing (e.g., Doney, Cannon and Mullen, 1998; Geyskens, Steenkamp and Kumar, 1998), trust as a construct which is defined as partner interacting generous and truthful. This research has adopted a “narrow scope trust” in which a reference is made to customers and members of individual organizations and to their exchange partners. The definition of this trust is narrow in the sense that it solely has an effect on the relation between the parties (customers/members and representatives of the organization). Thus this kind of trust construct has a relatively limited range of influence outside the range of relationship of the parties.
2.3.3 Relationship Satisfaction
Over the past few years academics have called attention to the concept of customer relationship in academic papers and made practitioners aware of this interesting construct. As a matter of fact relationship marketing have been given popularity through the presumption that working on customer relationships contributes in a positive sense to the increase of customer satisfaction, loyalty, word of mouth and consumptions.
Berry and Parasuraman (1991) and Czepiel (1990) has found that working on the construction of customer relationship to upsurge the customer satisfaction and
trustworthiness. Griffin (1995) has found that customer relationship gives rise to the extent of advantageous word of mouth, while again Berry and Parasuraman (1991) have found that this kind of relationship had a significant and positive effect on purchases. To give an example, (Prokesch, 1995) has sketched in his paper that British Airways have initiated a test on a group of loyal “premium customers” with which they have maintained an experimental relationship strategy. After the test period they have noticed that the volume of produced business from these test population have risen the revenues by approximately ten percent (Prokesch, 1995).
members (Fornell, Johnson, Anderson, Cha and Bryant, 1996; Anderson, Fomell and Lehmann, 1994). Maddox (1977) tells that in the retailsector the "institutional" satisfaction model offers customers satisfactions which are different in characteristics from the goods they retail. Singh (1991) mentions that there possibly various types of different goods or object characteristics which are judged by customers on the level of satisfaction they enjoy from. To give an illustration, if consumers “buy’ health care facilities from a hospital, they evaluate each of the levels of service they enjoy in terms of satisfaction. The levels of service can involve the physician, the hospital accommodation and the services from the insurance provider.
The literature has found that customers or members of an organization have very distinct expectations from the exchange partners of the organizations. The expectation of consumers relates to interactions with the different objects and goods of the selling party. So the objects are isolated and evaluated independently. Researchers like Swan and Oliver (1989), have empirically found differences between customers or members of an organization concerning the relation between customer related antecedents and relationship satisfaction. In addition experience is something untouchable earned by customers being satisfied by the service or good purchase from their vendor and their interaction with their salespartner (Westbrook, 1981). Consumers can be satisfied by their relation with the vendor or the shop and these satisfaction arising from these two “objects” can be related, however this thesis treats these two as separate constructs.
Relationship is the exchange of a buyer’s loyalty and a vendor’s good quality service on a serviceplatform on which these dynamics are played altogether within the rules of a
customer’s view of satisfaction. In plain words, the customer has a prejudgement about the service he or she would like to enjoy from the vendor. If this is lived up to the high
expectations of the customer then satisfaction is the outcome (Crosby, Evans and Cowles, 1990). In fact Gwinner et al. (1998) have empirically found that customer’s perception of the benefits they have received from service was significantly and positively related with
3 Conceptual Model and Hypotheses
3.1 Conceptual Model
The conceptual model is based on an analytical framework in the relationship marketing field of study by Palmatier, Grewal and Evans (2006). Sweeney and Soutar (2001) made some adaptations to their model by representing three dimensions represented by functional,
emotional and social benefits. The emotional or social dimensions are represented by relational benefits, perceived service quality; and functional dimension is represented by the relationship investment and seller expertise. Gordon, Mckeage and Fox (1998), concluded that commitment on the part of the consumer with regards to thought, feelings and behavioral response to a product, is significant to the long duration of client relation. Highly committed customers or members will show longer relationship duration as well as a higher interaction frequency. The main contribution of this study is the introduction of a moderator variable called membership. This study will examine whether the strength of the relationship between the antecedents and outcomes in the form of trust, commitment, and relationship satisfaction is influenced by type of formal relationship that exists in the shape of the customer or member of a financial cooperative organization. Based on the conceptual model the goal is to examine the differences in appreciation of the antecedents by members, compared with customers of a financial cooperative organization. In the following sections these variables are defined and explained and hypotheses are formed.
Figure 1: Conceptual model
3.2 Customer-Focused Antecedents
3.2.1 Relational Benefits
Customers can value a relationship when they receive relational benefits. In other words, time savings, convenience, and companionship. These advantages are proven to have a positive effect on relational moderators (Morgan and Hunt, 1994; Renolds and Beatty, 1999). This makes a very important point which is that one must assume that both parties should obtain benefits from the relationship. For customers, these benefits can concentrate on the core of the service or the relationship with the supplier itself (Henning-Thureau, Gremler and Gwinner, 2000). It is possible to distinguish three typologies for relational benefits. Relational benefits that arise from reduced sense of fear, anxiety or stress. The customer knows what to expect when he purchases the service. There are also social benefits. These relate to the emotional part of the relationship and are characterized by personal recognition of the customer by the supplier. This develops friendships between the client and service provider. Finally, special treatment benefits that the customer can obtain like discounts, faster service and additional individual custom services.
According to Beatty et al. (1996) the importance of personal interaction between customers and sellers in persuading relationship outcomes is not astonishing agreed that relationships are fundamentally social processes. For instance, nearly five decades back in time, Stone (1954) stressed the importance of social interaction of people in identifying the presence of customers who appreciate personal interaction in a store. Evans, Christiansen and Gill (1996) advocate that the social interaction has been proposed to be "the major stimulus for particular consumers to visit sales establishments." Examples of social interaction benefits are shaped by feelings of familiarity, friendship, trust, commitment and social support (Berry, 1995); personal acknowledgement and the usage of a customer's name (Howard, Gengler and Jain, 1995); recognizing the customer as a person; enjoying in kindly conversations; and demonstrating friendliness (Crosby, Evans and Cowles, 1990). Therefore the following can be hypothesized:
H1a. Relational benefits positively influences commitment; H1b. Relational benefits positively influences trust;
3.2.2 Perceived Service Quality
If customers assessments of the service quality is high, the customers behavioral intentions are favorable, which strengthens his or hers relationship with the company. When service quality
assessments are low the customer’s behavioral intentions will be unfavorable and the relationship will be more likely to be weakened (Zeithaml, Leonard and Parasuraman, 1996). It is therefore
hypothesized that perceived service quality will have a positive effect on the degree of commitment, trust and relationship satisfaction of a customer or member of a cooperative financial organization.
H2a. Perceived service quality positively influences commitment; H2b. Perceived service quality positively influences trust;
H2c. Perceived service quality positively influences relationship satisfaction.
3.2.3 Dependence on Seller
Dependence on seller shows the customers evaluation on the value of the resources that the seller adds to the relationship, for which few alternatives are found (Hibbard, Kuma and Stern, 2001). Researchers found both negative and positive influences in the empirical results for the relative dependence on relational moderators (Anderson and Weitz, 1989; Morgan and Hunt, 1994). This implies that the impact depends on the context for each situation. Price can play different roles in the customer decision making process. Prior to the actual purchasing customers can use price as an indicator of quality. This will create expectations about the level of the service the customer can expect (Zeithaml & Bitner, 2003). After they have consumed the service price will be a cost that a customer is willing to pay for the service. This in turn contributes to the value a customer assigns to the service (Lovelock, 2001). It is
therefore hypothesized that dependence on seller will have a positive effect on commitment, trust and relationship satisfaction the higher the perceived dependence on seller the more committed customers are to participating in the relationship they will also have a higher level of trust in de seller and will perceive the overall relationship as better. The following can with this respect be hypothesized:
H3a. Dependence on seller positively influences commitment; H3b. Dependence on seller positively influences trust;
3.3 Seller Specific Antecedents
3.3.1 Relationship Investment
Relationship investment is the time, effort and resources that the seller is willing to invest in the relationship with the customer. Such investments often generate expectations and reciprocal behavior that can help strengthen the relationship and therefore a positive effect on relational mediators (Anderson and Weitz, 1989; Ganesan, 1994). It is hypothesized that both members as customers will perceive relationship investment by the selling organization as a positive reinforcement of the relationship. By improving the overall feeling a customer of member has with the relationship it also hypothesized to have a positive effect on
commitment, trust and relationship satisfaction. According to de Wulf et al. (2001) the link between relationship investment and relationship quality has seldomly been examined empirically. A prominent exemption is the robust support Crosby, Evans, and Cowies (1990) find for a positive effect of relational selling behavior on relationship quality. Additionally, Wray, Palmer and Bejou (1994) discover a positive effect of salesperson's customer
orientation on relationship quality. Lastly Lagace, Dahlstrom and Gassenheimer (1991) have discovered a positive effect of ethical salesperson behaviour on relationship quality. Even though these constructs are not fully comparable to the constructs of this research’s
relationship investment, they deliver an preliminary foundation for the following hypotheses. However stronger proof can be established for the influence of relationship investment on the dimensions of relationship quality. It has been shown that relationship investment predicts satisfaction in business promoting interactions (Anderson and Narus, 1990; Ganesan, 1994; Smith and Barclay, 1997). Baker, Simpson, and Siguaw (1999) show that customers have a habit being extra pleased with sellers who make careful efforts in the direction of customers. Moreover, relationship investment has been exposed to cause trust. For instance, Ganesan (1994) advocate that relationship investments made by a vendor result in improved trust by a customer. Bennett (1996) claims that the power of customers’ commitment is subject to their sensitivities to the efforts made by the vendor. Consequently, this research hypothesizes the following:
H4a. Relationship investment positively influences commitment; H4b. Relationship investment positively influences trust;
3.3.2 Seller Expertise
Seller expertise consist of, the knowledge, experience and general competence of the seller. When customers interact with a competent vendor, they get a higher value for that relationship back. The relationship becomes more important and the customer is more likely to invest in the relationship in order to preserve and strengthen is (Crosby, Evans and Cowles, 1990; Lagacé, Dahlstrom and Gassenheimer, 1991). It is therefore hypothesized that the more expertise a selling organization can display towards their customers or members, the more positive these customers or members will evaluate the relationship. This in turn will have a positive effect on their commitment to maintaining the relationship, their trustworthiness of the relationship and their overall relationship satisfaction. The respective hypotheses will be:
H5a. Seller expertise positively influences commitment; H5b. Seller expertise positively influences trust;
H5c. Seller expertise positively influences relationship satisfaction.
3.4 Dyadic Antecedents
These antecedents are important for both the customer and the seller. They also requires an active involvement by both parties.
3.4.1 Communication
Communication is needed to create a strong relationship. This helps to synchronize goals, solve problems and identify new value-creating opportunities during the relationship (Morgan and Hunt, 1994). It is hypothesized that communication will have a positive influence on the value a customer or member experiences. Without communication commitment, trust and relationship satisfaction will not be able grow. For communication the next hypotheses will be tested:
H6a. Communication positively influences commitment; H6b. Communication positively influences trust;
3.4.2 Similarity between Buyer and Seller
Similarity in this context means the degree of similarity in appearance, lifestyle and status between individual boundary spanners. Or it means the comparability of the culture, values and goals of the organizations concerned. This similarity may indicate that the partner might be prepared to actively assist in achieving important goals. It has also been proven to have a positive effect on the relational mediators (Crosby, Evans and Cowles, 1990; Doney and Cannon, 1997), i.e. the next hypotheses are relevant for this research:
H7a. Similarity positively influences commitment; H7b. Similarity positively influences trust;
H7c. Similarity positively influences relationship satisfaction.
3.4.3 Relationship Duration
Relationship duration is the length of time that the relationship between the exchange partners (Scheer and Steenkamp, 1995). It is important to know the orientation of the
customer. To little understanding of this approach can cause a company to decide to develop a long term relationship while a short term relationship would be more appropriate. Customers with a long-term orientation will not quickly change vendors. This makes them very attractive for organizations. Organizations will achieve more benefits when they concentrate on
targeting the right segments within a market for which a long-term relationship is the best (Ganesan, 1994). In addition the hypotheses to test are:
H8a. Relationship duration positively influences commitment; H8b. Relationship duration positively influences trust;
3.4.4 Interaction Frequency
Interaction frequency means the number of interactions per unit of time between the exchange partners (Doney and Cannon, 1997). Both these antecedents reciprocal relationship duration and interaction frequency provide the partners with more behavioral information of the other. This leads to more accurate predictions of the behavioral actions the partner will undertake which in turn increased confidence in the exchange partner (Anderson and Weitz 1989, Doney and Cannon 1997). The hypotheses to be tested are:
H9a. Interaction frequency positively influences commitment; H9b. Interaction frequency positively influences trust;
H9c. Interaction frequency positively influences relationship satisfaction.
3.4.5 Manifest Conflict
Conflict means the degree of disagreement between the two partners. When conflicts increase customers will have less confidence that building a long term relationship is the right way to proceed. It would thus have a negative impact on trust and commitment by the
customer on behalf of the seller (Anderson and Weitz 1992, Kumar). Therefore logically the following hypotheses are necessary:
H10a. Manifest conflict negatively influences commitment; H10b. Manifest conflict negatively influences trust;
3.5 Moderator of Antecedents’ Influence on Relationship Quality
This thesis has described testing the effects hypothesized thus far and also carry out an preliminary step in the direction of considering the role of moderators that influence the usefulness of perceived relationship investment. An analysis of such moderators permits vendors to comprehend when investing in relationships is anticipated to be extra effective or less effective. Given the focus of this thesis and a general interest in the costumer and members of an organization, this research examines whether tbe effects of perceived
relationship investment are depending on any of two consumer features: being a customer or being a member. Christy, Oliver, and Penn (1996) have prompted the notion that certain customers are essentially motivated to involve in relationships. Even with the acknowledged importance of customers' likeliness to involve in relationships with vendors, no academic study has thus far examined the moderating influence of customers and members on relationship effectiveness. It is assumed that relationship-prone customers should make a return for something given or done by a retailer, because by meaning, relationship-prone customers are most expected to develop relationships.
It is expected that being a member of a cooperative financial organization will affect the strength of the relations between antecedents of relationship marketing and the relationship quality (Bijman, 2007)). Besides, Chen, Chen, Chen and Wey (2011), explore the notion of relationship marketing to define empirically whether a salesperson’s exertions in the area of relationship marketing can produce significant outcomes and whether customer characteristics moderate the relationship between antecedents of relational mediators and customer loyalty. Further a different study of Seider et al. (2005) and Evanschitzky and Wunderlich (2006) recap and extend the prior studies by suggesting that characteristics of customer moderate the relationship between satisfaction and repurchase behavior. In addition to the prior literature, this study makes a contribution by assuming that customers and members have a moderating effect between antecedents and relationship quality, hence the hypotheses:
H11a. Membership of an organization positively enhances the effect of RM strategies on communication;
H11b. Membership of an organization positively enhances the effect of RM strategies on trust;
4 Methodology
This chapter explains the collection of research data, the method by which the hypotheses are tested to eventually answer the research questions. usable
4.1 Data Collection
The study took place in May 2010 among 3573 members and 3764 customers of Rabobank. Respondents received an e-mail with the request to participate in this study. A pilot was conducted among 20 members of the bank. In addition, a questionnaire designed and presented to four managers representing the marketing communications department. After their evaluations the formulation of some questions were changed to avoid
misunderstandings. A week after this survey, a request to 4000 customers and members of Rabobank was sent to participate to the survey. The board of the Rabobank Northeast Drenthe decided not to give an incentive to people to induce the number of participations to this investigation. After several reminders 440 members and 343 clients completed the survey. The 783 returned questionnaires represent a 22% response rate. The sample was randomly taken with the preconditions that customers or members should at least be older than 18 years and focus was on personally business with the bank only. Only business customers and members were excluded from this study.
4.2 Measurements
For the constructs considered in this thesis, measures were borrowed from preceding studies. Especially, social and emotional treatment relational benefits of the customer are measured with the items delivered by Gwinner, Gremler and Bitner (1998); satisfaction items are a subclass of the items from Oliver (1980); and the commitment construct is a collection of a subclass of items from Morgan and Hunt (1994). In agreement with the intention of this thesis the trust items come from Morgan and Hunt (1994). The construct relational benefits is in the prior literature also depicted as emotional and social value or social benefits. The construct dependence on seller is in earlier studies also depicted as switching costs.
assessed by the following procedure. First, all raw questionnaire items listed in Table 2 below were factor analyzed. This analysis produced 13 factors that mirrored the predetermined scales: Relational Benefits, Perceived Service Quality, Dependence on Seller, Relationship Investment, Seller Expertise, Communication, Similarity, Relationship Duration, Interaction Frequency, Manifest Conflict, Commitment, Trust and Relationship Satisfaction. Then multi-item scales were constructed by averaging multi-item responses, correcting for reverse coding where appropriate.
4.3 Content and Face Validity
Scale development was based on the review of relevant literature on relationship marketing. From literature review an initial set of items was proposed. Where necessary the items were adapted to the cooperative banking sector. This adaptation had the objective of guaranteeing the face validity of the measurement instruments. Face validity is defined as the degree that respondents judge that the items are appropriate to the targeted construct. Content validity is de degree to which items correctly represent the theoretical content of the construct and it is guaranteed by the in-depth literature review undertaken. In line with Lichtenstein et al. (1990) an item was retained if a high level of consensus was observed among the experts.
Table 1: Cronbach’s Alpha
To assess the reliability of the scales Cronbach's alpha is used and composite reliability. Scales that have an α as low as ≤ 0.600 were considered unusable (Nunnally and Bernstein,
1994). Table 1 shows the resulting Cronbach’s alphas of this research. Because many of the items proposed by literature were conducted under other circumstances, reformulation of the questions was necessary. All items were reviewed and assessed by experts in the cooperative financial sector on how to adjust the formulation to match the current empirical situation. This research retained items that minimally loaded 0.600 on the proper latent factor and
maximally. As Cronbach's alpba values ranged between 0.689 and 0.959, reliability was uniformly high in all samples for all 13 constructs. This provided with enough confidence to calculate averages for all the constructs based on the amount of items of each construct and use these averages as indicators of the construct relationship quality (see Crosby, Evans, and Cowles 1990; Fosdakoffand Mackenzie 1994).
4.4 Method of Analysis
4.4.1 Univariate Model
The bivariate relationships among the variables were assessed via linear regression analysis (Joreskog and Sorbom, 1984). They also used to estimate model parameters with each of the antecedents as independent variables and each of the relationship quality variables as dependent variable as data input. The univariate linear regression is utilized to measure one-to-one the effects of antecedents on the relationship quality indicators. Gerbing and Anderson (1985) resolved that a sample size of 150 or more typically offers reliable estimations which produce standard errors small enough to be suitable in experiments (Anderson and Gerbing, 1988). Hence the present study meets the minimum sample size condition. This thesis accomplished analyses for the effect of each antecedent on one of the three relationship quality constructs which allows to compare the impact of each antecedent on each construct. Also in line with the prior research of Palmatier et al. (2006) this thesis estimates the following model:
i i
i Antc
RQ Eq. 1
Where RQi is the dependent variable and representing one of the three relationship
quality constructs commitment, trust or relationship satisfaction and Antci is the independent
variable and represented by one of the antecedent constructs: relational benefits, perceived service quality, dependence on seller, relationship investment, seller expertise,
communication, similarity, relationship duration, interaction frequeny or manifest conflict. α, βand εi are respectively the intercept, effects captured by the independent variable, and
residuals.
4.4.2 Multivariate Model
models are presented for each of the relationship quality construct commitment, trust and relationship satisfaction as dependent variable:
i i i i i i i i i i i i MC IF RD Sim Com SE RI DoS PSQ RB Comm 10 9 8 7 6 5 4 3 2 1 Eq. 2 i i i i i i i i i i i i MC IF RD Sim Com SE RI DoS PSQ RB Trust 10 9 8 7 6 5 4 3 2 1 Eq. 3 i i i i i i i i i i i i MC IF RD Sim Com SE RI DoS PSQ RB RS 10 9 8 7 6 5 4 3 2 1 Eq. 4
This method of analysis is again in line with the work of Palmatier et al. (2006), where Commi, Trusti and RSi are dependent variables and representing respectively the constructs
commitment, trust and relationship satisfaction. The independent variables are given by: RBi ,
the relational benefits, PSQi , the perceived service quality, DoSi , the dependence on seller,
RIi , the relationship investment, SEi , the seller expertise, Comi , the communication, Simi ,
the similarity, RDi , the relationship duration, IFi , the interaction frequeny and MCi , the
manifest conflict. α and εi are resp. the intercept and residuals. α, β1 through β10 and εi are
respectively the intercept, effects captured by the independent variables, and residuals.
4.4.3 Moderation Effects in Multivariate Models
Besides the aforementioned regression analyses on the complete sample, a multivariate regression analyse is performed on the members versus the customers. First the population is divided in the subsamples members and customers. Subsequently, a multivariate regression analysis is performed on the two subsamples. Afterwards, the significance of the relational effects between the dependent and independent variables of the regression analyses for each of the subsamples are compared.
i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i Mod Mod MC Mod IF Mod RD Mod Sim Mod Com Mod SE Mod RI Mod DoS Mod PSQ Mod RB MC IF RD Sim Com SE RI DoS PSQ RB Trust 21 20 19 18 17 16 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 Eq. 6 i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i Mod Mod MC Mod IF Mod RD Mod Sim Mod Com Mod SE Mod RI Mod DoS Mod PSQ Mod RB MC IF RD Sim Com SE RI DoS PSQ RB RS 21 20 19 18 17 16 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 Eq. 7
Where Commi, Trusti and RSi are dependent variables and representing respectively the
constructs commitment, trust and relationship satisfaction. The independent variables are given by: RBi , the relational benefits, PSQi , the perceived service quality, DoSi , the
dependence on seller, RIi , the relationship investment, SEi , the seller expertise, Comi , the
communication, Simi , the similarity, RDi , the relationship duration, IFi , the interaction
frequeny and MCi , the manifest conflict. The moderation effects of the members versus the
customers for each of the aforementioned antecedents are represented by the abbreviations: RBiModi, the moderation effect for relational benefits, PSQiModi , the moderation effect for
the perceived service quality, DoSi Modi , the moderation effect for the dependence on seller,
RIiModi , the moderation effect for the relationship investment, SEiModi , the moderation
effect for the seller expertise, ComiModi , the moderation effect for the communication,
SimiModi , the moderation effect for the similarity, RDiModi , the moderation effect for the
relationship duration, IFiModi , the moderation effect for the interaction frequeny and
MCiModi , the moderation effect for the manifest conflict. Modi is a dummy variable and
representing the moderation of the members versus the members, where members are coded 1 and customers are coded 0. α, β1 through β21 and εi are respectively the intercept, effects
5 Results of Empirical Study
5.1 Bivariate Outcomes
In Table 5 below, the results of the regression estimations are summarized for the direct effects of antecedents on the relationship quality constructs. Overall, except for some of the relations, the results support that the antecendents have an significant effect on the constructs commitment, trust and relationship satisfaction. The average of the estimated effects among antecedents and mediators range from -0.698 for manifest conflict to 0.807 for similarity.
Of all antecedents Similarity has the greatest average positive impact on the constructs (β = 0.807). The importance of similarities between buyers and sellers suggests that without common reference points, exchange partners may find it difficult to move the exchange from a purely economic or transactional basis to a relational basis.
Table 3: Influence of Antecedents on each of the Constructs of Relationship Quality. The population consists of n = 783 individuals.
Estimated Model: RQi = α + βAntci + εi
Antecedent Relationship quality β R2
Similarity is a proxy for customers’ perceptions of a seller’s fit with their in-group (Devine 1995). Similarity is the commonality in appearance, lifestyle, and status between individual boundary spanners or the similar cultures, values, and goals between organizations. Such similarities between people or organizations may provide cues that the exchange partner will help facilitate important goals and has been shown to affect contructs positively (Crosby, Evans and Cowles, 1990; Doney and Cannon, 1997).
Table 3 (Continued)
Influence of Antecedents on each of the Constructs of Relationship Quality. The population consists of n = 783 individuals.
Estimated Model: RQi = α + βAntci + εi
Antecedent Relationship quality β R2
Similarity Commitment 0.929 *** 0,529 Similarity Trust 0.729 *** 0,541 Similarity Relationship satisfaction 0.762 *** 0,495 Relationship duration Commitment -0.044 *** 0,020 Relationship duration Trust -0.014 0,003 Relationship duration Relationship satisfaction -0.019 ** 0,005 Interaction frequency Commitment -0.118 0,001 Interaction frequency Trust -0.039 0,000 Interaction frequency Relationship satisfaction -0.197 * 0,004
Manifest conflict Commitment -0.657 *** 0,134 Manifest conflict Trust -0.694 *** 0,247 Manifest conflict Relationship satisfaction -0.744 *** 0,238 * Significant at 10% level ** significant at 5% level *** significant at 1% level
The second largest average positive impact on relationship constructs come from communication (β = 0.514). The large positive effect of communication on all the independent varianbles is consistent with its role in both uncovering opportunities and resolving problems.
Perceived service quality (β = 0.446), seller expertise (β = 0.446), relationship benefit (β = 0.433) and relationship investment (β = 0.419) have the third largest average positive impact on relational mediators. The strong effect of perceived service quality on the
employees. The positive impact of the relationship benefits again indicates that managers should engage in proactive spending. The strong positive impact of the seller’s relationship investments indicates that managers should engage in proactive spending.
Of all antecedents conflict manifest (β = -0.698) has the largest negative impact on the constructs. This finding supports the importance of resolving problems and disagreements to prevent relationship-damaging conflicts (alternatively, the presence of conflict may seriously undermine the effect of other antecedents). The last three antecedents – dependence on seller (β = 0.045), relationship duration (β = -0.019) and interaction frequency (β= -0.118) – have notably smaller effect on the independent variables.
5.2 Multivariate Outcomes
In Table 6 below, the outcomes of the multivariate regression model is shown. As can be seen in each column the estimates are given for each estimate with respectively the dependent variables commitment, trust and relationship satisfaction. Besides interaction frequency and manifest conflict al the remaining independent variables (antecedents) are significant at 1% or 5% for the estimate with commitment as dependent variable. The explanatory power of this estimate is relatively high, namely R2 = 0.630.
Table 4: The simultaneous effects of all antecedents on each of the Constructs of Relationship Quality.
Dependent variable Commitment Trust Relationship satisfaction
Independent variables Beta P-value Beta P-value Beta P-value
Intercept 0.446 0.643 1.842*** 0.009 0.659 0.315
Relationship benefits 0.267*** 0.000 -0.010 0.839 0.052 0.241
Perceived service quality 0.094*** 0.007 0.145*** 0.000 0.123*** 0.000
Relationship investment 0.213*** 0.000 0.069*** 0.004 0.250*** 0.000 Dependence on seller 0.123*** 0.000 0.007 0.658 -0.002 0.885 Seller expertise 0.076** 0.021 0.122*** 0.000 0.129*** 0.000 Communication -0.135** 0.038 0.100** 0.035 0.313*** 0.000 Similarity 0.605*** 0.000 0.381*** 0.000 0.232*** 0.000 Relationship duration -0.027*** 0.000 -0.011** 0.036 -0.018*** 0.000 Interaction frequency 0.022 0.795 0.085 0.176 -0.037 0.528 Manifest conflict -0.074 0.125 -0.137*** 0.000 -0.084** 0.011 R2 0.630 0.674 0.761 F-statistic 131.63 159.63 245.31 P(F-statistic) 0.000 0.000 0.000 Obs. 783 783 783
The third column shows the estimate with trust as dependent variable. In this model specification, dependence on seller and interaction frequency are the only independent variables which are not significant. The remaining variables are significant at at 1% or 5%. The explanatory power of this estimate is relatively to the previous estimate higher, namely R2 = 0.674. The fourth column shows the estimate with relationship satisfaction as dependent variable. In this model specification, again dependence on seller and interaction frequency are the only independent variables which are not significant. The remaining variables are
significant at at 1% or 5%. The explanatory power of this estimate is relatively to the previous estimate higher, namely R2 = 0.761.
The consequences for the hypothess according to the results in Table 4 are briefly summarized:
Accept H1a: relational benefits positively influence commitment;
Reject H1b and H1c: relational benefits do not positively influence trust and relationship satisfaction;
Accept H2a , H2b and H2c: perceived service quality positively influences commitment, trust and relationship satisfaction;
Accept H3a: dependence on seller positively influences commitment;
Reject H3b and H3c: dependence on seller does not positively influence trust and relationship satisfaction;
Accept H4a, H4b and H4c: relationship investment positively influences commitment, trust and relationship satisfaction;
Accept H5a, H5b and H5c: seller expertise positively influences commitment, trust and relationship satisfaction;
Reject H6a: communication does not positively influences commitment;
Accept H6b and H6c: communication positively influences trust and relationship satisfaction;
Accept hypothesis H7a, H7b and H7c: similarity positively influences commitment, trust and relationship satisfaction;
Reject H9a, H9b and H9c: interaction frequency does not positively influence commitment, trust and relationship satisfaction;
Reject H10a: manifest conflict does not negatively influence commitment;
Accept H10b and H10b: manifest conflict negatively influences trust and relationship satisfaction.
5.3 Multivariate Outcomes with Moderation Effects
In Table 7 below, the results are presented for the moderated effects by membership dummyvariabele in the multivariate case. The purpose is to demonstrate whether there exists a difference between the customers and members in the relation between antecedents and relationship quality constructs. The moderator variable is a dummy variable for which the members are coded as 1 and customers are coded 0, and this is the membership variable. The premise that the members have a greater average impact on relationship quality constructs in situations in which relationships are more critical to the success of a company is – for most of the constructs - not supported by the estimation results.
In the second column the regression estimates with commitment as dependent variable is shown. In this specification the moderator variable is not significant, however the product of dependence on seller with the moderator is significant at 5% level. The effects of the
dependence on seller on the commitment construct is supported by members, however this effect is relative to customers smaller, which is explained by the negative sign (β = -0.105)
