About NTF
 Career Links
 Hall of Fame
 Industry Links
 News
 Research & Resources
 Scholarships
 Special Events
 Scholarship Winner Resumes

Search this site:

 
   

Relationship Selling and Trust: Antecedents and Outcomes



CHAPTER ONE Systemic Thinking
This chapter reviews the basics of the systems approach. The chapter opens with a short introduction to systems thinking, defines a system and describes the process by which systems analysts define a system. Since system analysts, particularly in interdisciplinary fields, apply a multiple perspectives view (MPV) this chapter reviews the MPV. The chapter closes with principles of systems thinking in learning organizations. Companies are complex, dynamic systems that adapt to changes in their environment. Every complex system is best analyzed from multiple perspectives. Companies today are increasingly becoming interdisciplinary and therefore apply systems thinking more than before. Systemic solutions provide work teams with a way to look at complexity in a way that enables them to see the big picture and the details. Thus, systems thinking enables to see a whole, a critical frame that otherwise would be obscured from our view.
Introduction: Systems Thinking
Systems thinking (ST) is the discipline for seeing wholes. It is a framework for seeing interrelationships among events and actions. This framework facilitates an ability to identify underlying structures, which are patterns of change. Instead of explaining problems, events, and actions by snapshots, details, and events, analysts explain actions problems, and events by underlying structures. ST answers the question: can the set of objects that are pieced together, with reciprocal relations, in fact generate behavior patterns that are being produced by the actual system. Systems thinking is a set of general principles, a body of knowledge and tools that was distilled over the course of the 20th century. Systemic solutions help to make the full patterns clearer and help us see how to change them effectively. Systems thinking spans many fields as diverse as the physical and social sciences, engineering and management. The tools and techniques originated from two threads: feedback concepts of cybernetics and servo-mechanism engineering theory, dating back to the 19th century. Today, successful problem solvers use problem solving processes, ideas, and procedures that are based on systems thinking and have come to be known as the systems approach. Systems thinking is needed because while the web of interdependencies is rapidly tightening, our development of capacity for thinking in terms of dynamic interdependency has not kept pace. Our thinking resembles a laundry list that reflecting our thinking of factors as acting independently and as having fixed weights. As interdependency increases, we must increase our appreciation for the growing systemness of our reality and begin to function as responsible web-mates. As interdependency increases we must learn a new way to learn. We must develop a common language and a framework to share our knowledge. The systems approach provides practitioners with tools that organize complexity into clear ideas, enabling the use of a common professional jargon by practitioners from different disciplines free of context.
The Process of a System Definition
A definition of a system is based on perception (Lendaris, 1986). Perceptions are subjective and systems practitioners should be aware that different people look upon the same situation differently. Each person sees the world through a unique set of perceptual filters based on previous experience, pre-dispositions, cultural influences etc. Knowledge cannot correct the illusion but, awareness of the practitioner of her or his own personal filters and aspects of illusions is crucial. Awareness of the bias will lead practitioners to change the internal processing of sensory data and to thereby achieve different perceptions that will allow practitioners to declare a system. Defining a system, practitioners should ask two major questions (Lendaris, 1986): is there a whole involved and are there parts that operate together to manifest that whole? A system is a) a unit with certain attributes perceived relative to its environment and b) a unit that has the quality that it internally contains subunits, and those subunits operate together to manifest the perceived attributes of the whole. To assert systemness, an explicit observer role is required. Next, the role of the observer is described. Assumptions regarding the critical role of the observer contrast traditional assumptions found in the classical scientific methodology. Traditional assumptions contend that the observer of the system is separated from the observed. In the current literature however, the observed depends on the observer and this is a fundamental dependency (Lendaris, 1986). The observer engages in three levels of perception: the environment, the unit, and the subunit. The environment is the context. It is everything outside the unit that gives the unit meaning in the eyes of the observer. Thus, meaning is dependent upon the focus of the observer and her or his goals in observing the unit. The environment therefore is everything but the unit (e.g., the retail industry), confined to aspects that are relevant to what the observer chooses to focus on (e.g., consumer relations). Thus, the chosen context and the meaning are both subjective, there is no such thing as an independent system. Therefore, the observer who perceives the unit distinguishes it from its perceived environment. It is very important that the systems practitioner keep the context consciously in mind during this process since context is the single most important notion for the systems practitioner. A proper formulation of the context is crucial for successful problem solving. Operational means to define the context is to define a suprasystem relative to the system that is in the focus of the observer (e.g., service). A suprasystem is a system in which the observed unit functions as an element. The observer should operate in a systematic manner moving across perceptions. The observer is perceiving the system in terms of the environment and the unit (Matches part A in the definition, hence stance A). The unit (e.g., a company) has objects outside it and objects inside it, and in that sense is a whole. The whole at stance A however, is the unit in its environment, presenting the notion of wholism, whereas the whole in stance B is the unit and its subunits, presenting the notion of holism. Observing the whole and the environment (A stance), the analyst asks how does the unit relate to its environment. What are the attributes of the unit as they are being perceived in that environment (e.g., position, relationships data bases). After characterizing the environment and identifying the properties of the unit in that environment, the observer perceives the system in terms of the unit and the subunits inside it (Matches part B of the definition, hence, stance B). Observing the unit and its subunits (stance B), the analyst identifies the subunits (e.g., departments in a retail store), describes them in detail, and examines the relationships among them. The observer analyzes how the relationships among subunits allow them to comprise the unit (e.g., standards of service). Thus the observer perceives the subunits operating together as a whole. After identifying the unit and its subunits, the observer focuses on the subunits (e.g., departments) as the unit and identifies its subunits (e.g., SA-consumers dyads). The manifestation of the whole by the subunits ties both stances of observation together. It is only through the coordinated operation of the subunits together, that the whole emerges. The whole is greater than the sum of its parts. There are two factors underlying this claim. First, the observer cannot deduce the properties of the whole by studying properties of the parts, their individual operation, and then adding them up. Second, the parts, when jointly operating according to some organizing principle, do something which is greater than a simple collection of their individual uncoordinated operation. The properties of the parts operate in a coordinated manner and only then contribute together to the emergence of a whole. A declaration of a system requires the observer to take both stance A and stance B. The observer needs simultaneous awareness of the stances combined. Because the observer is biased, and because it is hoped that the observer is aware of the bias, key advice for the observer is to consciously adopt perceiving roles separately both at the horizontal level and at the vertical level of observation (For more on the horizontal and vertical observations levels, turn to Appendix A2), in a systematic manner, thus adopting a multiple perspectives view (MPV). The Multiple Perspectives View and its Purpose Linstone (1984) claims that rational systems analysts use different perspectives than perspectives used by organizations and individuals. These perspectives are underlying paradigms that are used for gathering information, analysis of information, and decision making. The difference in underlying paradigms leads to decision making processes that are based on a single paradigm, creating a chasm between the analysis of the system and decision making regarding that system. To address this chasm, Linstone (1984) presents the MPV. This view addresses complex socio-technical systems in which issues must deal not only with technological and technical issues but also with social and human issues. MPV provides the human mind with strategies for the data gathering, chunking, and abstracting processes that are based on a variety of perceptual filters and inquiry models, rather than one. The chasm is therefore bridged by the MPV leading to a higher quality decision making. MPV does not call to abandon conventional models of analysis, but to create an awareness of the need to step beyond confines of current paradigms and adopt a multiplicity of paradigms called perspectives. Next, a description of the current paradigm and its implication. Decision making processes today are science and technology based. This basis provides us with a sense of confidence, strength and power. Thus, problems solving starts with the assumption that problems undoubtedly can be solved. Yet, often a new solution or technology creates a new problem. The problems are not solved, they only shift. In search for the best solution a reductionism approach is taken. The solution is confined to one possible option rather than to maximum options. Reductionism, reduces the problem, in some of its aspects, and uses statistical inference to examine the truth in the achieved solutions. Problem solvers, from all sciences, rely on models and data as models of inquiry, they quantify information, and guard to maintain objectivity. Reductionism and quantification, however, as much as they contribute to seeing a clear big picture, ignore the individual. These activities also ignore the various movements of time assuming it is linear and that all can be predicted. Modeling activities, each have their emphasis leading to the groupthink phenomenon, to misuse of assumptions, to jumps from simulation to duplication, and to often becoming the end rather than means. Models strive to represent complex systems and seek abstract models to aid problem solvers in their thinking. Two sets of inquiry models that go beyond the confines of the analytic problem solving, are the Hegelian and the Sinergian inquiry systems (Mitroff & Turoff, 1973; Turoff & Mitroff, 1974). (For a further discussion regarding inquiry systems please turn to Appendix A2.) The analytical approach also uses quantification. Quantification leads to drive out qualitative analysis and entails the risk of underestimates or overestimates of probabilities of events. Objectivity is a myth and although problem solvers strive for it, given the human perceptual filter, objectivity does not exist. The unawareness to this myth could be crucial in decision making.
What Should We Be Looking At
Components of socio-technical systems are: technology, physical environment, techno-personal setting, techno-personal setting, organizational actors, political action, and decisions which are all related. Linstone (1984) simplifies the components into three overlapping components: technical, organizational and personal. To deal with all components of the sociotechnical system, all three facets are to be considered. Since the emphasis in Linstone’s view is on how we look rather than what are we looking at, the facets are called perspectives. A perspective is a perceptual filter we use when we look at an issue. To balance the picture of issues at hand, problem solvers should employ multiple perspectives. Each perspective may illuminate any element but the other two perspectives may not be neglected for they contribute important insight. While some people have all three perspectives as their way of seeing the world, others can acquire a MPV. Those who use a MPV succeed in understanding the problem and in designing a good decision, better than those who use one perspective and an unbalanced view. Next, perspectives are briefly discussed. The technical perspective has been the dominant perspective and has an analytical nature. The world is seen by quantitative terms of models, optimization, probability, game theory, cost benefit analysis etc. There is much interest in categorizing and classifying information. While the ‘what’ varies, the ‘how’ never changes. The more reality is complex, the less suitable is the technical perspective. As much as organizations can be different: hierarchical, decentralized, or ad hoc, all humans are socialized into sociocultural organizations and are influenced in their actions and decisions by the organizational perspective. The organizational perspective addresses components that can never be adequately encompassed by the technical perspective: values, culture, myths, interests. Users of this perspective often mistrust analytical tools and view them as uncontrollable. The focus of this perspective is short term. Organizational conflicts and contradictions, lead problem solvers to use this perspective in analysis of organizational politics. The organizational perspective reflects decision making best when opposing views, communication issues, decision makers are at issue. This perspective however, discounts future problems in contrast to short term problems. Thus, problems are fragmented between terms, creating reductionism in this perspective too, but in a different form than that caused by the technical perspective. Short range consequences are given high priority. Rationality of decision makers is always limited. There is a strong tendency to break problems in accordance to organizational responsibilities (reductionism again) and a fear of making errors across levels. Technology is perceived as a potential disruptive force, and there are difficulties in adapting to changing needs. Decisions often depend on the extent of agreement. All these create difficulties in using this perspective. This, however, is the only perspective which provides leverage in dealing with power issues. The organizational perspective is complemented by the personal perspective. The personal perspective is the most difficult to define and most subtle. This perspective involves perceptual filters of individuals: intuition, charisma, leadership, and self interest, which play vital roles in decision making. Politicization and its effect on decisions is explicit. The addition of this perspective to the analysis, brings us to open ourselves to deeper mental levels of behavioral patterns, beliefs, and personalities. These mental levels have potential value in understanding decisions. The difficulties with this perspective derive from ambivalent behavior, a fuzzy decision making process, and difficulties to approach issues objectively. Issues are influenced by background experience, roles, personality structure, and changing environments. The multiple perspectives view encourages dynamic interdependencies among perspectives over time. This interdependency creates a whole. As mentioned before, the discipline of seeing wholes is systems thinking (ST). How Does Systems Thinking Work? Since each element and event is connected within the same pattern and has an influence that is usually hidden from our view, ST enables a systems practitioner to step back far enough and see both the trees and the forest: to see details and patterns leading to the use of fundamental solutions rather than symptomatic short term relieves (Richmond, 1991). Systemic solutions provide a way to look at complexity. There are two kinds of complexity. Complexity for detail and complexity for dynamics. Most systems analysts focus on detailed complexity using simulations to deal with many variables. The engagement in details however, obscures the view of the big picture, of the pattern and interrelationships underlying these variables. ST provides the perspective of looking at the whole. Next, principles of systems thinking. One principle relates to the source of leverage. The real leverage is in understanding dynamic complexity not detailed complexity. Leverage is the ability to be well focused and bring best results from small focused actions. The leverage principle is the bottom line of ST. Leverage lies in the ability to see interrelationships and not cause and effect chains. We see straight lines but reality is made of circles. Thus, even our language distracts us from seeing reality. A new language will have an effect on ST perception. The other principle relates to the practice of ST which starts with understanding the concept of feedback. It shows how actions can counteract or reinforce each other. Thus the idea of feedback can overturn deeply engrained ideas such as causality. It builds learning to recognize the type of structures that recur again and again and how they predict behavior. It simplifies the complexity by enabling to see the deeper patterns that are lying behind the events and the details, so that the focus will not be on symptoms and short term change. Thus, ST organizes the complexity into a coherent story that illuminates the causes of problems and how they can be remedied in enduring ways. One organization which applies systems thinking is the learning organization. The Learning Organization Learning organizations are organizations in which people can continually expand their capacity to create results they truly desire. New and expansive patterns of thinking are nurtured, collective aspirations are set free and people continuously learn how to learn together. In order for the learning organization to move from being invented to being practiced, five components converge. These components are disciplines in human behavior. A discipline is a body of theory and technique that must be mastered to put into practice. The five components of the learning organization together make a whole (A stance). The whole is of systems thinking. Systemic thinking destroys the illusion that the world is created of separate unrelated parts. Only when we give this premise up, can we build learning organizations. The five disciplines develop separately, but are each critical to the others success. From a B stance, the five components of systems thinking are: personal mastery, shared vision, mental models, and team learning (Senge, 1994). Next, each component is described setting a B stance level of looking at the parts mindful of the whole (Holism). Personal mastery is a life long process of learning. Learning is not acquiring information, it is the continuous expansion of the ability to create the truly desired results, the personal vision. It involves both the continuous clarifications of the things that are important to us and the difficult learning to see the current reality more clearly. Only through these two processes, we see and sense a gap between where we are and where we want to be. Therefore, only when one constantly engaged in these two activities, personal mastery becomes a discipline. This gap between the two is a creative tension which sustains our lives. The creative tension may be hard to contain and therefore may lead to frustration and anxiety that are called emotional tension. it is very important to contain this tension and bring the reality to the vision rather than compromise and bring the vision closer to reality. Mental models are deep internal images of how the world works. Mental models can be simple generalizations or complex theories. Mental models are a subjective reality, but their existence is an objective reality, a given. Therefore, the problem does not lie in whether or not they exist, but in the extent to which we are aware of their existence and their affect on our cognition and behavior. Mental models are tacit. If they exist below the level of awareness, they can impede learning. New insights are not translated to practice because they conflict with deep internal images. We are thus limited to familiar ways of thinking and acting. Mental models can accelerate learning when awareness is on the surface. Therefore, learning organizations manage mental models by surfacing, questioning, testing, and improving these internal images of the world. To manage mental models two skills are required: the skill of reflection and the skill of inquiry. An adequate environment must exist for members to feel secure and engage in these activities. In reflection, hidden assumptions are brought to the surface and through exposing the left hand column (what we normally say), leaps (from observation to generalization) of abstractions are recognized. In inquiry, an honest investigation and a balance between inquiry and advocacy, lead to distinctions between espoused theories and theories in use. The distinction between those leads to the change in existing mental models. Thus, these skills must be developed. A shared vision is an inspiring and compelling emotional force that connects people. It is what we want to create. The shared vision provides focus and energy for learning through generative learning. Shared vision evolves from personal visions and is an overreaching goal affecting cognition and actions. It gives people commonality (no our/their mentality) and coherence to diverse activities. Generative learning is meaningless without a shared vision. Shared vision fosters risk taking, long term perspective and experimentation. To develop the shared vision discipline, organizations must encourage personal visions, give up the tradition of shaping the shared vision by top management and telling the vision across organizational levels. Managers should aspire to have members design the vision and be committed to it. Managers are important role models in the commitment process as they share their personal vision and encourage personal mastery of others. Team learning is the fourth discipline. Team learning is the process of team alignment and generative learning. In alignment the team members are not at cross purposes but are all directed and unified to the same goals. Therefore, alignment is a condition for empowering individuals. In learning the team develops the capacity of to create results members truly desire. It builds on the shared vision and on personal mastery. Since many important decisions are made by teams that are interconnected, mastering team learning is more important than it was before. To master team learning skills of dialogue and discussion must be developed and constantly practiced. In dialogue the team engages in a free and creative exploration of complex issues in an environment of openness and in a divergent manner. Reflective openness leads to a deep listening to each other and suspension of one’s own views. In discussion different views are presented and defended but the goal is to decide upon the best solution and a new view in a convergent manner rather than each member defending own view. The above disciplines develop as an ensemble and the fifth discipline of systems thinking integrates them, fusing them into a coherent body of theory and practice. Without the systemic orientation there is no motive to look at how disciplines interrelate. To realize its potential, systems thinking needs the other disciplines. Systems thinking makes the subtlest aspect of the learning organization understandable. People adopt a new way of perceiving themselves and the world. As the five disciplines develop into an ensemble, a shift of mind takes place: A METANOIA. A shift from seeing ourselves as separate from the world, to seeing ourselves as connected to the world. From seeing problems as related to one field and caused by an outside ‘enemy” to seeing ourselves responsible for creating connected problems that we experience.
The Future
Marketing is moving from viewing the marketplace as comprised of separate, independent players that compete against each other, to a systemic view. The marketplace is viewed as a network of related players, with growing partnerships, alliances, and relationships. Systems thinking is viewed as a formula for competitiveness. In competitive companies systemic thinking is starting to predominate internal processes. The field has been heavily impacted by disciplines such as psychology, organizational behavior, sociology, technology, enriching the marketing perspective and contributing to a new whole. A learning organization can learn faster than its competitors, encouraging all levels within the organization to engage in learning and turn everyday practice into organizational knowledge bases. Learning companies examine their mental models, inspire individuals and units to create visions that relate to the organizational vision, and extend the maximum degree of power and authority as far as possible from the top or corporate center. These companies localize. In a world where information surrounding the product is rapidly becoming at least as important as the product itself, intellectual capital is a commercial asset. Next, a review of the basic concepts in marketing, trends in the retail industry, and future trends.
CHAPTER TWO Basic Concepts in Marketing
Chapter two discusses basic concepts of marketing, outlines future trends in the marketplace and in the retail industry, and closes with the retail industry as a learning entity. Marketing starts with human needs and wants. Consumers choose among many products that might satisfy a given need by value and satisfaction. The key, therefore, to achieving organizational goals consists of being more effective than competitors in determining and satisfying needs and wants of target markets. Buyers will buy from a firm that they perceive to offer the highest customer delivered value. To be successful, companies need to look for competitive advantages beyond their own operation into value chains. Often, the adoption of new opportunities requires a change in mental models.
What is Marketing? The Core Concepts
Kotler (1997) defines marketing as a social and managerial process by which individuals and groups obtain what they need and want, through creating, offering, and exchanging products of value in the market. This definition of marketing relies on the following core concepts: needs, wants, demand, products, value, cost, satisfaction, exchange, transactions, relationships, networks, markets, marketers, and prospects. Next, each of these concepts is briefly described. Human Needs and Wants Marketing starts with human needs and wants. A human need is a state of deprivation of some basic satisfaction (Kotler, 1997). Marketers do not create needs. Needs preexist in the market. Wants are desires for specific satisfiers of needs. Although people’s needs are few, their wants are many (Mowen & Miner, 1998). Human wants are continually shaped and reshaped by social forces and institutions such as churches, families, schools, and businesses. Demands are wants for specific products that are backed by an ability and a willingness to purchase them. Effective marketers, therefore, measure not only how many potential customers want the product, but how many potential customers are actually willing and able to pay for it (Kotler, 1997). People satisfy their needs and wants with products. A product is anything that can be offered to satisfy a need or want. Occasionally the terms offering or solution replace the term product. A product or an offering can consist of three components: physical goods, service, and ideas (Kotler, 1997). The importance of physical products lies not so much in owning them, as in obtaining the services they render. Value, Choice, and Satisfaction Consumers choose among many products that might satisfy a given need by value and satisfaction. Value is the consumer’s estimate of the product’s overall capacity to satisfy her or his needs. Products differ from other products by features and price. Features may satisfy the interest of the customer, but when expensive they may actually lower the overall value. Thus, customers choose the product that produces the most value per dollar. Value is the satisfaction of customer requirements at the lowest possible cost of acquisition, ownership, and use (Kotler, 1998).
Exchange and Transactions
Exchange is the act of obtaining a desired product from someone by offering something in return. Five conditions must be satisfied for exchange to exist (Kotler, 1997). First, there are at least two parties. Second, each party has something that might be of value to the other party. Third, each party is capable of communication and delivery. Fourth, each party is free to accept or reject the exchange offer, and lastly, each party believes it is appropriate or desirable to deal with the other party. Whether exchange actually takes place depends upon whether the two parties can agree to terms of exchange that leave both parties better off than they were before the exchange. The exchange is a process rather than an event. Two parties are in a negotiation process. When the parties reach an agreement a transaction takes place. A transaction is a trade of values between two or more parties. A transaction involves several dimensions: at least two things of value, agreed upon conditions, a time of agreement, and a place of agreement. In the most generic sense, companies seek to elicit the behavioral response of buying. To affect successful exchanges, marketers analyze what each party expects to give and get from the transaction. If there is a sufficient match in the wants list of the parties, a basis for a transaction exists. The legal system which arises to support and enforce compliance on the part of the transactor, enables the parties to approach transactions with less distrust. The Market The concept of exchange leads to the concept of a market. A market consists of all the potential customers sharing a particular need or want, who might be willing and able to engage in exchange to satisfy that need or want. Thus, the size of the market depends on the number of people who exhibit a need or a want, have resources that interest others, and are willing and able to offer these resources in exchange for what they want. Economists use the term market to refer to a collection of buyers and sellers who transact over a particular product or product class. Marketers see the buyers as constituting the market and the sellers as constituting the industry. There are different markets: need markets, product markets, demographic or psychographic markets, and geographic markets. Sellers send goods and services, as well as communication to the market. In return they receive money and information. All modern economies abound in markets. Basic markets dynamics is illustrated in Figure 1. Manufacturers go to resource markets, labor markets, money markets, buy resources and turn them into goods and services, and then sell the finished products to intermediaries who sell them to consumers. A variety of factors influences buying behaviors: culture, subculture, social class, reference groups, family, age and stage in the life cycle, occupation, economic circumstances, lifestyle, personality, and motivation. Different theories of motivation (Herzberg, 1966; Maslow, 1954) explain how needs drive people at particular times.
The Concept of Marketing
The concept of markets brings the concept of marketing. Marketing means working with markets to actualize potential exchanges for the purpose of satisfying human needs and wants (Kotler, 1998). Marketers can be sellers or buyers, but in a normal situation, the marketer is a company serving a market in the face of competitors. The company and competitors send their respective products and messages directly and through marketing intermediaries to end users. Their relative effectiveness is influenced by their respective suppliers as well as major forces such as: demographics, economics, physical, technological, political, legal, social, and cultural. Thus, putting all the above elements together, marketing is defined as a social and managerial process by which individuals and groups obtain what they need and want through creating, offering, and exchanging products of value with others. Marketing Management Marketing management is the process of planning and executing the conception, pricing, promotion, and distribution of ideas, goods, and services, to create exchanges that satisfy individual and organizational goals (American Marketing Association, 1999). Marketing management can be applied in any market. Traditionally, however, marketing management has been identified with tasks and personnel dealing with the customer market. Marketing management has the tasks of influencing the level, timing, and composition of demand in a way that will help the organization achieve its objectives. Within marketing planning, managers must make decisions on target markets, market positioning, product development, pricing, distribution channels, communication, and promotion. The Orientation of a Company Toward the Marketplace There are five competing concepts under which companies can choose to conduct their marketing activities: the production concept, the product concept, the sales concept, the marketing concept, and the societal marketing concept (Kotler, 1997). The production concept holds that consumers favor products that are widely available and low in cost. Managers in production oriented organizations concentrate on achieving high production efficiency and wide distribution. This concept holds in at least two situations. First, where the demand for the product exceeds supply and consumers are more interested in obtaining the product than in its fine features. Second, where the product’s cost is high and has to be decreased in order to expand the market. The product concept holds that consumers favor products that offer the most quality, performance, or innovative features. Managers focus their energy on marketing superior products and improving them over time. Product oriented companies often design their products with no customer input (Day, 1990) The sales concept holds that customers, if left alone, will ordinarily not buy enough of the company’s products. The company must therefore adopt an aggressive selling and promotion effort. Most firms that practice the sales concept, have an over capacity. Their aim is to sell what they made rather than to make what the market wants. In modern industrial economies buyers are dominant and sellers have to scramble hard for customers. The sales concept carries high risks. It assumes that customers who are coaxed into buying a product, will like it, will not bad-mouth the company, will not bad-mouth the product if they don’t like it, will not complain, will forget the disappointment, and will buy again. The marketing concept is a business philosophy that challenges the first three concepts (Borch, 1957; Kieth, 1960; McKitterick, 1957). The marketing concept holds that the key to achieving organizational goals consists of being more effective than competitors. Higher effectiveness is feasible through integrating marketing activities toward determining and satisfying needs and wants of target markets. The marketing concept has been expressed in many colorful ways: “meeting needs profitably,” “find wants and fill them,” “love the customer and not the product” (Kotler, 1997). Levitte (1960) draws a perceptive contrast between the sales concept and the marketing concept. While selling focuses on needs of sellers, marketing focuses on needs of buyers. While selling is preoccupied with the need to convert the product into cash, marketing is preoccupied with the idea of satisfying customer needs by means of the product and the whole cluster of things associated with creating, delivering, and consuming it. The marketing concept rests on four pillars: the target market (Mallory, 1993), customer needs (Albrecht, 1992; Hamel & Prahalad, 1994; Sellers, 1989; Zemke, & Connelian, 1993), integrated marketing (Kotler, 1997), and profitability (Bonoma & Clark, 1988; Uttal, 1978). The societal marketing concept holds that the organization’s task is to determine the needs, wants, and interests of target markets and to deliver the desired satisfaction more effectively and efficiently than competitors. The value should be provided in a way that preserves or enhances the consumer’s and society’s well- being (Collins, 1993; Elliot, 1990; Frederick & Webster, 1994). These authors call marketers to build social and ethical considerations into their marketing practices. Marketers must balance the often conflicting criteria of company profits, consumer wants and satisfaction, and public interests. Building Customer Satisfaction Through Quality, Service, and Value Customers are value maximizers (Hamel & Prahalad, 1994). They form expectations of value and act on it. Buyers will buy from the firm that they perceive to offer the highest customer delivered value, defined as the difference between total customer value and total customer cost. Total customer value (TCV) is the bundle of benefits customers expect from a given product or service. Total customer cost (TCC) is the bundle of customers expect to incur in evaluating, obtaining, and using the product or service. Superior value is the ratio of TCV divided by TCC compared between competitors (Hamel & Prahalad, 1994). Sellers, therefore, must assess the total customer value and total customer cost associated with each competitive offer to know how their own offers stack up. Sellers who are at delivered value disadvantage can either try to increase total customer value or decrease total customer cost. Increasing customer value, companies can augment the service, the product, the personnel, or add image benefits (Day, 1990). Buyers’ satisfaction is a function of the product perceived performance and buyer expectations. If the performance falls short of expectations, the customer is dissatisfied. Customer expectations are influenced by past buying experience, friends, family, marketers information and premises. Many companies are aiming at high satisfaction because customers who are highly satisfied are much less ready to switch providers or brands. High satisfaction or delight, creates an emotional affinity with the company or the brand, not just a rational preference. The result is customer loyalty (Berry & Parasuraman, 1991; Kotler, 1997; Riechheld & Sasser, 1990). As Anderson (1992) notes, in one consumer packaged goods company, 44% of those reporting satisfaction subsequently switched brands. Those who reported high satisfaction with the offer’s quality and value were much less ready to switch. Recognizing that high satisfaction leads to high customer loyalty, many companies today are aiming for total customer satisfaction. For customer centered companies, customer satisfaction is both a goal and a marketing tool. These companies form high expectations and deliver performance to match customer expectation. Strong companies develop superior technological capabilities in managing the four core business processes that provide customer value: the new product realization process, the inventory management process, the order to remittance process, and the customer service process. By focusing on these four processes, companies create customer value through the value chain. The value chain (Harmon, 1997; Porter, 1985) identifies strategic activities that create value and cost in a specific business and look for ways to improve them across the company. The firm’s success depends not only on how well each department performs its work, but also on how well the various subunits and their activities are coordinated to operate as a whole. Often, companies achieve sub-optimization (Lendaris, 1986) leading to the maximization of departmental interests and goals rather than to the optimization of the company’s and customers’ interests. The solution to sub-optimization is to place more emphasis on smooth management of core business processes, most of which involve cross-functional inputs and cooperation. To be successful, a firm needs to look for competitive advantages beyond its own operation, into value chains of its suppliers, distributors, and customers. Faced with intense competition, many companies today have partnered with specific suppliers and distributors forming a superior value delivery network. In addition, to improving their relations with their partners in the supply chain, many companies are intent in developing stronger bonds with their ultimate customers. In the past many companies took their customers for granted. Their customers may not have had many alternative sources of supply, or all suppliers were equally deficient in service, or the market was growing so fast, that the company did not worry about satisfying its customers. Businesses today face several major challenges. Advances in technology and telecommunications have brought all the world’s countries together into one global economy. Companies must focus on the customer if they are to achieve success in the global marketplace. Companies must pay close attention to their customer defection rate. Losing profitable customers can dramatically impact a firm’s profits. Also, the cost of attracting a new customer is estimated to be five times the cost of keeping a current customers satisfied. The cost of attracting a new customer is higher than the customer’s life time value for the company (Riechheld & Sasser, 1990). Therefore, one of a marketer’s main tasks is customer retention and the reduction of customer defection rates. Sheth and Parvatiyar (1995) found a few explanations to customer defection: dissatisfaction, disagreement, higher value provided by competitors, social influence, and boredom. Most of the marketing theory and practice focus on the art of attracting new customers rather than retaining existing ones. Traditionally the emphasis has been on selling rather than on caring for the customer afterward. In mature industries, in which differentiation among companies is a challenging task, and in which high quality is a standard, companies have no choice but to implement total quality management programs, if they want to remain solvent and profitable. Unless companies can sign up customers with fewer marketing efforts, spend less per a sales call, stimulate higher new customer annual spending, retain customers longer, or sell customers higher profit products, the company is headed for bankruptcy. There are two ways to strengthen customer retention. One way is to erect high switching barriers. Customers are less inclined to switch to another supplier when the switch would involve high capital costs, high search costs, loss of discounts for loyal customers, etc. (Hamel & Prahalad, 1994). High switching costs, however, would make buyers hesitant to agree on these transaction terms in the first place, reducing the number of acquired customers. The better approach to retain customers is to deliver high customer satisfaction. This makes it harder for competitors to overcome switching barriers by simply offering lower prices or switching inducements. The key to customer retention is relationship marketing - creating strong customer loyalty. Relationship marketing embraces all the steps that a company needs to undertake in order to know and serve their valued individual customers better. Relationship marketing is the practice of building long term satisfying relationships with key parties (e.g., customers, suppliers, distributors) in order to retain their long term preferences and business. Transaction marketing could be viewed as a part of relationship marketing. Good marketers build long term relationships by delivering high quality, good service, and fair prices to other parties over time (McKenna, 1991). Relationship marketing is an application of a systems thinking which results in strong economics, technical, and social ties among the parties. In the next chapter, a comprehensive discussion of relationship marketing. Prior to that discussion, trends driving the marketplace are reviewed. Trends in the Marketplace Marketing is one of the most dynamic fields within the management arena. Companies must respond continuously to new challenges. To meet these challenges new marketing ideas surface. There are eleven emphases in marketing thinking. First, there is a growing emphasis on quality, value, and customer satisfaction. Different buying motivations (e.g., convenience, style, features, service) play a strong role at different times and places. Today’s customers are placing greater weight on quality and value in making their purchase decisions (Anderson & Narus, 1990). The market is comprised of affluent customers demanding personalized superior service, resulting in a highly fragmented market. Second, there is a growing emphasis on relationship building and customer retention (McKenna, 1991). Much marketing theory in the past has focused on how to make a sale without knowing much about the customer and whether she or he will ever buy again. Today’s marketers are focusing on creating life-long customers (Berry, 1995;Gronoos, 1995). The shift is from transaction thinking to relationship building. Companies are now building customer databases containing customer demographics, lifestyles, level of responsiveness to different marketing stimuli, past transactions, and orchestrating their offerings to produce delighted customers who will remain loyal to the company. Third, a growing emphasis on managing business processes and integrating business functions (Kotler, 1997). Today’s companies are shifting their thinking from managing a set semi-independent departments, each with its own logic, to managing a set of fundamental business processes, all of which impact customer service and satisfaction. Companies are assigning cross disciplinary personnel to manage each process. Marketing personnel are increasingly working on cross-disciplinary teams rather than only in the marketing department. This is a positive development that broadens marketers perspectives on the business and gives them a greater opportunity to broaden their perspective from one focal point to a cross department multiple perspectives of workers from other departments. Fourth, a growing emphasis for global thinking and local market planning (Goldman, Nagel, & Preiss, 1994). Companies are increasingly pursuing markets beyond their borders. As they enter these markets, marketers must drop their traditional assumptions about market behavior and adapt their offerings to other countries cultural prerequisites. They must be aware of the local economic, political, legal, and social realities facing the firm. Companies must think globally adopting an A stance view, but plan and act locally practicing a B stance view. Fifth, there is a growing emphasis on building strategic alliances and networks (Hamel & Prahalad, 1994). As companies globalize, they realize that no matter how large they are, they lack the total resources and requisites for success. Viewing the complete supply chain for producing value, marketers adopt systems thinking and recognize the necessity of partnering with other organizations who supply different requirements for success. Those marketers give up the premise of separate independent parts in their world. Sixth, there is a growing emphasis on direct and on- line marketing (Toffler & Toffler, 1994). The information and communication revolution promises to change the nature of buying and selling (Toffler, 1994). People anywhere in the world can access the internet and company home pages to scan offers and order goods. Via on-line services, they can give and get advice on products and services by chatting with other users, determine the best values, place orders, and get next day delivery. As a result of the advances in database technology, companies can do more direct marketing and rely less on wholesale and retail intermediaries. Beyond this, much of companies buying is now done automatically through electronic data interchange links among companies (Hamel & Prahalad, 1994). These trends portend greater buying and selling efficiency. Seventh, growing emphasis on service marketing (Kotler, 1997). The U.S. population today consists of only 2.5% farmers and about 15% factory workers. Most people are doing service work: field sales people, retailers, craft people, and knowledge workers such as physicians, engineers, accountants, and lawyers. The service industry captures 80% of the market and is expected to grow further. Because services are intangible, perishable, variable, and inseparable, they pose additional challenges that are not found in tangible-goods marketing. Marketers are increasingly developing strategies for service firms. These strategies reflect a move from economical exchanges to economical and social exchanges (Foa, Converse, Tornblom, & Foa, 1993) . Eighth, there is a growing emphasis on high tech industries (Toffler & Toffler, 1994). Much economic growth is due to the emergence of high tech firms, which differ from traditional firms. High-tech firms face higher risks, slower product acceptance, shorter product life cycles, and faster technology obsolescence. Ninth, there is a growing emphasis on ethical marketing behavior (Kotler, 1997). The general public is wary of ads and sales approaches that distort product benefits. Marketers must hold to high standards in practicing their craft. Tenth, most marketing processes today are controlled and measured using technological systems that provide abundant data. The data is partially processes into information which serves to improve processes. Eleventh, processes are more systemic (Senge, 1994). Management is much aware of the positive affects of interactions among units, people, processes, relationships among components on customer service, retention, and employee commitment (McKenzie & Fetter, 1996). Cross functional service and departmental interface are growing. Adopting these trends, companies can form linkages and feedback loops among underlying processes. This will enable companies to be proactive, informed, and systemic. Thus, companies will be able to either increase market share, increase business share, or maintain their current market position competitively. Figure 2 presents trends and their affect on processes.
PROACTIVE Trends in the Retail Industry
Retailing includes all the activities involved in selling goods or services directly to final consumers for their personal non business use. A store whose sales volume comes primarily from retailing is defined as a retail store (Kotler, 1997). Retail stores offer one of four service levels: self service, self selection, where customers serve themselves, but can also ask for help, limited service retailing, and full service retailing. The strongest trend in the retail industry is the emergence of new retail forms (i.e. delivering banking services to the office or the home) which threaten established forms of retail (Nelson-Forsyth, 1993). Second, as a result, retail life cycles are shortened. New forms are duplicated and rapidly lose their novelty. Third, the electronic age has significantly increased the growth of non store retailing (Eldridge, 1993). Consumers receive sales offers over their television set, computer, telephone, to which they can immediately respond calling a toll free number, usually 24 hours a day, or via computers. Fourth, competition is increasing today between and among types of outlets (Coughlan & Vilcassim, 1994). Discount stores, catalog showrooms, and department stores all compete on the same market (Web-Fressler, 1995). The arrival of superstores, with bulk buying power, favorable terms, increased squared footage, and enhanced amenities for customers, have forced nearby independent stores out. Fifth, small store branches, independent stores, are finding that the size can be an advantage for a personal touch and special niches for a loyal customer base (Tearcy & Wiersema, 1993). Sixth, increasing inter type competition has produced retailers positioning themselves on extreme ends of the number of product lines carried. High profitability and growth have been achieved both by mass merchandise and specialty stores. Thus, superpower retail stores are emerging (Rickard, 1995). Through superior information systems and immense purchasing power, those retailers can offer consumers strong price savings. Department stores which used to be a one-stop shopping convenience, gave way to small stores in malls. Specialty stores within malls are becoming increasingly competitive with department stores (Stern & El- Ansary, 1996). Some retailers are combining groceries with a huge selection of non-food merchandise, and even specialty items (e.g., Fred Meyer) pushing out malls. Seventh, marketing channels are becoming professionally managed and programmed (Koselka, 1992). Large retail stores extend their control over marketing channels. Seventh, retail stores are moving into a mix of business, launching new store formats targeted at different lifestyle markets. Eighth, retail technology is growing in importance. Retail technology is becoming critical to competitiveness (Vloski, Smith, & Wilson, 1994). Computers are used to produce better forecasts, control inventory costs, order electronically from suppliers, send electronic mail between stores and sell to customers within stores. They are adopting checkout scanning systems, electronic funds transfer, electronic data interchange, in store television, traffic scanner, and merchandise handling systems. (Vlosky, Smith & Wilson, 1994). Ninth, a global expansion into other countries (Rapoport & Martin, 1995). Retail stores with unique brand positioning (The Gap, Toys R Us) are becoming globally prominent. Due to the maturity and saturation of markets in the US, marketers are pursuing overseas markets to boost profits. Retailing in America has never been more competitive and there is much to be learned from companies that excel in fierce competition. In a study about the common traits of successful retailers (Berry, Seider, & Gresham, 1997), although the sample companies differed in size, scope, and nature of their business, they demonstrated similar traits. These common traits were found to fuel sustainable marketplace achievements and create a perpetual state of readiness for the retailing revolution. The traits are rooted in the companies’ respective values driven cultures, forming another common success factor. In each store among the sample companies, a set of core values permeates the organization and both motivates and guides the company’s dynamism. While each company’s value system has distinct properties, all revolve around creating a superior shopping experience and compelling value for customers. Services represented a radical departure from conventional practices. The sample companies were also highly focused on strategies and on their effective execution, on credibility as merchants in their customer’s eyes, and committed to products and retail concepts. These companies also seek to strengthen credibility with suppliers and customers. Speed of service and making the store a fun place to shop, have become competitive tools. Lastly, in all the companies in the sample, employee and community investments were considered important extensions of customer treatment (Berry, Sieder, & Gresham, 1997). As a learning organization, for the retail industry, the growing importance of knowledge and knowledgeable workers, means more resources are being extracted to drive the everyday business. Manufacturers are continuing to adopt a team approach and brokers are continuing their efforts to use a multi-function team approach to manage distributor-supplier interface. All of these changes reflect a change in mental models. This change is based on a bottom up theory of learning. This theory argues that those closest to the activity are in the best position to see the trees and the forest, to best learn how to learn, to be leaders, and to assess what is best needed in the industry. Appendix B presents trends in specific retail segments, the travel industry and the professional hair care industry
CHAPTER THREE Relationship Marketing
Chapter three focuses on the strategic trend of relationship marketing (RM). The chapter opens with a description of RM, discusses its dimensions, its development, and key success factors. The final section deals with managerial and research implications of RM. Faced with intense competition many companies today have partnered with specific suppliers and distributors, forming a superior value delivery chain. To improve their relations with their partners in the supply chain, many companies develop stronger bonds with their consumers. Relationship marketing is a key to retain satisfied consumers and to provide them with added value through the value chain. This chapter describes the relationship marketing domain, its development, types of relationships, key success factors, and managerial implications.
Relationship Marketing: Definition and Description
Relationship marketing is a consumer centered approach seeking long term relationships with existing as well as with prospective partners to the marketing process i.e., suppliers, alliances, competitors, distributors, employees, and consumers (Evans & Laskin, 1994; McKenna, 1991; Aijo, 1996). Since not all relationship marketing instances have customers as exchange partners (i.e., Government), Morgan and Hunt (1994) defined relationship marketing as attracting, developing and retaining relationships. Thus, relationship marketing is the creation of a network with close relationships of parties who take part in the marketing process and work together towards understanding consumer needs, treating the customers as a business partner, ensuring that employees satisfy consumers wants, and providing the consumer with the best quality (Evans & Laskin, 1994). Relationship Marketing: Where Were We, Where are We and Where are we Going Marketing in the past century is cyclical. In early decades of this century, merchants closely knew every consumer, her or his tastes, needs, and could even forecast needs of single consumers and their families. With the industrial revolution and urban movement, communities grew in size and merchants in small stores were gradually replaced by mass marketing and long distribution chains delivering products to consumers. Towards the year 2000, companies are adopting the early mindset and seek direct relationship with consumers. Each consumer as a unique individual with different needs and value perceptions. The emergence of relationship marketing is attributed to several developments in marketing that resulted from major paradigm shifts over the past three decades. The early marketing paradigm was based on the microeconomics model which emphasized profit maximization, demand analysis, breakeven points, and margins. In practice, centralized brand management systems were at the core of marketing organizations. Since the goal was profit maximization, large, hierarchical, integrated organizations, with segregated marketing departments as extensions of sales departments, were a common structure. This centralization enabled the development of specialized expertise, allowed economies of scale and tighter control of brands and sales effort in the national market. The major task of marketing departments was to understand the marketplace, produce required products, generate demand for production, use mass marketing to create consumer demand, and manage distribution channels through which products flow to the market. Global competitors introduced products with higher quality yet lower prices, telecommunications and growing information processing erased the importance of the country of origin and geographical distances in the marketplace. Since the post depression times in the 1930’s the market has become increasingly affluent. Affluence brings demand for quality and intense marketing fragmentation. Consequently, product assortment and service variety change quickly in a continuous manner. This new reality brought deregulation, slow growth, unlimited customer choice, high competition, and high pace of change in a complex environment. Survival and success in this new reality required lower costs, improved efficiency, higher quality, higher service quality, and speeded innovation cycles. Strategic responses to this reality were flexibility, specialization, standartization, information, and customization. Consequently, the old organizational structures became disfunctional, leading to inflexibility, complexity, unresponsiveness and delays. Reorganization and restructuring were crucial for survival. To be flexible, innovative, and responsive, firms needed a structure that moves them closer to customers. Analyzing marketing structures of global competitors, particularly Japanese competitors, marketers identified two prototype models. One model emerged from the Total Quality Management philosophy and stressed that quality sells better and for less. Total Quality Management encouraged a move away from a focus on transactional competition, where many suppliers are invited to submit bids and competition among them is maximized, towards a focus on long term contractual commitment, where companies work with a few suppliers (Aijo, 1996; Sheth & Sharma, 1997; Webster, 1990, 1992). Contractual commitment with a few suppliers reduces uncertainty and cycle times, enables to track units and determine their profitability, provides higher information exchange and customer service, and increases competitiveness due to locking-in suppliers and providing better offerings to customers (Smith, 1997). A second model was that of complex groupings of firms with inter-linked ownership and trading relations. This structure introduced organizations which are not hierarchical nor decentralized but are bound together in a long term relationship providing support, reciprocity and information sharing. The new reality and these functional models propelled a change in the underlying paradigm. Instead of decreasing cooperation under circumstances of increasing competition, the new underlying paradigm suggests increasing cooperation under circumstances of increasing competition. Creation of networks provides partners with: information regarding customer needs and changes in value perception, ability to react quickly, and shared resources. As the new paradigm expanded, the microeconomics model was replaced by an interdisciplinary model that emphasizes the understanding of relationships, processes of cooperation, coordination and negotiations that define the relationships (Webster, 1990). Figure 3 presents the paradigm shift. During the shift in paradigms marketing organizations reorganized and pyramids decentralized. With the era of relational marketing, marketing business units have turned into wheels with spokes representing knowledge links between the core organizational unit and strategic partners. Boundaries between organizations and their environment became blurred. Figure 4 presents new structures of marketing organizations. Next, additional managerial and research implications of relationship marketing are discussed.
Marketing
The types of relationships firms have with their network partners is perceived to be the next generation of competitive advantage. Effective relationship marketing will lead to a higher percent of customer satisfaction, to greater customer loyalty, to a better image of the firm as a quality provider, and to increased profits (Evans & Laskin, 1997). Several types of relationships are evident. These relationships could be located on a continuum from a single transaction to networks. One extreme on relationships continuum is the single transaction. In a single transaction the exchange with partners is not based on prior or subsequent interaction. There may not be a brand name, credit extension, loyalty, but there is some preference that differentiates one provider from another. On the other extreme on the relationships continuum are networks. In networks, there is a confederation, a loose coalition with a marketing function that is responsible for keeping all of the partners focused on the customer and informed about competing product offerings and changes in customized needs and expectations. Relationships in-between the two extremes are: repeat transactions, long term relationships, buyer seller partnerships, strategic alliances, and joint ventures. Next, each type of relationship in the relationship marketing domain is briefly reviewed. Figure 5 presents relationships by dimensions of scope and depth. In repeat transactions, repeat sales are important but are usually not perceived by the company as a meaningful long term relationship with consumers. There is however, a presence of a brand loyalty. Rudiments of trust and credibility are present and could serve as a foundation of a relationship. If products require service or warranty there is an ongoing relationship after sale, but even that responsibility may be a source of conflict. In long term relationships the relationship is based on contractual commitment. Many suppliers are invited to submit bids, the prices are determined on negotiations which are based on mutual dependence, quality, delivery and technical support (Sheth & Sharma, 1997; Webster, 1992) In buyer-seller partnerships, the company and a few suppliers have a long term supportive relationship. Suppliers are brought in at an early stage to help with design, manufacturing, delivery, and service policy decisions. Suppliers are partners in the development problems of the firm (Evans & Laskin,1994; Han et al., 1993; Sheth & Sharma, 1997). In strategic alliances the relationship is intended to move each party towards the achievements of some long term strategic goal and improve its strategic position. Strategic alliances are key in developing a company’s competitive advantage. Beyond shared objectives the parties commit resources. This type of relationship involves partnerships with customers, resellers, and competitors. One special case of a strategic alliance is a joint venture in which a separate entity is created with its own capital structure as well as with shared resources of the founders. Networks are perceived as the next generation of relationship marketing (Aijo, 1996). Networks differ from strategic alliances in that they are not based on individual agreements and collaborations between partners, but rather on multiple joint ventures and multiple strategic alliances. Thus, networks are complex multi- faceted organization (Aijo, 1996; Gronoos, 1994; Webster, 1992). Lastly, in relationship selling one-on one relationships with existing customers are managed. The relationship itself is the added value for consumers. Relationship selling is best applied in service environments, where the emphasis on customization, personalization, and superior quality service, align with the marketing strategy and the market. Relationships selling is based on personal selling. Personal selling is an ancient art. Three major aspects of personal selling are: sales professionalism, negotiation, and relationship marketing. In sales professionalism and in negotiation the sales associate’s task is to identify customer needs and come up with sound product solution. Customer needs analysis skills are critical. This approach assumes that customers have latent needs that constitute opportunities, that they appreciate constructive suggestions, and that they are loyal to sales associates who have their interest at heart (Anderson & Rosenblum, 1992; Paley, 1994). These principles of personal selling and negotiation are transaction oriented because they aim to help sales associates close deals. Notwithstanding, marketers are seeking accounts not transactions. Thus, to serve accounts while providing superior added value over time, companies need to develop relationships with consumers. Relationship selling has an underlying consumer orientation which focuses on problem solving for customers. Relationship selling is built on the premise that important accounts need focused, continuous attention that could only be provided when sales associates manage customers not products. Relationship marketing however, is not effective in all situations. Jackson (1991) claims that relationship marketing is not effective when the customer’s time horizon is short and when switching costs are low. Jackson, uses this criteria to link transaction selling and relational selling to industries. Anderson and Narus (1990), claim that it is not so much the issue of the industry, but the issue of customer wishes, and the extent to which the customer is willing to pay for premium service. To enhance customer loyalty, companies must use data bases which will build customer’s interests, preferences, and through analysis turn them into valuable data for life long sales. Relationship Marketing: The Relationship Development Process There are two theories of relationship marketing development in the literature. Dwyer et al. (1987) present a five phase development model and Wilson (1995) presents a similar four phases model. Next, the models are briefly reviewed. The first stage in Dwyer et al.’s (1987) model is awareness. In this stage a party recognizes feasible exchange parties. Any interaction in this stage may be the beginning of the next stage. The second stage is the exploration stage in which the parties consider obligations, benefits, and burdens. Trial exchanges are made. This is a very fragile stage and therefore minimum investment is made and termination of the relationship is simple. As the parties feel that the interaction rewards them of outcomes in excess of the cost and in excess of the minimum level outcome attraction increases. The gratification due to the association among partners increases due to similarity of beliefs and values. If the relationship is to survive, disclosure must be reciprocated. This stage of the relationship development determines norms. Understanding expectations for cooperation and planning depends on trust. When trust is established the parties will take high risk actions. The third stage is a stage of expansion. Risk is increased and the range of mutual dependence and motivation increases. The probability of alternatives is reduced. The next phase is commitment in which a level of satisfaction from the exchange is achieved. Commitment is measured by inputs, durability, and consistency. The last development stage is dissolution. If the relationship did not proceed to and from the exploration stage, the dissolution begins with one party evaluating its dissatisfaction concluding that continuation costs outweigh benefits. Critics of this model claim that the model lacks detail and operationalization of variables to empirically examine the theory. The second model theorized by Wilson (1995) presents the first stage as search and selection. Since partners are untested, performance reputation and trustworthiness become the measures. In early stages interaction creates a social bonding process. Mutual trust is develops despite high uncertainty. In the second phase of the relationship development, the parties define the purpose by designing an agreed upon set of mutual goals and objectives. In the absence of shared goals and objectives, a failure will occur. The relationship needs to reach a business friendship level . Actions of the parties will define the level of trust and shape the future. In the third stage boundaries are defined. The definition relates to the degree to which each party penetrates the other organization and achieves joint action. Appropriate resources to complete the task are required. When each partner calls upon resources of the other the adaptation to a structural bond begins. The fourth and last stage is a stage of creating value. The relationship enhances competitiveness of parties. Shared values will result in a balanced relationship. Extracting values will determine the extent of trust and cooperation. In both models relationships develop based on information accumulated during the exploration stage. Communication in both models is a key for trust development but Wilson (1995) highlights friendship and personal compatibility as necessary for a successful bonding relationship to deepen. In both models once the structural bond is stable parties can gain benefits. This may indicate that a stable structural bond signals cooperation and trust among the parties. Relationship Marketing: Key Success Factors Under the relational paradigm longer duration and greater depth signal that the firm manages relationships and not just deals with them. Effective management of relationships is required in order to be closer to the customer and translate the closeness to a competitive advantage. Several indicators of effective networks and channels relationships were studied in past research. Trust was the a core construct explained by communication, previous experience, and cooperation. Communication was found to be an exogenous variable that explained the variance from both perspectives of the manufacturer and the distributor. (Anderson & Narus, 1990). Trust and cooperation were reciprocal with cooperation leading to trust and trust leading to further cooperation. Influence was determined by relative dependence and the perception of dependence of both parties. Examining effective relationships in strategic alliances, Smith (1997) found trust, open communication, and perceived interdependence with goal compatibility, as the most important predictors of success. Open communication was the strongest predictor of trust. Interdependency and goal compatibility resulted in high commitment and reflected cooperation. Han et al. (1993) studied relationships in strategic alliances and supplier relationships and found trust to be the strongest predictor, followed by commitment, cooperation, and open communication. Relationship Marketing: Managerial Implications To be more effective firms need to integrate relationship marketing processes into the strategic planning so that resources are well managed and future needs are met. The process of relationship marketing should be conducted systemically, incorporating needs of all parties at all organizational levels: the corporate, the business unit, and the functions. Relationship marketing should be implemented through the organizational culture supported by the top management. Next, implications are presented by three organizational levels: the corporate level, the business unit level, and the operational level. Implementation of RM at the Corporate Strategy Level At the corporate level the scope of the business is defined. Values concerning customer centrality are shaped. Decisions concerning the engagement in relationships and dependence on long term partners are made. At this level marketers analyze customer needs, examine competitive offerings, and assess attractiveness. The unique role of negotiating strategies with partners may be crucial in assuring flexibility. The core competency of relational marketing is the ability to design, manage, and control partnerships (Aijo, 1996; Webster, 1992). Competitiveness also depends on providing a customer perceived value that matches the firm’s position in the value chain. Ravald and Gronoos (1996) claim that organizations applying relational marketing must be market driven. In a market driven framework companies add quality for which customers are willing to pay. Instead of adding extras that are engineering driven features, and supportive services, added value should be determined by customer wants and needs taking in account the decision maker needs, awareness, motivations and the purchase process (Mowen & Miner, 1998). Since consumers are sensitive to a loss more that they are to a gain, added value should add benefits by reducing sacrifice. Relationships provide safety, credibility, stability, confidence, reduce uncertainty and perceived risk. By reducing psychological costs related to the lack of these benefits that relationships provide, the firm reduces customer perceived sacrifice and encourages customer loyalty. Value can also be added by reducing cost through increasing convenience, delivering goods, etc. Ravald and Gronoos (1996) found that perceived value based on the core business was fundamental but was not the ultimate purchase reason. The kind of a relationship a company is perceived to maintain shaped the total perceived value for consumers. Relationships create a dependency of the firm on the success of its relationship partners, posing greater risks of cost efficiency, delivery times, image, and therefore may have negative implications in case of distrust and inadequate cooperation. Implementation of RM at the Business Unit Strategy Level At the business unit level processes of segmentation, targeting, and positioning of the unit takes place. A careful analysis of the market, competitors, customers, firm’s resources and skills to compete, has to be conducted for different segments. Marketers in relational marketing organization need to decide which functions are to be performed by strategic alliances, which functions are to be out-sourced, and which functions are to be internally performed. The segmentation of the market in relational marketing combines differentiated segmentation with advantages of mass marketing. A network organization can identify and operate in super segments rather than in isolated markets. Segments are grouped into super segments based on synergies of materials, facilities, distribution channels, etc. Within the super segments, the use of direct marketing will enable firms to break the market into customer groups and offer a variety of products to distinct groups of buyers who have differentiated needs and responses. The possible negative implications of relationship marketing at this level are break ups of the existing relationship structures that customers currently have with their suppliers. Also, potential opportunities that a firm may have will not be pursued due to its commitment to the network. Implementation of RM at the Operational Function Level At the operational level the marketing strategy is implemented through the marketing mix. The optimization of the marketing mix, however, is no longer the ultimate marketer goal. The marketing mix and product development cycles are too narrow to include all role dimensions in a firm that practices relational marketing. The marketing mix activity is not based on direct communication with consumers, limiting the extent to which the mix is market driven. Also, the marketing mix activity does not hold a dominant perspective of creating and distributing values among partners and consumers. Nevertheless, each element in the marketing mix is affected by relational strategies. Products are market driven and therefore their development is based on information systems and close relationships with customers, enabling response measurement for every customer and defection rates, rather than on surveys that examine general satisfaction of customers. Distribution is based on alliances and partnerships linked to the company through telecommunication. Adequate processing of information enables to integrate the information from the network, be continuously informed, coordinated and responsive. Using data bases, companies can access information regarding customer purchases, purchase frequency, habits, preferences, that connect between manufacturers and customers needless of middlemen. Data mining and other techniques to process information are being introduced to the market. Decision makers are faced with a revolutionary amount of data and information, but little ability to use and process all the information. A future challenge is to build and use systems that integrate and process network information. Pricing combines value pricing with negotiations based pricing, rather than break even analyses plus markups. Promotion through mass media is less effective and is therefore becoming personal and targeted. Sales are broadly defined with responsibilities for relationships management. The use of information technology provides constant, close, location free communication with customers. The use of databases enables direct marketing through telemarketing, direct response radio and television, electronic shopping, customized catalogs etc. The sales are ordered directly by targeted customers. Advertisement is through cost effective media that allows two-way communication, convenience, and measurement of results. The readiness of customers that buy through interactive marketing systems, develops in a later stage. Therefore, advertisement may be longer in direct marketing so that customer awareness is maximized, both rationally and emotionally (Mowen & Miner, 1998) and the firm can trigger a higher level of customer readiness to purchase. Promotion costs in network organizations may be reduced per unit due to shared cost and economies of scale. Public relations will focus on greater reputation due to the image of the network in addition to those of each partner. Relationship Marketing Research Shifts in marketing research were parallel to the shifts in the market. Along with the paradigm shift, research on economies of scale has become absolete. Research on single components such as the reduction in time cycles, the maximization of inventory (EDI), and the improvement of quick responses (JIT) has become less relevant since the application of these systems in processes. The current research deals with networks and examines effective linkages within and among components of the network: customers, employees, distributors, alliances, technology, value concepts, customer and their effect on customer satisfaction, customer loyalty and performance. Organizational behavior research influences channel research in topics of dependence, trust, conflict, and power. Information systems research has brought abundant knowledge concerning technologies and their use. Information processing, however, is still at a dissemination stage in firms. Organizational behavior, information systems, systemic approaches will be integrated strongly in the future into the current marketing research, reflecting the multidisciplinary nature of the current marketing model.
CHAPTER FOUR Trust
Chapter four discusses definitions of trust, integrates divergent perspectives of those definitions, and reviews the evolution of different types of trust. Among the types of trust this chapter overviews interpersonal trust, initial trust, trust as conditional and unconditional, trust as a dichotomy, and institutional trust. Lastly, this chapter reviews the measurement of trust. Trust is an essential component in any relationship, interpersonal or in the marketplace. There is no substitute for this critical element. The core concept of trust is an agreed upon concept. Notwithstanding, since trust theory has developed from different disciplines, it is fragmented. The most recent studies of trust, however, reflect a scholarly interest in the multidimensionality of trust and in broadening the study across disciplines, building a whole out of the fragmented theory.
The Study of Trust
The study of trust may be categorized based on how trust is viewed (Lewicki & Bunker, 1995). Trust may be viewed as an individual characteristic, as a characteristic of interpersonal transactions, and as an institutional phenomenon. According to these authors specific disciplines have been historically associated with each of these approaches. Personality psychologists have defined trust as an individual characteristic (Rotter, 1971, 1980). Social psychologists have defined trust as an expectation about the behavior of another in transactions, focusing on the contextual factors that enhance or inhibit the development and maintenance of trust (Lewicki & Bunker, 1995). Economists and sociologists have been interested in how institutions and incentives are created to reduce the anxiety and uncertainty, and thus increase trust associated with transactions among relative strangers (Goffman, 1971; Zucker, 1986). By concentrating only on specific aspects of the trust concept, each of the different disciplines provides only a partial description of what would be recognized as trust. As a result the description of the different mechanisms to form and enhanced trust in specific transactions and in society in general, are also incomplete (Battacharya & Pillutla, 1998; Linstone, 1984). The most recent trust literature reflects scholarly interest in studying the multi dimensionality of trust and in broadening the study across disciplines, building a whole out of a fragmented theory (Senge, 1994). Definitions of Trust The phenomenon of trust has been researched from different perceptual filters: economics, social psychology, developmental psychology, political sciences and business. Economists tend to view trust as either calculative (Williamson, 1993) or institutional (North, 1990). Psychologists commonly frame their assessment of trust in terms of attributes of trustors and trustees and focus upon a host of internal cognitions that personal attributes yield (Rotter, 1967; Tyler, 1990). Sociologists often find trust in socially embedded properties of relationships among people (Granoveter, 1985) or institutions (Zucker, 1986). Rotter (1971) defines trust as a generalized expectation held by an individual or a group, that the word, promise, verbal, or written statement of another individual or group, can be relied on. This definition is close to that which is found in the Oxford English Dictionary: “confidence in or reliance on some quality or attribute of a person or a thing, or the truth of a statement.” In contrast to Rotter’s “ generalized expectancy” which is a relatively stable personality characteristic, social psychologists view trust as an expectation that is specific to a transaction and to a person with whom one is transacting. Like Rotter, (1971), Lewicki and Bunker (1995) define trust as a general expectation of one person that the promises of the other person can be consistently relied upon. Lewis and Wiegert (1985) define trust as a psychological event within the individual (Lewis & Wiegert, 1985). These authors further claim that trust is a functional prerequisite for the possibility of society, reducing uncertainty and replacing predictability with faith in situations of risk that otherwise may bring individuals to be paralyzed with fear in chaotic societies. It is an implicit contract (Lewicki & Bunker, 1995) in a social exchange (Clarck & Millis, 1979). Social psychologists define interpersonal trust as one’s confidence in the words, actions, and decisions of the other. It is manifested by one’s willingness to be vulnerable and act by that confidence in risk entailing situations (Mayer, Davis, & Schoorman, 1995; McAllister, 1995; Morgan & Hunt, 1994; Rempel, Holmes, & Zanna, 1985). Mayer, Davis and Schoorman (1995) define trust as the willingness of one party to be vulnerable to the actions of another party based on the expectations that the other party will perform a particular action which is important to the trustor, irrespective of the ability to monitor and control the party (1995: 712). Economists (except Williamson, 1993) have no such restriction on their definitions of trust. Their view is that trust follows from the ability to structure contracts or rewards and punishments so that individuals may be inherently trustworthy. Although the implicit assumption is that individuals are not inherently trustworthy in most cases, economists are concerned with the costs and benefits of specific behaviors. It is not a requirement of economic models that all people are trustworthy. Kreps (1990) and Dasgupta (1988) provide the most well known examples of the economic modeling of trust and represent this approach. In their models trust serves less as an inherent concept and more as a label describing such as equilibrium behavioral outcomes not to cheat one’s opponent or partner. Regardless of the underlying discipline of the author, from psychology and micro-organizational behavior (Lewicki, McAllister, & Bies, 1998; Mishra & Spreitzer, 1998) to strategy and economics (Bhattacharya, 1998; Devinney, 1998; Pillutla, 1998), confident expectations and a willingness to be vulnerable are critical components of all definitions of trust reflected in the recent literature (Cummings, & Chervany, 1998; Jones & George, 1998; McKnight, Mishra & Spreitzer, 1998). The most frequently cited definition is “willingness to be vulnerable” which was proposed by Mayer, Davis, and Schoorman, (1995). Thus, although each perceptual filter has examined trust from a different angle, trust is fundamentally, an agreed upon concept. Other authors who define trust say the same thing with different words “willingness to rely on another” (Doney, Cannon, & Mullen, 1998) and “confident positive expectations” (Lewicki et al., 1998). A few authors define trust as positive expectations of others (Das & Teng, 1998; Elangovan, 1998; Hagen & Choe 1998). Trust is also viewed as a positive attitude toward others (Whitener, Brodt, Korsguard, & Werner, 1998). Bigley and Pearce (1998), who argue that an overarching definition of trust does not exist, characterize trust in a very similar way to its definition by other authors. Trust is defined by these authors as “vulnerability,” “perception,” and “preconscious expectations.”
Trust Definitions: Integrating Diverging Perspective
Three attempts have been made to integrate definitions of trust (Barney & Hansen, 1994; Lewicki & Bunker, 1995; Rousseau et al., 1998). Rousseau, Sitkin, Burt, and Camerer (1998) integrated the different theoretical perspectives by introducing the following widely held definition of trust: “a psychological state comprising the intention to accept vulnerability based upon positive expectations of the intentions or behavior of another.” Barney and Hansen (1994) borrowed the terminology from the financial literature, and suggest three kinds of trust: weak form, semi form, and strong form. They define trust as: “the mutual confidence that no party to an exchange will exploit another’s vulnerability.” Weak trust refers to relationships where the parties in the relationship have no vulnerabilities that can be exploited. In that case there is no need to set up contractual agreements between the parties to form trust. In case that there is an opportunity to exploit existing vulnerabilities, requiring contractual and governing mechanisms, it is called a semi trust. In its strong form trust emerges from significant exchanges. Vulnerabilities exist independent of whether or not elaborate social or economic governance mechanisms exist. This is because opportunistic behavior would violate values that are internalized by the parties to the exchange. Barney and Hansen’s (1994) model suggests that both economic models which disregard individual differences on the propensity to be trustworthy, and psychological models which are preoccupied with individual differences regardless of situational factors, are better explanations of trust when they are combined than when they are individual. Lewicki and Bunker (1995) also integrated the theories by defining trust as: “a state involving confident, positive, expectations about another’s motives regarding oneself in situations of risk.” These expectations may be based on rewards or punishments that guide the behavior, on the predictability of other’s behavior, or on a full internalization of the other’s intentions and desires. Thus, in integrating the theories, Lewicki and Bunker’s categories depend on the source from which expectations arise. Another approach to integrating divergent perspectives lies in conditions of trust. Across disciplines there is agreement on the conditions that must exist for trust to arise. Risk is one condition considered essential in psychological, sociological, and economical conceptualizations of trust (Coleman, 1990; Rotter, 1967; Williamson, 1993). Risk is the perceived probability of a loss as interpreted by a decision maker (Chiles & McMackin, 1996; MacCrimmon & Wehrung, 1986). The path dependent connection between trust and risk taking arises from a reciprocal relationship (Rousseau et al., 1998). Risk creates an opportunity for trust, which leads to risk taking. Moreover, risk taking buttresses a sense of trust when the expected behavior materializes (Coleman, 1990; Das & Teng, 1998). Trust would not be needed if actions were taken with complete certainty and no risk (Lewis & Wiegert, 1985). The second condition of trust is interdependence, where the interests of one party cannot be achieved without reliance upon another. Although risk and interdependence are needed for trust to emerge, the nature of the risk and trust changes as interdependence increases (Sheppard & Sherman, 1998). Degrees of interdependence actually change the form of trust. Several authors link forms of trust to the context of the relationship (Lewicki, McAllister, & Bies, 1998; Sheppard, & Sherman, 1998). Thus, scholars appear to agree fundamentally on the meaning of trust and on the conditions of trust. Trust is neither a behavior (e.g., cooperation) nor a choice (e.g., risk taking), it is a psychological condition that can form or cause or result from such actions. Because risk and interdependence are necessary conditions for trust, variations in these factors over the course of a relationship between parties, can alter both the level and the form that trust takes. This author adopts the definition of trust by Mayer, Davis, and Schoorman (1995). As mentioned, trust is defined by these authors as “the willingness of one party to be vulnerable to the actions of another party based on the expectations that the other party will perform a particular action which is important to the trustor, irrespective of the ability to monitor and control the party” (1995 :712). This definition pertains to both conditions of trust. It regards trust as interpersonal and suggests that one’s expectations that others will cooperate or behave benevolently simply because of external incentives or sanctions, does not count as trust. Thus, predictability does not mean trust. One’s ability to predict an untrustworthy behavior of another, does not imply that one will be willing to be vulnerable. Likewise trust is not a condition for cooperation. Cooperation may be simply a result of a need to cooperate. The Evolution of Interpersonal Trust Five Theorists theorize the development of trust (Jones & George, 1998; Lewicki & Bunker, 1995; Lewicki et al., 1998; Lewis & Wiegert, 1985; McKnight et al., 1998; McKnight et al., 1998). The development of trust was first theorized by Lewis and Wiegert (1985) and a decade later by Lewicki and Bunker (1995). Lewis and Wiegert (1985) examined the multi-facets of interpersonal trust and presented trust as a three level phenomenon. Lewicki and Bunker (1995) focused on interpersonal professional trust, and presented it as a three base phenomenon. Trust is based on three bases which are distinct levels of trust: cognitive, affective, and behavioral. The authors assume that the three level typology will enable a distinction among relationships, and also between processes and states of trust. Trust is cognitive in that we choose whom to trust, in what and under what circumstances. To develop cognitive trust one needs to constitute evidence of trustworthiness. One gathers information and knowledge until trustworthiness or untrustworthiness is attributed to the other. The cognitive trust level exists in all types of relationships. Trust can develop to an emotional level. We decide in which relationships to invest intense emotional energy. In primary relationships this level of trust would be dominant. Emotional trust contributes to cognitive trust which in turn further contributes to emotional trust. Cognitive trust however, must proceed emotional trust. There has to be some cognitive trust in order for emotional trust to develop. While in cognitive trust behaviors are specific, in emotional trust they are open ended and diffused. Behavioral trust is the manifestation of emotional trust. Once one feels safe, one can undertake a risky course of action with the confident expectation that all people will act dutifully. The behavioral level of trust reinforces the emotional level of trust. Lewicki and Bunker (1995) present three bases to interpersonal professional trust: deterrence based trust, knowledge based trust, and identification based trust. The three bases are linked and sequential. The drives underlying each base and the actions to maintain trust in each base vary. Deterrence based trust is grounded in fear or in a threat of acting differently than expected. This threat may be influenced by one’s orientation toward risk, or in one’s expectation of potential rewards. Rousseau et al. (1998) call deterrence based trust a calculus based trust and claim that it is based on a rational choice. It is a characteristic of interactions based on economic exchange. Trust emerges when the trustor perceives that the trustee intends to perform an action that is beneficial. The perceived positive intentions in calculus based trust derive from the existence of deterrence and also because of credible information regarding the intentions or competence of another (Barber, 1983). Credible information about the trustee may be provided by others or by certification signaling that the claims of trustworthiness are true (Doney et al., 1998). Rousseau et al. (1998) claim that deterrence based trust emphasizes utilitarian considerations that enable one party to believe that another will be trustworthy, because costly sanctions in place for breach of trust exceed any potential benefits from opportunistic behavior (Ring & Van De Ven, 1992, 1994; Shapiro, Sheppard & Cheraskin, 1992). Although asset specifity effects in the form of switching costs economics, are examples of deterrence based trust, Rousseau et al. (1998) question whether sanctions foster or substitute trust, particularly in interfirm situations (Hagen & Choe, 1998). Some authors have raised the issue that deterrence based trust is not trust at all (Sitkin & Roth, 1993). It is clear that trust sanctions can foster or obstruct cooperation, which is a behavior. However, although trust promotes cooperation, cooperation can occur for other reasons as well (e.g., coercion or fear of loss). Since trust is not a control mechanism, but rather a substitute for control, it may reflect a positive attitude about another’s motives. There is an incompatibility between strict controls and positive expectations about the intentions of the other party. Control comes into play only when adequate trust is not present (e.g., a detailed legal contract is one mechanism for controlling behavior). Some controls signal the absence of trust, and therefore, can hamper its emergence, perhaps by limiting the degree of interdependence that develops between the parties. Macaulay (1963) observed that detailed contracts can get in the way of creating an effective exchange relationship: in effect, people do not need to develop trust when their exchange is highly structured and easily monitored. Although detailed contracts promote limited cooperation based upon deterrence, most firms that form alliances do so because of a social network of prior alliances, which makes detailed contracts less necessary (Kogut, Shan, & Walker, 1993). Moreover, beliefs in the absence of negative intentions is not the same as beliefs in the presence of positive ones. Thus, deterrence based trust may not be trust at all, but closer to low levels of trust (Lewicki et al., 1998), where costs of breaching trust are high and the involvement between the parties is limited in the first place. The greater the interdependency between parties, the greater the necessity to go beyond short transactions, and the less parties are repulsed by what they learn about the other, the higher the probability of a transition from a deterrence based trust to a knowledge based trust. A knowledge based trust is grounded in the understanding of the other’s motives, intentions, actions, and thoughts. This predictability requires repeated interactions in multidimensional relationships. While deterrence based trust focuses on sensitivity to differences, knowledge based trust focuses on sensitivity to similarities and on expanding one’s knowledge about the other. The learning process continues until an internalization of the other’s needs takes place. With this internalization, unless the parties lack the time, energy, interest or, have not established an effective understanding of the other, a transition from a knowledge based trust to an identification based trust takes place. An identification based trust is grounded in the effective understanding of the other, the endorsement of the other’s needs and wants and in the actions one can take to substitute for the other in interpersonal transactions. The focus in this stage shifts from expanding the knowledge about the other to identifying with the other. Identification or affective based trust is called by Rousseau et al. (1998) relational based trust. Relational trust derives from repeated interactions over time between trustor and trustee. Information available to the trustor from within the relationship itself forms the basis of relational trust. Reliability and dependability in previous interactions with the trustor, give rise to positive expectations about the trustee’s intentions. Emotion enters into the relationship between the parties because frequent, long term interaction leads to the formation of attachments that are based upon reciprocated interpersonal care and concern (McAllister, 1995). This explains the reason for referring to this form of trust as affective trust and as identity based trust (Coleman, 1990). Repeated interactions can vary in the resources exchanged and in the scope of interdependence between the parties. Repeated cycles of exchange, risk taking, and successful fulfillment of expectations strengthen the willingness of trusting parties to rely upon each other and expand the resources brought into the exchange. Thus, an exchange can evolve from a transaction into a relationship. Organizational citizenship behavior and organizational support are characteristics of high levels of relational trust based upon experience within the relationship (Eisenberger et al., 1986; Organ, 1990). In relational based trust the array of resource exchange is much broader than in any other type of trust (Rousseau et al., 1998). Both socio-emotional and concrete resources are exchanged entailing a greater level of faith in the intentions of the other party. To sum, in any professional relationship trust is initially deterrence based, with a calculus based trust and behavior control. As parties determine their interdependencies, risks, benefits, and enjoy joint outcomes they may grow to like each other and develop an identification based trust.
Initial Trust
Several trust theories have stated that trust develops gradually over time (Blau, 1964; Rempel, Holmes, & Zanna, 1985; Zand, 1972). When contrasted with recent empirical findings, these theories present an interesting paradox. By positing that trust grows over time, these trust theorists implicitly assume that trust levels start low and gradually increase. In some studies however, trust levels were surprisingly high when parties first interacted (Berg, Dickhout, & McCabe, 1995; Kramer, 1994). This initial trust was evident despite the lack of accumulated knowledge or experience with the other person as suggested by Lewicki and Bunker (1995) and Holmes (1991). McKnight, Cummings, and Chervany (1995) explain the paradox of high trust in initial relationships by identifying hidden factors and processes that enable trust to be high when people first meet. A person exhibits a disposition to trust to the extent that she or he demonstrates a consistent tendency to be willing to depend on others across a broad spectrum of situations and people. In contrast to other authors, McKnight et al. (1995) distinguish between two types of disposition to trust. Each disposition affects the trusting intention in a different way: faith in humanity and trusting stance. McKnight et al. (1998) relate to effects of dispositional trust by the relationship time frame. The time frame is important in predicting the effects of the disposition to trust. While in ongoing relationships the effect of a person’s trusting tendency declines, in new organizational relationships it is assumed to have a significant effect on trusting beliefs and intentions. Trust as a dichotomy Lewicki, McAllister, and Bies (1998) posit the framework of incorporating trust with distrust. Just as love and hate exist simultaneously in one relationship, trust and distrust also exist in one relationship. Under these conditions trust is limited to specific exchanges, and on other exchanges or facets (i.e., financial but not personal), distrust exists. Opportunities are pursued and risks are monitored. The range of a calculus based trust is often limited to situations where evidence of failure to perform can be obtained in the short term. Risk may entail short term performance losses but not threaten the trustor’s broader interests (Rousseau et al., 1998). While the information creating trust and distrust is accumulated at the same time, once trust and distrust are established, there is a dynamic tension between them, reflecting pressure towards homogeneity that facilitates harmonious interaction and coordination. In this trust/distrust framework, relational trust entails both beliefs in the positive intentions of the trustee, and also in the absence of negative intentions giving rise to the condition of high trust/low distrust. Interdependence between the parties to relational trust is likely to increase over time as new opportunities and initiatives are pursued. A dynamic of relational trust is its potential for expansion or contraction, where experiences over time can escalate positive beliefs regarding intentions of the other or conversely, exacerbate negative beliefs (Lewicki et al., 1998; Sitkin & Roth, 1998). Figure 8 presents the integration of trust and distrust.
Conditional and Unconditional trust
Jones and George (1998) assume that people act in social situations based on the meanings that they have learned to associate with them, and these meanings are acquired by interactions with other people so that a definition of the social situation is created over time. Thus, in any encounter two parties mutually develop and negotiate a definition of the social situation. People use their values to decide whether or not to trust the other. People decide whether differences in values appear to be so divergent, that they are making themselves vulnerable to the other party. The point at which two parties have strong confidence in each other’s values and trustworthiness, have favorable attitudes toward each other, and experience positive affect in the context of the relationship, is crucial in the evolution of trust. To distinguish between the experience of trust before and after this point, Jones and George (1998) make a distinction between states of conditional and unconditional trust. Conditional trust is a state of trust in which both parties are willing to transact with each other, as long as: each behaves appropriately, each uses a similar interpretation to the situation, and each can take the role of the other. Attitudes of one party towards the other, are favorable enough to support future interactions. Conditional trust is sufficient to facilitate a wide range of exchanges. It is consistent with knowledge based trust, and with positive expectations of the other. Unconditional trust characterizes an experience of trust that starts when individuals abandon the pretense of suspending the belief. Because shared values now structure the social situation and become the primary vehicle through which those individuals experience trust. With unconditional trust each party’s trustworthiness is assured based on confidence in the other’s values. Previous repeated interactions created knowledge which is contained in each one’s attitudes. Positive affect increases as positive moods and emotions strengthen the affective bonds between parties, bolstering the experience of trust. Thus, when unconditional trust is present, relationships become significant and often involve a sense of mutual identification (Lewicki & Bunker, 1995; Shapiro et al.,1998).
Institutional Trust
Institutional trust refers to organizational and societal factors that support the mass of trust that sustains trust behavior (Gulati, 1995; Ring & Van De Ven, 1992; Sitkin 1995). These may be cultural controls or legal controls that protect individual rights and property. The literature questions whether institutional trust is a control or a form of trust. regardless of the answer to this question, the factors defined as institutional trust serve as a springboard for the creation of trust in the interpersonal level (Hagen & Choe, 1998; Ring, 1998). Institutional controls can also undermine interpersonal trust (Rousseau et al., 1998) where legal mechanisms give rise to rigidity in responses to conflict and substitute high levels of formalization for more flexible conflict management (Sitkin & Bies, 1994). Zucker (1986) also views institutional trust as reducing the opportunity for creating interpersonal trust. The role institutional trust plays in shaping interpersonal trust is yet to be studied. Table 1 presents the foci of researchers in all previous trust studies as presented by Bigley and Pearce (1998).
The Measurement of Trust
The common meaning of trust does not imply that all operationalizations of trust reflect the same thing. Inter organizational and interpersonal trust are different because the focal object differs. Lewicki and Bunker (1995) claim that the existing theories regarding the nature and dynamics of trust are fragmented and simplistic. Each discipline assumes its own frame and perspective without articulating the parameters of that frame. Each scholar, therefore, is a blind person describing her or his small piece of the elephant. Mayer, Davis, and Schoorman (1995) support this view claiming that trust research is hindered by a lack of clear differentiation among antecedents and outcomes of trust. Previous trust research concerning general trust (Rotter, 1967) and trust as a social phenomenon (Lewis & Wiegert, 1985), do not clarify the relationships and how to measure them (Mayer & Davis, 1995). Since fundamental elements of the definition of trust are comparable across research and theory, focusing on parties inside and outside companies, researchers can examine trust relationships from the different disciplinary perspectives, with each perspective adopting a different foci. Trust - a Continuous or a Discrete Variable Intuitively, trust is a continuous variable with different relationships positioned on a continuum of trust. There is evidence however that relational trust is discrete (Sheppard & Sherman, 1998). First, Haslam and Fiske (1992) found evidence for four discrete cognitive structures. Second, encompassed within the psychological contract literature, there is a distinction between two forms of contracts: transactional and relational. Relational contracts, however, involve open - ended relationships. Third, a recent factor analysis of trust scales shows that three of the four allied forms of trust are quite distinct (Lewicki, Stevenson, & Bunker, 1997). Fourth, different disciplines seem to adopt different aspects of trust as their study purview, whereas relationships in reality entail some combination of them all. Sheppard and Sherman (1998) further claim that this evidence for distinct forms does not imply that trust is entirely discrete. The depth of trust and the level of interdependence across relationships clearly differs and the relationship changes its form. A supplier relationship becomes an integrated design, which becomes an alliance. Thus, although the forms of relational trust are discrete, Sheppard and Sherman (1998) claim that they are also highly linked. One form serves as the building block for the other creating relationships that have different bandwidth of trust. Social psychologists often see trust as either/or, where one person completely trusts or distrust another (Gabarro, 1990). This static view of trust is related to the predominance in early trust research of laboratory studies which focused on highly structured games (i.e., the prisoner’s dilemma). Under such conditions, the level of trust reflects a single point, rather than a distribution along a continuum. The fact that trust changes over time, from developing to building to declining, in long term relationships indicates that trust is indeed a continuum. Rousseau et al. (1998) contend that scholars do at times focus on multiple phases of trust (Bhattacharya et al., 1998; Bigley & Pearce, 1998; Jones & George, 1998; Lewicki et al., 1998). However, the tendency of authors is to focus upon one stage and to specify a conceptual framework within a particular phase. Rousseau et al. (1998) claim that such an emphasis on phase-specific trust may be necessary at this point in the development of trust scholarship.
The Status and the Level of Trust
Examining if differences in the status of trust as a cause, effect, or interaction, vary across disciplines, Rousseau et al. (1998) found that discipline differences do not account for the status given to trust. Trust was an independent, dependent, or moderator variable and its function in the causal framework reflects a rich cross disciplinary view. The multi-levels of trust exist in and between organizations reflecting an array of entities: individuals, dyads, groups, networks, firms, and interfirm alliances in which trust and its related processes play a role. It is commonly assumed that different disciplines occupy different turf in organizational science. Psychologists, the individual and occasionally the group. Sociologists, the group and society. Economists, the individual or the firm. Although micro-organizational behavior researchers predominate the individual level, as a trustor (Doney et al., 1998; Lewicki et al., 1998; McKnight et al., 1998; Mishra & Spreitzer, 1998) or a trustee (Elangoven & Shapiro, 1998; Whitener et al., 1998), and economists predominate the firm level (Das & Teng, 1998; Hagen & Choe, 1998; Sheppard & Shapiro, 1998), neither turf is sacred to a particular field. Scholars from a variety of backgrounds address phenomena at any given level of analysis. Regardless of the disciplinary focus of the scholar, trust might require a multi-level analysis (Rousseau et al., 1998). Institutions promote or constrain trust relations (Fukuyama, 1995; Hagen & Choe, 1998; Smitka, 1994). Thus, microlevel trust relations are constrained and enhanced by macro processes (Sitkin, 1995). Conversely, broader forms of trust, particularly between firms, can be influenced by microlevel arrangements in particular how individuals representing each firm treat each other (Fichman & Goodman, 1996). Thus, multilevel processes underlie trust, and scholars across a variety of disciplines have embraced a multi - level view. One of the most important considerations in trust measurement is a match between the level of the studied phenomenon and the analysis level of the instrument used to measure the phenomenon. Interpersonal trust requires an individual analysis level, inter-group trust requires a group analysis level, and intra organizational trust requires an institutional analysis level. Ring and Van De Ven (1994) claim that all kinds of trust are established in the interpersonal level. Organizations are comprised individuals working with individuals of other organizations. Thus, although group and organizational trust are influenced by the national cultural, trust itself is formed at the interpersonal level. Previous Measures of Trust Cook and Wall (1980) described three approaches to measure trust. The three approaches differ by the directness of the measurement. The most indirect approach is to infer trust from other behaviors. A second approach is to create situations where trust is critical for performance, and measure performance (e.g., prisoners dilemma). The direct approach to measure trust is through measurement of affective response using self reports. These authors measured reliability and faith by separate measures for peers and managers and claimed that there are not enough measures for trust in the work context. Trust has been measured using 13 instruments (Butler, 1991; Cook & Wall. 1980; Curral & Judge, 1995; Hart et al., 1986); Gabbaro, 1978; George & Swap, 1982; Jennings, 1971; Larzelere & Huston, (1980); McAllister, 1995; Rempel et al., 1985; Roberts & O’Rielly, 1974; Rotter, 1971; Scott, 1981). Next, these instruments are briefly discussed. George & Swap (1982) measured dimensions of the trust construct. They measured interpersonal trust between specific individuals by the Specific Interpersonal Trust Scale (SITS). This scale measures the general tendency of a specific person to trust another person in specific situations. Some researchers (Butler, 1983; George & Swap, 1982; Remple & Holmes, 1986) advocated the relevance of situational trust in specific others as opposed to global trust in generalized others. Scott (1981) focused on trust in four specific groups. Gabarro (1978) and Jennings, (1971) identified ten conditions, which can be useful to those more interested in building trust rather than in understanding the construct of trust. Jennings (1971) used clinical interviews and found trust to be a condition of mobility in organizations. Gabarro (1978) also used clinical interviews and found that there are nine bases of trust. These bases include integrity (honesty and moral character), motives (intentions and agenda), consistency of behavior (reliability related to the concept of predictability), openness (leveling and expressing ideas freely, one concept of accessibility), discreteness (keeping confidences), functional/specific competence (knowledge and skills related to a specific task), interpersonal competence (people skills), business sense (common sense and wisdom about how a business works), and judgment (ability to make good decisions). The relative importance of antecedents of trust was tested for upward and downward trust (Butler & Cantrell, 1984). For both directions the importance was in the following order: competence, integrity, consistency, loyalty (making the target person look good), and openness. Many of the psychometric properties of these instruments, however, have not been reported and therefore, these scales are hard to evaluate. Only Hart et al., (1986) reported psychometrics including factor analysis of the items and the reliabilities of each of their three scales. Other scales reported psychometric properties in a partial manner. George and Swap (1982) showed that their measure was discriminable from measures of liking and loving. Larzelere and Huston (1980) distinguished their measure from those of self disclosure, social desirability, love, and generalize trust. Roberts and O’Rielly (1974) found the predicted correlation of trust with several organizational variables. Although divergent validity is fundamental, it is only secondary to the reliability of the scale. Unfortunately, in all of these studies no test- retest coefficients were reported. Butler (1991) developed the Conditions of Trust Inventory (CTI), extending and validating the list that was identified by Gabarro (1978). Instruments which were introduced prior to the Conditions Trust Inventory, although well validated, did not attempt to measure a complete, exhaustive, set of concepts that represents the conditions leading to trust. The CTI purports to measure conditions of trust in any relationship. The ten conditions, however, may not represent a complete domain of trust conditions in family relationships. The CTI was found to encompass nine distinct factors and to possess good psychometric properties. The two other measures of trust have been recently developed and reflect the growing importance of trust in the work context and specifically in marketing. Curral and Judge (1995) developed an instrument to measure trust between organizational boundary role persons (OBP). The survey instrument measures the most proximal antecedent of trusting behavior. Specifically, behavioral estimation was used to assess the estimation of the OBP that she or he will engage in trusting behavior toward a counterpart OBP. The manifestation of trusting behavior was conceptualized into dimensions that are characteristics of OBPs in particular. The dimensions were open and honest communication, informal agreement with the counter OBP, maintaining surveillance over the counterpart OBP, and task coordination. In the dyadic level, individual level variables were applied. High mutual trust in a dyad depended on positive attitudes members had toward trusting each other, normative pressure to trust each other, past trustworthiness, and trusting personalities. The OBP scale adequate psychometric properties were fully reported. McAllister (1995) developed a scale that measures levels of trust, specifically cognitive and emotional trust. This scale is based on the other measures of interpersonal trust (Cook & Wall, 1980; George & Swap, 1982; Rempel et al., 1985; Rotter, 1981), but is the only one that measures cognition and emotion based trust between managers and employees. This study will use Butler’s CTI scale. Three reasons explain this choice. First, the scale is targeted to measure global interpersonal trust and possesses good reliability (.88) and validity. Second, the CTI scale is used to answer research questions concerning conditions leading to interpersonal trust and their relative importance in respect to each other. This advantage enables this researcher to test which conditions of trust underlie each RS level. Third, in contrast to other scales that indicate only whether or not trust exists, the CTI scale enables to identify aspect of trust on which there may be a problem hindering high trust.
CHAPTER FIVE Social Exchange
Chapter five defines social exchange, distinguishes between social and economical exchanges and presents theories which bridge the two types of exchanges. The last part of this chapter reviews a special case of social exchange, the leader member exchange theory. Human needs are seldom satisfied in solitude. People depend on each other for the material and psychological resources they need for their well being.
  • Social exchanges differ from economic ones in that they may not have any
  • economic utility but rather an intrinsic utility. Many theories of trust are
  • grounded in the social exchange theory since a social exchange is primarily
  • based on reciprocity, gratitude and trust.
The Study of Exchange
Many theories of trust are grounded in a social exchange theory (Blau, 1964). The social exchange theory assumes that trust emerges through the repeated exchanges of benefits between two individuals. Because it addresses the process of initiating and expanding social exchanges, this theory provides useful lenses for examining motivational mechanisms underlying the initiation of trustworthy behavior (Whitener, Brodt, Korsgaard, & Werner, 1998). Human needs are seldom satisfied in solitude. People depend on each other for the material and psychological resources they need for their well being. People associate to exchange their resources through interpersonal behavior. In the study of these exchanges there has been a traditional division of tasks. Economists have long been concerned with the exchange of money with goods, and more recently, with labor and information. Psychologists and sociologists have investigated transactions that involve more subtle resources (Etzioni, 1968; Maslow, 1954) such as self esteem, attraction, affect, respect, and status (Foa, Converse, Tornblum, & Foa, 1993).
Social Exchanges Versus Economical Exchanges
Social exchanges differ from economic ones in several fundamental ways (Blau, 1964). First, social exchanges may involve extrinsic benefits with economic value (e.g., information and advice) or intrinsic benefits without any direct objective economic utility (e.g., social support). Second, extrinsic benefits are often expressions of support and friendship and thus, have intrinsic value as well. Therefore, exchanges that have little or unclear economic benefit, may have a strong impact on the social dimension of a relationship. Third, whereas benefits in economic exchanges are formal and often contracted explicitly, such benefits are specified or negotiated in a social exchange. Thus, providing benefits is a voluntary action. Fourth, because such behavior is voluntary, there is no guarantee that benefits will be reciprocated or that future benefits will be provided. The exchange of benefits therefore involves uncertainty, particularly in early stages of the relationship, when the risk of non reciprocation is high. Consequently, relationships evolve slowly, starting with exchange of relatively low value benefits and escalating to high value benefits, as the parties demonstrate their trustworthiness. Behavior, however, could be influenced by both economic and non-economic factors. Thus, the dichotomy between exchanges seemed artificial and had to be bridged. The first attempt to bridge the dichotomy was by sociologists and social psychologists (Blau, 1964; Homans, 1961; Longabough, 1963; Taibut & Kelly, 1959). These authors sought to interpret every interpersonal behavior as an exchange, characterized by profit and loss. The social exchange theory, however, had difficulties explaining situations in which the taker does not lose anything. Resources like information and love can be given without reducing the amount that the giver possesses, contradicting the basic notion of exchange. Another difficulty was the difference between resources that one exchanges with himself. While money can not be given to the self, one can provide the self with esteem or even information after an exploratory behavior. Thus, various exchanges follow different rules and require a theory that provides order among these exchanges. Bridging Social and Economic Exchanges with Order One such theory classifies resources by two coordinates: concrete (i.e. a physical object) versus symbolism (i.e., a smile, language), and particularism (a specific person provides the resource) versus universalism (any person can provide the resource). To facilitate the mapping of exchanges on these two coordinates, Foa and Foa (1971) grouped resources into six groups. All resources were divided into six categories: love, information, status, money, goods, and services (Foa, Converse, Tornblum, & Foa, 1993). Love is defined as an expression of affection, warmth, or comfort. Status, is an expression of evaluative judgment which conveys high or low prestige, regard or esteem. Information includes advice, opinions, instruction, or enlightment, but excludes those behaviors which could be classed as love or status. Money, is any coin, currency, or token which has some standard unit of exchange value. Goods are tangible products, products, objects, or materials, and services, involve activities on the body or belongings of a person which often constitute labor for another. Each of the six resource types can be classified on the basis of the two suggested coordinates concrete- symbolic and particularistic-universal. On the first coordinate, concreteness, services and goods involve the exchange of some overtly tangible activity or product and are classed as concrete. Status and information however, are typically conveyed by verbal paralinguistic behaviors and are more symbolic. Love and money are exchanged in both concrete and symbolic forms and thus occupy intermediate positions on this coordinate. Leader Member Exchange Theory One type of a social exchange which has been extensively studied in the past two decades, is the social exchange between leaders and their followers, called in the literature leader-member exchange theory (LMX) (Dienesch & Liden, 1986). In contrast to classical leadership theories which focus on the leader traits, situational factors etc., LMX regards leadership as a multi-domain consisting of the leader, the follower and the relationship between them. A high quality exchange is characterized by relations of trust, respect, obligation, loyalty, interaction, support and formal and informal rewards exchanged between the parties. A low quality exchange is characterized by low trust, low interaction, low support, and low rewards. Research has proposes trust to be highly related to the quality of LMX both as a determinant, when both parties attribute trustworthiness to the other and choose to expand the relationship beyond a transactional exchange (Bauer & Green, 1996: Dienesch & Liden, 1986), and as a mediating outcome of the exchange when the parties invest in a relationship (Wayne & Green, 1993) leading to higher performance, higher organizational commitment, faster career progression, higher satisfaction and more functional communication patterns (Graen & Uhl Bien, 1995). The development of LMX was first theorized by Dienesch and Liden (1986) who presented LMX as an interaction-attribution-delegation process. Graen and Scandura (1987) also theorized about the development of LMX and presented the core of the developmental process as role making. Next, a review of LMX models. The LMX development process as presented by Dienesch and Liden (1986) begins with an initial interaction. The leader is a source for information and support of newcomers who are engaged in both a socialization process and LMX. The leader is engaged in a judgment and attribution process and assigns the member with trial assignments. The performance of the member on these assignments serves as an indicator of ability, potential future performance, dependability in task completion. This information assists the leader in shaping her or his attributions about the member. Satisfactory performance results in delegation activities of the leader and reciprocal trust that leads to high LMX. Graen and Scandura (1987) present a similar development model. In the first stage, role taking, the member is assigned tasks that are specified in her or his job description. The leader initiates job requests and communicates them to the member. The leader gathers important information to assess the member’s motivations, behaviors and potential performance on unstructured tasks. The leader questions the extent of energy she or he should invest in the member. In the second stage, role making, the member is provided with a greater latitude on the job. Through a set of dyadic transactions the parties inter-connect. Through the unstructured tasks the parties evaluate their work together and test the dyad as an alternative to problem solving. Each party examines the resources and contributions of both parties. These must be perceived as equitable. In the third stage, role routinization, a dyadic understanding is created. This understanding grows deeper as the parties collaborate on unstructured tasks and informal activities. This process produces stable forms of dyadic relationships. Trust and LMX Bauer and Green (1996) propose that LMX is a trust building process. Outcomes of high exchanges are encompassed by identification based trust. One is confident that her or his interests will be defended and protected with no monitoring. The three levels of trust presented by Lewis and Wiegert (1985) are parallel to the development of LMX. In the cognitive level of trust, information is gathered to constitute evidence of trustworthiness (Lewis & Wiegert, 1985; Mayer et al., 1995). This level parallels the role taking stage in which the leader assesses the dependability of the member in completing tasks, the willingness to exert effort at work, and the potential loyalty (Graen & Scandura, 1987). This trustworthiness perception in the cognitive level of trust proceeds the emotional level of trust in all models (Bauer & Green, 1996 ; Lewis & Wiegert, 1985; Lewicki & Bunker, 1995; McAllister, 1995). Similarly, an evaluation of performance in LMX proceeds the later LMX relationship which is an affect outcome. Affective trust is the emotional bond in the relationship. In LMX it is the bond that grows deeper once a dyadic relationship is established in the role routinization stage (Graen & Scandura, 1987) or in interlocking behaviors (Bauer & Green, 1996; Dienesch & Liden, 1986). The behavioral level of trust is parallel to the role making stage of LMX development or the delegation stage in which an action entailing risk is taken by the leader as she or he delegates responsibility and both parties progress to the formalization of the relationship. Thus LMX is a trust building process (Bauer & Green, 1996). CHAPTER SIX Hypothesis Development This study focused on relationship selling, a dyadic consumer-sales associate relationship. Levels of relationship selling, moderated by consumer desire for service relationship, were hypothesized to predict trust. Trust, and relationship selling levels, moderated by perceived quality, were hypothesized to predict consumer loyalty to the sales associate, and consumer loyalty to the store. Consumer loyalty was hypothesized to predict sales performance of the SA and be mediated by purchase frequency. Introduction Until recently marketing was focused on acquiring new customers and managing transactional exchanges. Berry (1983) was first to stress that acquiring new customers should be an intermediate step in the marketing process rather than the ultimate objective. The current marketing focus is on transforming frequent customers into loyal customers (Barnes, 1994; Berry, 1983; Gronoos, 1994, 1996). This is a paradigm shift (Aijo, 1996; Gummeson, 1997; Senge, 1994). The core of the new marketing paradigm is relationship selling which refers to forming relational exchanges rather than transactional exchanges with consumers. Relationship selling (RS) is defined as a one-on- one close relationship with the ultimate customer (Beatty, Mayer, Coleman, Reynolds & Lee, 1996; Freiertag, 1994; Gronoos, 1994; Levine, 1993; Morris, 1996; Schultz, 1993). RS is a special case of relationship marketing. While relationship marketing addresses relationships with various network players such as suppliers, employees, competitors, and partners, in any business environment, RS specifically addresses consumers in service environments (Riechheld, 1993). RS includes attracting, maintaining, and enhancing relationships with customers who are highly involved in a purchase due to the importance, the variability, and the complexity of the desired product. This high involvement brings customers to reduce their choice and desire a continuity with the same provider (Sheth & Parvatiyar, 1995). Thus, services offering the above benefits develop relational exchanges in addition to transactional exchanges. Four convergent influences that have propelled the focus on RS were identified: customer benefits (Berry, 1995; Jackson, 1993), company benefits (Reichers, 1994; Reichheld & Sasser, 1990), information technology, (Egolf, 1997; Endicot, 1997; Larson, 1996; Melvin, 1996; Pollack, 1997; Stacey, 1994), and maturity of services (Aijo, 1996; Berry, 1995; Gummeson,1996; Kawelski, 1997). Several trends in the marketplace are expected to strengthen and further stimulate RS. These trends are: the slow growth and maturity of service, the growing emphasis of quality, value, and customer satisfaction, the growing emphasis on relationship building and customer retention, and advantages that information technology provides in tracking consumer purchases and habits. Organizations applying relationship selling desire high quality interactions based on the premise that developing better relationships with customers will not only reduce costs, but will also provide them with greater sales, market share, and business share. Only two studies, however, examined the connection between relationship and financial performance. Crosby, Evans, and Cowles (1990) found no evidence that the relationship is linked to performance in service marketing. In contrast, Leuthesser (1997) found that quality interactions do increase business share, but only in long established relationships. In such relationships, relational behaviors explained 11% of the variance in business share. Thus, the importance of relationship marketing as a strategy that both maintains market share in slow growth, high competition business environments, and increases business share is growing. Companies need information regarding antecedents, outcomes, and implementation requirements. Nevertheless, little is currently known about the forces underlying successful RS. The State of the Literature From January 1995 to August 1998, 257 articles were published on the topic of relationship marketing (computerized data bases and shelf search). Of the 257 articles however, 110 (43%) deal with RS and of those only 17 are empirical studies (7%). Table 2 presents a summary of the empirical studies of RS to date. Table 2 The Study of RS to Date STUDY YEAR TOPIC FINDINGS Cross and Smith 1995 Social bonding Social bonding increases dependence of consumer on SA Peterson 1995 Consumer benefits Money savings explain consumer engagement in relationships Prince 1995 RS as promotion RS helped market high quality estates Beatty et al., 1996 RS development Four phases in the relationship Hausman 1996 Social benefits Commercial friendships develop between consumers and SA Goodwin and Gremler 1996 Social aspects of RS and consumer loyalty Marshall 1996 RS and consumer rewards Need differential rewards Mitchelle and Singh 1996 RS orientation in health context In hospitals RS orientation prevented hospital decline Schijns and Schroder 1996 Dimensions of relationship strength Behavioral measures should be combined with psychological measures Williams and Attaway 1996 Relationship among organizational culture, customer orientation, and RS development Influence of organizational culture on relationship development is mediated through customer oriented behaviors of SA Ziethaml 1996 Social benefits Social benefits come from the SA but also from other consumers in the store Barclay and Smith 1997 Examining a trust based model in dyads of selling partners Identification of obstacles and ways to overcome them Bitner 1997 Service encounters Employee is the most important factor in shaping the encounter Dorgan 1997 Communications with consumers Outlines information that need to get Gengler, Leozczyc and Popkowski 1997 RS and customer satisfaction Customer satisfaction both measures consumer attitudes and manages relationships Leuthesser 1997 Relational behaviors, RS and sales Relationships between RS and performance depends on relationship duration. The longer the relationship, the greater the affect on business share. Ramsey and Sohi 1997 RS skills Listening is the most important skill Stump and Sriram 1997 RS and Information technology investments Information technology investments affect closeness in relationships Bolton 1998 Role of customer satisfaction in RS Satisfaction evaluation depends on relationship duration. The longer the relationship, the less new information affects the initial evaluation. Gwinner, Gremler, and Bitner 1998 Customer motivation for relational exchanges Three benefits: confidence, social, special treatment Thus, the knowledge in the RS arena has been broadening rather than deepening in one topic. Much of the theoretical work on RS engaged in perspectives such as the framework, principles, and RS as a paradigm shift (Bendapudi & Berry, 1997; Berry, 1983, 1995; Berry & Parasuraman, 1990; Bitner, 1996; Czepiel, 1990; Gronoos, 1990, Gummesson, 1987; Lovelock, 1983; Moorman, Zaltman, & Deshpande, 1992; Sheaves & Barnes, 1996; Sheth & Parvatiyar, 1995). The early studies of RS, from 1983 to 1995, dealt with the question of “why use RS” and focused on potential benefits that companies that applied RS could expect (Berry, 1983; Cziepiel, 1990; Gronnos, 1990; Gummesson, 1987; Lovelock, 1983). The studies of RS in 1996 dealt with the question of “what is RS” and focused on the relationship itself. Study topics were relationship strength (Shijns & Schroder, 1996; Wilson, 1990), underlying conditions of RS (Hausman, 1996; Marshall, 1996), relationship development (Beatty et al., 1996), and the linkage between relationship development and organizational culture (Williams & Attaway, 1996). Realizing the positive affect of RS on consumer loyalty, the study in 1997 dealt with the question of “how to use RS” and focused on how to retain customers through better communication (Bitner, 1997; Dorgan, 1997; Gengler et al., 1997; Leuthesser, 1997; Ramsey & Sohi, 1997) and through investments in information technology (Stump & Sriram, 1997). The study of RS and of trust in the marketing context is evolving into an important topic. Previous studies of RS pertained to antecedents and consequences of RS. Authors focused on antecedents of RS (Dorgan, 1997; Smith, 1997), on relational behaviors (Leuthesser, 1997; Ramsey & Sohi, 1997; Scott, 1995), and on indicators of relationship strength (Schijns & Schroder, 1996). Only one study focused on the developmental process of RS (Beatty, Mayer, Colemen, & Reynolds, 1996). Only one study examined the effects of RS on marketing enhancement (Prince, 1995) and only one study examined the long term effect of RS on sales (Leuthesser, 1997). To date, the RS research has added valuable insight into extending the literature of niche marketing and service. This author identified three gaps in the literature. First, although the field of marketing, which was traditionally influenced by economics, is increasingly influenced by other fields (e.g., fields of organizational behavior, sociology, psychology, political science, and systems thinking), none of the previous studies integrated RS with other fields. One topic which is studied by other fields and is related to RS is trust. Trust, a critical component in any relationship, and an important marketing tool (Berry, 1995; Bitner, 1996; Crosby et al., 1990; Czepiel, 1990; Morgan & Hunt, 1994; Parasuraman, Berry, & Ziethaml, 1991) has not been empirically studied in SAs- consumers dyadic relationships and is the focus of this study. This study integrates trust into RS testing antecedents and consequences of trust in SA-consumer relationships. The other two gaps relate to the methodology of previous studies. Previous studies analyzed relationships on a business to business level, on a business to consumers level, or on a SAs consumers level. Only one study, however, which studied the formation of SAs-consumers relationships, analyzed the dyad (Beatty et al., 1996). The dyad is comprised of a SA and her/his consumers. This dyad is important since it is the smallest possible unit which goes beyond one person, thereby facilitating an examination of interrelationships and revealing variations in the internal components of the two parties (Graen & Scandura, 1987; Senge, 1994). This study will focuses on the dyadic level while analyzing the SA-consumer relationship. Third, although both the consumer and the SA manage the interaction, and although the relationship is retained only when both parties benefit from the relationship, to date, all empirical studies but one used data from one party (e.g., firm, managers, SAs, consumers), without equal and concurrent emphasis on other parties. Surprisingly, only one qualitative, empirical study has been done (Beatty et al., 1996) and collected data from both parties to analyzed the consumer-SA dyad. This study will also collect data from both parties, SAs and their consumers. The Contribution of the Study It is important to understand antecedents and consequences of a trustful SA-customer relationship for three reasons. First, relationships were shown to enhance sales performance of the firm (Leuthesser, 1997; Riechheld & Sasser, 1990). For companies that wish to maximize business opportunities with their existing consumers, understanding what determines trust in sales relationships is necessary in order to establish and facilitate long term relationships. Second, knowing antecedents that establish trust, companies will be able to adjust selection, training and development, and compensation, to the goal of building long term relationships. Third, it is important to know consequences of trustful SAs-consumers relationships, in order to estimate expected benefits versus potential costs of this strategy before it is chosen. The most important antecedents of RS were found to be communication and trust. Trust was found to be predicted mainly by communication (Bitner, 1996; Morgan & Hunt, 1994; Ramsey & Sohi, 1997; Smith, 1997). While the centrality of trust is well acknowledged and supported, previous research has not yet examined the association between trust and groups of incentives (RS levels). This study extends the existing knowledge on both antecedents and outcomes of one-on-one marketing. The purpose of this study was to extend the existing knowledge testing groups of incentives, provided in different RS levels, as antecedents of the SA-customer trustful interaction. This study tested the affect of these incentives on consumer loyalty as well as on sales performance. This study makes three specific contributions to the literature while taking a systems view, by drawing upon the marketing, systems science, trust, psychology, and social exchange literatures. First, this study tested previously theorized but unexplored hypotheses while integrating between the literature of RS and the literature of trust. Second, this study was conducted at the dyadic level examining relationships between each SA and her/his consumers. Third, this study used multiple sources of data examining the relationship from both perspectives, that of the SA and that of the consumer. Berry (1995) theorized three RS levels as bases of the relationship. In each level of RS different incentives are provided to consumers: pricing incentives, social incentives, and structural solutions. A review of the relationship and service literatures (Barlow, 1992; Berry, 1995; Czepiel, 1990; McCallum & Harrison, 1985) led this author to believe that different types of relational benefits exist. Most of the previous research regarding benefits (Crosby, Evans, & Cowels, 1990; Morgan & Hunt, 1994; Peterson, 1995) focused on benefits of customer loyalty to the company rather than to the consumer. Many scholars have theorized potential benefits to consumers (Barnes, 1994; Bendapudi & Berry, 1997; Berry, 1995; Bitner, 1996; Sheth & Parvatiyar, 1995), but only one empirical study had been done (Gwinner, Gremier, & Bitner, 1998). Their findings indicate that consumer relational benefits can be categorized into three distinct benefit types: confidence, social, and special treatment. The benefit of confidence relates to trust in the service provider and supports Sheth and Parvatiyar (1995) who theorized that consumers want relationships in order to reduce risk. Thus, trust in the service provider and confidence in quality service are major consumer motivators for a relational exchange. Since Berry (1995) presents levels of RS as bases of the relationship, and since trust is a critical foundation of any relationship, this study integrated trust into RS hypothesizing that different RS levels predict trust. This was a pioneering effort to examine the relationship between RS levels and trust. Berry (1995) also hypothesized that RS levels predict loyalty. This study associated between RS levels, trust, and loyalty, examining whether RS levels and trust have a unique contribution in explaining the variance in loyalty beyond the contribution of factors that are known to explain loyalty. Finally, congruent to Berry’s theory (1995) this study also tested which RS levels underlie each type of loyalty (i.e., loyalty to the SA versus loyalty to the store), and how each type of loyalty is related to sales performance. The second contribution of the study lies in its analysis level. Selling involves three domains: the buyer, the seller, and the relationship between them. Within each domain researchers can adopt a different level of analysis (i.e., the seller’s behavior can be examined from the perspective of the buyer, or the relationship can be examined from the perspective of the buyer, with customer expectations from the relationship as the focus of the researcher). Since selling involves all these three domains, studies could involve each domain singularly, contributing valuable information, or multiple domains in combination. Borrowing from the LMX study in its analysis, this study also obtained a multi-domain representation (Graen & Uhl-Bien, 1995) of the one-on-one selling process. In contrast to previous RS studies that focused on business- consumers or SAs-consumers level of analysis, this study focused on the dyadic relationship, examining the relationship between a SA and her/his consumers. The third contribution of this study lies its data sources. This author collected data from both parties of the dyad, thereby examining the relationship from both perspectives, that of the buyer and that of the seller. This extended the multiple perspectives view of the system (Lendaris, 1986; Linstone, 1984; Mitroff & Turoff, 1973; Turoff & Mitroff, 1974) thereby facilitating a better understanding of forces that drive customer loyalty, of the customer-SA reciprocity, of characteristics that underlie customer receptivity to different forms of RS, and of effects of RS. For example, assuming that agreement itself may indicate relationship quality, to what extent is there an agreement between the parties on the provided incentives. Implications of the Study This research has practical implications as well, particularly for the retail or personal services industry. Companies are striving to achieve a sustainable advantage. Most competitive strategies adopted by companies in these industries (e.g., product differentiation, brand management, pricing, cost control, information technology), can be duplicated. Although companies spend huge sums on opening new stores, refurbishing existing stores, improving operations, acquiring the best sites, locking in the best suppliers, increasing distribution efficiency, they are still seeking competence. In search for an enduring formula retailers apply new trends in the marketplace. One such trend is the establishment of loyalty programs. Loyalty programs aim at maximizing customer loyalty thereby maximizing business opportunities with existing consumers over a life time. Simply filling out forms and holding cards, however, does not ensure consumer loyalty. The relationship that customers establish with SAs, however, could impact loyalty and ultimately the bottom line. This study made five practical contributions. First it enabled companies to test how members of loyalty programs, who behave differently, can become more loyal through relationships and specific aspects of trust that may be reinforced. Second, it tests if relationships with SAs contribute to loyalty beyond that gained by the use of external cues. If trustful relationships do have a unique contribution to loyalty, companies can better control relationships to enhance loyalty. Third, it tests which loyalty, that to the SA versus that to the store, is primary. A primary loyalty to the SA may lead to consumer defection when SAs leave, resulting in a loss of future revenues and in no return on expenditures that have been invested in building the relationship thus far. Knowing which loyalty is primary, and how each loyalty is related to sales, companies can estimate potential benefits and risks, and reduce risks of consumer defection if a RS strategy is chosen. Fourth, companies will be able to better direct investments towards selection, training, incentives, and/or promotions in loyalty programs. Finally, incentives that consumers desire and the match between consumer desire and the incentives that SAs provide bring insights as to the reciprocity between SAs and consumers. One important implication of understanding the reciprocity may be the ability to target individuals who are more predisposed to give a company their loyalty. This study explored possible variables that affect performance through RS. Independent variables are consumer desire for service relationships (which incentives does the consumer want), incentives provided to consumers by levels of RS, and perceived quality of service and merchandise. Dependent variables of the study are trust, consumer intentional loyalty to the sales associate, consumer behavioral loyalty to the store, purchase frequency, and individual sales performance. Next, foundations of the study model: relationship bases, relationship development, relationship strength, and trust, are discussed. Bases of Relationships The nature of the customer-supplier interaction has an important effect on relationship formation. Relational behaviors that were found to predict relationship formation are: proactively initiating interactions with customers in order to understand the customer better, signaling impending changes, and disclosing information that puts the sales associate (SA) at risk and implies that the supplier trusts the customer (Leuthesser, 1997). The frequency of the interaction was found not to be significant in predicting the relationship quality. Only when product importance was low, was the influence of social interaction high (Leuthesser, 1997). Attempting to further explain the formation of relational exchanges, Berry (1995) introduced the notion of multiple levels of RS. Each relationship level reflects a different type of bond used to foster customer loyalty. This notion is congruent with findings of other studies (Berry & Parasuruman, 1991; Turnbull & Wilson, 1989) which found that various types of linkages exist between sales associates and their customers. In Level 1, relationship selling relies primarily on pricing incentives to secure customers’ loyalty (e.g., frequent flier programs, a free rental video after ten paid rentals, etc.). From a multiple perspectives view (Linstone, 1984) this level parallels the technical perspective involving electronic data interchanges (EDI), coupons, newsletters, and membership programs designed to enhance customer loyalty. In Level 2, relationship selling relies primarily on social bonds. Social bonds encompass the interpersonal interactions between a SA and her or his customers. Social bonds concern the manner in which the SA relates to the customer at a human level (e.g., courtesy, politeness, getting to know the customer, engagement in a friendly conversation, and exhibition of personal warmth in the SA’s behavior). The role of interpersonal interaction in positively influencing customers was recognized in the service marketing literature (Crosby et al., 1990; Mittal & Lasser 1996). From a multiple perspectives view (Linstone, 1984) this level parallels a personal perspective involving charisma, intuition, and needs of customers and sales associates. In Level 3, relationship selling relies primarily on structural solutions to important customer problems (e.g., personalized service, selling products that are compatible to existing products of the customer, when the customer perceives those as adding superior value, offering value-added benefits that are expensive or difficult for customers to provide, and may not be readily available elsewhere). From a Multiple Perspectives View (Linstone, 1984) this level parallels an organizational perspective involving organizational structure, leadership, decision making which affect the service system. The service leadership rather than individual SAs designs solutions to customers’ problems bonding the customer to both the company and the sales associate. RS Developmental Process Beatty et al. (1996) present the only existing model of relationship development in a retail context. The process is comprised of four phases: initiation, formation, enhancement, and maintenance. The initiation phase begins with the initiative of any party. Either the consumer has a question or request, or the sales associate serves the consumer while initiating communication. The formation phase takes place when the SA recognizes the consumer during a repeat sale. For example, the SA may ask the consumer if she or he wishes to receive a catalog every season or be on the mailing list for special sales. An affirmative answer of the consumer will be followed by a recording of the demographic data of the consumer. The SA then would send the consumer a business card stating the hours that she/he is at the store. After the visit the SA will do something special for the consumer to lock the consumer into the relationship. As the consumer uses the service of the SA, data regarding consumer needs, habits, family members, preferences, styles, are collected. A consumer file is created. The SA notifies the consumer about merchandise that fits the consumer’s preferences and style and may put it on hold for the consumer before it reaches the sales floor. The SA invites the consumer to enjoy special events at the store or at suppliers. The SA also accompanies the consumer throughout all departments and functions as a personal server, solving any problem of the consumer. The enhancement phase takes place as the SA uses exceptional latitude to serve her or his consumers in any way possible to exceed consumer satisfaction. Responding to special needs and in extreme examples of established relationships, the SA decides what to purchase for the consumer, uses a credit card, and delivers the merchandise to the consumer. The maintenance phase is a phase of stability in the relationship. The SA has to creatively maintain it. Although Sheth and Parvatiyar (1996) claim that boredom is one reason for consumer defection and call for variety in incentives, Kahn (1998) contends that there is a limit to variety in promotions and incentives. In either case, RS assists the SA in assessing what the consumer expects and how various incentives can be used to maintain the relationship.
Relationship Strength
While most authors distinguish betweentransactional interactions and relational interactions, Schijns and Schroder (1996) distinguish between relational interactions and strong relational interactions. These authors claim that relationship strength should be measured by both behavioral dimensions and psychological dimensions. They explain that length is only one behavioral indicator of relationship strength. Moreover, monetary value including recency or frequency, reflects the importance and quality of the customer rather than the relationship strength. While importance is measured by the extent to which a customer contributes to the business share, strength is determined by the customer’s perception of the relationship. To measure relationship strength, Wilson (1990) developed a relationship audit that classifies customers by their attitudes towards the relationship. Attitudinal variables that were found to measure relationship strength are: satisfaction, involvement, perceived switching costs, attractiveness, and trust. These psychological dimensions were better indicators of the relationship strength than were behavioral dimensions and monetary value. Relationship strength did not necessarily correspond to relationship length. Among psychological dimensions, trust was the most important predictor of relationship strength (Wilson, 1989) and was therefore the focus of this study. Trust Trust is defined as the belief or the expectation that another can be relied upon with confidence to perform role responsibilities in a fiduciary manner (Morgan & Hunt, 1994; Smith, 1997; Young, Louise, & Wilkison, 1989). Trust is manifested as a willingness to voluntarily increase one’s vulnerability to another (Morgan & Hunt, 1994; Smith, 1997). Conditions of trust that are used to conceptualize the construct are: honesty/integrity, reliability/ dependability, responsibility, competence, judgment, discreetness, concerns, consistency, availability, openness, fairness, and motivations/intentions (Butler; 1991; Morgan & Hunt, 1994; Smith, 1997). The formation of trust is partly explained by openness and communication (Smith, 1997). In the social exchange literature, social exchange is distinguished from the economic exchange in that it engenders reciprocity and feelings of personal obligation, gratitude, and trust (Blau, 1964; Liden, Sparrowe, & Wayne, 1997). In trustful relationships there is a reciprocity between the parties (Liden et al., 1997; Sahlins, 1972). Social exchange theory (e.g., Foa, Converse, Tornblom, & Foa, 1993), classifies resources that may be exchanges between parties into six categories of money, affection, esteem, advice and information, goods, and services. Resources that are exchanges become either more symbolic or a mixture of symbolic and concrete resources (Bauer & Green, 1996; Lewicki & Bunker, 1995; Lewis & Weigert, 1985; McAllister, 1995). Leader member exchange theory (LMX) is a special case of social exchange. LMX is a dyadic approach to leadership (Dienesch & Liden, 1986), which holds the premise that within work units, different types of relationships develop between leaders and each follower. High quality exchanges are characterized by respect, loyalty, liking and mutual trust (Blau, 1964; Graen & Scandura, 1987). In these relationships exchanges of physical or mental effort, material resources, information, or emotional support, take place between the parties. Moreover, the exchange is a trust building experience (Bauer & Green, 1996; Blau, 1964). Thus, trust is a critical foundation in the formation of these exchanges in RS as well (Bitner, 1996; Morgan & Hunt, 1994). These social exchanges are parallel to relational exchanges between SAs and their customers. In each SA- customer dyad, different resources, concrete and symbolic, may be exchanged. Since in RS service is provided by the same SA or service team in all service encounters, the resource exchange is particularistic. In each dyad different resources are exchanged, differentiating one dyad from another. In the marketing literature Morgan and Hunt (1994) and Wilson (1990) showed that once trust was established, the psychological switching costs were so high that the propensity for customers to defect significantly decreased. High psychological switching costs characterize high quality relationships (Ravald & Gronoos, 1996). Examining relationship marketing across partnerships (lateral, internal, buyers, suppliers), Morgan and Hunt (1994) also found trust to be central in facilitating relational exchanges. Applying the Systemic Approach (Lendaris, 1986), when the property of trust exists between consumers and SAs that serve them, linkages among subunits (consumers, SAs) emerge, forming a new structure among SAs and consumers at the B stance level (Lendaris). From separate players who belong to two distinct groups, dyads of people from both groups are formed. Next, the model of the study is presented.
Relationship Selling as a Trust Building Process
In a study by Morgan and Hunt (1994) trust was found to be a key mediating variable in relational exchanges. Undoubtedly, trust or promise keeping (Bitner, 1995) is a particularly important dimension of the relationship from the customer perspective. In this study trust was hypothesized to be a mediator. Trust was hypothesized to be an outcome of the application of various incentive groups as well as an antecedent of loyalty. Trust was hypothesized to be an outcome since feelings of reduced anxiety and confidence in the SA are believed to develop over time, only after a relationship has been established. Trust was also hypothesized to be an antecedent since it pertains both cognition and emotion leading to behaviors (Lewicki & Bunker, 1995; Lewis & Wiegert, 1985). Next, how trust is established across developmental stages of the relationship. Behaviors of consumers and SAs differ across developmental phases of a relationship (Beatty et al., 1996). Underlying these different behaviors are resources that SAs and consumers reciprocally exchange, conditions that may lead to types of trust, and RS levels on which a relationship may be based. Table 3 presents these behaviors, conditions of trust underlying these behaviors, resources exchanged, types of trust, and RS level by relationship phase. In the initiation phase of the relationship the SA provides the consumer with information. The consumer assesses the potential performance of the SA to help her/him in the current shopping task. The SA examines the extent of energy to invest in this transactional exchange. In terms of behavior, the SA informs, guides (Beatty et al., 1996), and promotes higher value products which may have the consumer money. The exchanged resources are primarily concrete resources such as money savings, goods, and standardized services (Foa et al., 1993). The consumer may choose to use or not to use these services, to communicate or not to communicate with the SA, and to purchase or not to purchase at the store. Underlying these decisions is a deterrence/calculative based trust (Lewicki & Bunker, 1995). In deterrence based trust people act in a way that best serves their interest, either in preventing sanctions or in receiving rewards. This author expected that if while the SA provides the core service the consumer attributes functionality and professionalism to the SA (Dienesch & Liden 1986; Graen & Scandura, 1987), a cognition based trust (Lewis & Wiegert, 1985) is established. Antecedents of this cognition based trust are consistency, fairness, and promise fulfillment (Butler, 1991). Therefore, this author expected pricing incentives (RS Level 1) to be related to trust. If a consumer does not desire service relationships with sellers and aspires to receive only pricing incentives from the interaction with a SA, the exchange will remain transactional and will not progress to a relational exchange. Consumers who desire transactional exchanges only, will choose the store on a basis of perceptions of store image, store appearance, merchandise assortment, merchandise quality, service quality, and pricing incentives. These customers, among them loyalty program card holders, might shop for desired products in different stores buying the best deals in each store. Consumer high desire of a service relationship with sellers will lead to the formation of a relational exchange. Thus, this author expected that consumer desire for service relationships (DSR) will moderate the formation of a relational exchange. In the formation phase of the relationship, the SA has more opportunities to serve the consumer, to manage unstructured communication, and to invest energy in relationship building (Beatty et al., 1996). The SA listens and assesses resources of the consumer. The consumer assesses the potential contribution of the SA to satisfy her/his current and future wants and needs (Bauer & Green, 1996). If expectations are met, a relationship is formed. Satisfaction at this phase will facilitate a dependence of the consumer on the SA in the future. With this dependency, the consumer feels more belonging, pleasure, security, and self esteem. Cognition based trust develops into an emotion based trust (Lewis & Wiegert, 1985). Both parties invest energy in learning and knowing the other. Thus, the deterrence based trust now develops into a knowledge based trust (Lewicki & Bunker, 1995). Exchanges comprise symbolic resources in addition to concrete ones which were exchanged in the initiation phase of the relationship (Foa et al., 1993). Potential resources at this phase may be esteem, warmth, and comfort. Antecedents of trust at this phase of the relationship are those which facilitate investment in the relationship: openness, availability, integrity, and social competence (Butler, 1991). These bases of trust are added on to fairness, promise fulfillment, and consistency from the initiation phase of the relationship. Pricing is still important and should be perceived by the consumer as fair and consistent but, the SA is attentive to social incentives as well. The SA uses the consumer’s name, personalizes the relationship, and customizes the service in accordance to consumer wants and needs. In the enhancement phase of the relationship, understanding has been created and as expectations are met or exceeded, the customer-SA bond grows deeper. The consumer exposes information about her/his limitations, needs, and concerns with less selectivity. The relationship is stable and the SA invests less in building the relationship and more in creatively responding to problems and concerns that the consumer expresses (Beatty et al., 1996). The SA may be affectionate towards the consumer, providing all types of resources. The SA uses knowledge about products, services, and markets to best serve the consumer (Berry, 1995). The SA may be perceived as complementing and even replacing the consumer in thinking and solving consumer problems (Beatty et al., 1996). As the SA acts upon her or his genuine concern for the consumer, emotional trust grows deeper into an identification based trust (Lewicki & Bunker, 1995). Since in order to provide consumers with solutions the SA must truly understand the consumer, structural incentives subsume the social incentives which facilitate communication, and lead to a clear understanding of consumer value perceptions, reality, and expectations. Underlying social incentives and structural solution are conditions of trust such as: genuine concern, judgment, and functional competence which are added on to antecedents that underlie former relationship phases. In the maintenance phase the understanding of consumer value perception, and other ways of knowing the consumer, provide a strong foundation for forming, maintaining, and enhancing trust relationships. As the consumer relies upon the SA with confidence, behavior based trust is established. Behavioral trust reinforces the emotional trust, forming the strongest base of trust. Underlying behavioral trust are all trust conditions and therefore, this author expected all incentives/RS levels to underlie trust. The development of trust, however, depends on the consumer desire for service relationships with SAs, therefore, this author expected consumer desire for service relationships to moderate the development of trust. Figure 9 models the hypothesized relationship among consumer desire for service relationships, incentives, and trust. Hypothesis 1a: The relationship between incentives (RS Levels 1, 2 and 3) and trust will be moderated by consumer desire for service relationships such that if consumer desire for service relationships is high, incentives will be related to trust and if consumer desire for service relationships is low, incentives will not be related to trust. Hypothesis 1b: If the relationship between incentives (RS Levels 1, 2, and 3) and trust is not moderated by consumer desire for service relationships, the main effect will be significant such that incentives (RS Levels 1, 2, and 3) will be significant. H1b Figure 9. A Model of the Hypothesized Relationship Among Consumer Desire, Incentives, and Trust Trust and Consumer Loyalty in RS Loyal consumers lead to increased revenues (Riechheld, 1993, 1996; Schlesinger & Heskett, 1991), bring predictable sales and profit streams (Aaker, 1992), are more likely to purchase additional goods and services (Clarck & Payne, 1994; Riechheld, 1996), typically demonstrate low consumer defection (Riechheld & Sasser, 1990), and often, through word of mouth advertisement, generate new business for the company (Zeithaml, Berry & Parasuraman, 1996). Thus, positive effects of consumer loyalty are well supported. As for the study of antecedents of loyalty, only three studies (Ramsey & Sohi, 1997; Sirohi et al., 1998; Ziethaml et al., 1996). examined the question of how consumers develop loyalty (i.e., consumer retention) to a store or to a person. Ramsey and Sohi (1997) found that communication skills of the SA, specifically, listening skills predict consumer loyalty to the SA. The other two studies found loyalty to the store to be determined by perceived quality of merchandise and service. Perceived quality is defined as the consumer’s judgment about the extent of superiority or excellence of the product or the service (Zeithaml, 1988). Perceived quality reflects a user based approach. It is widely believed that consumers use cues to infer quality (Olshavski, 1985; Ziethaml, 1988). For example, cues are classified into intrinsic and extrinsic cues. Intrinsic cues involve physical characteristics of the product. Typically services are intangible and only after the consumer experiences the service, does she/he know its intrinsic cues (e.g., taste). Extrinsic cues include general aspects of the product or service, and may be the variety of brand names, pricing, advertisement, promotions, etc., affecting perceived quality. Most of the research on extrinsic cues has focused on brand names, pricing, store name, and level of advertisement (Dodds et al., 1991; Mazursky & Jacoby, 1985; Nelson, 1974; Rao & Monroe, 1989). The focus of these studies was almost exclusively on the relationship between perceived relative price and quality perception. Ziethaml et al. (1996) examined the relationship between perceived quality of service and loyalty and found a positive relationship between the two. Sirohi et al. (1998) examined three determinants of loyalty to the store: consumer perceived value for relative price, perceived service quality, and perceived merchandise quality. All determinants were measured both for a focal store and for a competing store. Testing these various antecedents of loyalty, Sirohi et al. (1998) surprisingly found the effect of relative price to be the smallest of all other effects on perceived quality (2%). Moreover, only when competition was high, did the small indirect effect of relative price exist. In the study by Sirohi et al. (1998) perceived quality of service explained 56% of the variance in perceived merchandise quality, perceived quality of merchandise explained 22% of the variance in loyalty to the store, and lastly, perceived quality of merchandise and service explained 45% of the variance in loyalty to the store. Thus, relative perceived quality is an important predictor of consumer loyalty. Loyalty may be determined by trust. As an independent variable, in relationship marketing, high trust was found to be related to lower uncertainty, higher stability, and higher satisfaction from relationships (Anderson & Narus, 1990). In other studies trust was found to be positively related to relationship duration (Anderson & Weitz, 1989) and to customer commitment to the relationship (Morgan & Hunt, 1994). These studies, however, focused on industrial relationships (relationship marketing) rather than on relationships between SAs and consumers. Since relationship selling is a special case of relationship marketing, interpersonal trust may effect loyalty in consumer marketing as it does in industrial marketing. Since trust was found to result in higher stability and lower uncertainty, and since higher stability and lower uncertainty were found to be benefits which consumers expect from relational exchanges (Gwinner et al., 1998; Sheth & Parvatiyar, 1996), this author expected trust to have a positive effect on consumer loyalty to the SA and to the store. Looking closely at the potential interplay between trust and perceived quality, however, this author believed that while perceived high quality is expected to contribute to loyalty, perceived low quality, may decrease loyalty despite a trustful relationship. Trust may exist, reflecting a strong relationship, but if the perceived quality is low, purchase frequency may decrease resulting in low loyalty. Thus, this author expected relative perceived quality of merchandise and service to moderate the relationship between trust and loyalty. Figure 10 models the relationship among trust, loyalty, and perceived quality of merchandise and service. Hypothesis 2a: The relationship between trust and loyalty to the SA will be moderated by relative perceived quality of merchandise and service such that when perceived quality is high, there will be a relationship between trust and loyalty to the SA and when perceived quality is low, there will not be a relationship between trust and loyalty to the SA. Hypothesis 2b: If the relationship between trust and loyalty to the SA is not moderated by perceived quality, the main effect will be significant such that trust will be related to loyalty to the SA. Hypothesis 3a: The relationships between trust and loyalty to the store will be moderated by relative perceived quality of merchandise and service such that when perceived quality is high, trust will be related to loyalty to the store and when perceived quality is low, trust will not be related to loyalty to the store. Hypothesis 3b: If the relationship between trust and loyalty to the store is not moderated by perceived quality, the main effect will be significant such that trust will be related to loyalty to the store. H2 a Loyalty to the SA H2b Trust H3a Loyalty to H3b the Store Figure 10. A Model of the Relationship Among Trust, Loyalty, and Perceived Quality Berry (1995) theorized that RS levels predict loyalty and differ on their importance in creating loyalty. Peterson (1995) also theorized a relationship between incentives and loyalty. Peterson theorized that money savings or pricing incentives are a primary motivation of consumers to engage in relationships. Social bonding of a consumer with a SA was found to increase perceived dependence of the consumer on the SA (Cross & Smith, 1995; Moorman et al., 1992). Since perceived dependence was found to be a strong predictor of relationship commitment in business to business relationships (Anderson & Narus, 1990; Dwyer, Schurr, & Oh, 1987; Smith, 1997), this author believes that it may also contribute to relationship commitment (i.e., retention) with consumers as well. Since RS levels were never tested as predictors of loyalty, they may actually be a distinct factor that has been confounded in the consumer perceived quality of service. Thus, loyalty might be determined by trust, perceived quality, and RS levels. Loyalty can be based on a bond to a SA or to the store (Beatty et al,. 1996; Berry, 1995). No empirical investigation concerning types of loyalty has been done. Since Level 3 of RS provides solutions which are facilitated by the store, whereas Level 2 of RS provides incentives that are fully controlled by the SA, this author also expected that loyalty to the store will be stronger in Level 3 of RS, than in Level 2 of RS. The relationship between RS levels and loyalty may also be moderated by relative perceived quality. Only when perceived quality is high, will incentives lead to consumers loyalty to the SA and to the store. Figure 11 models the hypothesized relationship among incentives, loyalty, and perceived quality. Hypothesis 4a: The relationship between structural solutions (RS Level 3) and loyalty to the SA will be moderated by relative perceived quality of merchandise and service such that if relative perceived quality is high, structural solutions will predict loyalty to the SA, and if relative perceived quality is low, structural solutions will not predict loyalty. Hypothesis 4b: If the relationship between social incentives (RS Level 2) and loyalty to the SA is not moderated by perceived quality, the main effect will be significant such that social incentives (RS Level 2) will be related to loyalty to the SA. Hypothesis 5a: The relationship between incentives (RS Level 3) and loyalty to the store will be moderated by perceived quality of merchandise and service such that if relative perceived quality is high, structural solutions will predict loyalty to the SA, and if relative perceived quality is low, structural solutions will not predict loyalty. Hypothesis 5b: If the relationship between structural solutions (RS Level 3) and loyalty to the SA is not moderated by perceived quality of merchandise and service, the main effect will be significant such that structural solutions will be related to loyalty to the SA. Hypothesis 6a: The relationship between social incentives (RS Level 2) and loyalty to the SA will be moderated by perceived quality of merchandise and service such that if relative perceived quality is high, social incentives will predict loyalty to the SA, and if relative perceived quality is low, social incentives will not predict loyalty. Hypothesis 6b: If the relationship between structural solutions (RS Level 3) and loyalty to the store is not moderated by perceived quality of merchandise and service, the main effect will be significant such that structural solutions (RS Level 3) will be related to loyalty to the store. Perceived Quality Incentive (social) H6a H4a H5a H4b Loyalty to Incentive the SA (structural) H5b H6b Loyalty to the Store Figure 11. The Hypothesized Relationship Among Incentives, Loyalty to the SA, Loyalty to the Store and Perceived Quality . Companies strive for consumer loyalty in order to increase profits, market share, and business share. Therefore, the association between incentives, trust, and attitudes of loyalty lacks interest if it does not result in a behavior, specifically, in greater spending of existing customers. Since loyalty was found to be related to higher sales performance (Riechheld & Sasser, 1990; Crosby et al., 1990), and since greater spending was found to be the key predictor of sales performance (Passingham, 1998), this author expected loyalty to positively effect the spending commitment, resulting in a higher sales performance of the SA. Hypothesis 7a: The relationship between loyalty and sales performance of the SA will be mediated by spending commitment. Hypothesis 7b: If no mediation effect is evident spending commitment will be significantly related to sales. Figure 12 presents a general overview of the study model. Consumer Desire Relative Perceived For Service Relationships Quality of Service (high, low) and Merchandise (high, low) H1a H2a, H3a H4a, H5a, H6a Incentives Trust Loyalty (pricing, (SA, store) H1b H2b, H3b SA Sales social H7a Spending Commitment H7b structure H4b, H5b Figure 12. A General Overview of the Study. Note that consumer gives data for all variables except “Sales.” Sales data are collected from the firm and “Incentives” data are collected from both consumers and SAs. CHAPTER SEVEN Methods This study was conducted in two settings: a hair salon chain store, and a travel agency. Approximately 40 hair stylists and at least three of each of their customers and 15 travel agents and at least five of each of their customers were analyzed. This author used both qualitative and quantitative information for data collection. The analysis level was dyadic examining the dynamics of the relationship between a SA and a consumer from perspectives of both parties. This chapter describes the sample population, materials, procedures, and measures. Sample Size and Participants The study was conducted in two companies, each from a different industry. The first company was a National hair salon chain store, positioned as a high quality, superior service marketer, and emphasizes personalized service. Forty stylists that have regular customers and their customers participated in the study. This company is operated on a franchise basis. Eleven salons located in the Pacific Northwest of the United States were included in the study. The second company was a full service travel agency positioned as a superior service provider. The travel agency is a privately owned company. It is rated as one of the top ten agencies in the Northwest employing 15 leisure travel agents and 11 corporate travel agents. Annual sales are $1.8 Billion. A power analysis revealed that at the p<.05 level, based on a one tailed test, a sample of 100 dyads in each setting yields sufficient power to detect a moderate-sized correlation of .19 which is in the recommended acceptable range of power (Cohen, 1977). In both companies, questionnaires were sent to regular customers who purchased services in the past six months. Each customer chose a stylist/agent whom she/he requests when purchasing a service or a product and filled the survey in regard to that stylist/agent. Assuming less than a 20% response rate (Aaker, Kumar, & Day, 1995), for an adequate representation of all SAs, 734 customers were sampled in the hair salon chain store and 330 customers were sampled in the travel agency setting. Respondents residing out of the United States did not participate in this study. For each stylist/agent, information from at least three customers was collected. In the travel agency setting 25 mailings were returned as wrong addresses. One hundred and twenty one respondents out of 305 returned the survey, resulting in a 39% response rate. In the hair salon setting 86 mailings were returned as wrong addresses and 134 surveys out of 648 were completed and returned, resulting in a response rate of 21%. The travel agency sample of 121 respondents was comprised of 62.5% women and 37.5% men. Age ranged from 23 to 87. Table 4 presents the distribution of age by group. Mean education ranging on a scale from none to a Ph.D. degree was a Bachelor’s degree (5 on a 7 point scale). Occupations were 58% professional such as finances (7%), lawyers (5%), executive administrators (13%), executive managers (7%), software and telecommunication company owners (4%), physicians (10%), engineers (8%), scientists (2%), and media (2%). Besides professionals some respondents were in sales (9%), some retired (16%) and some homemakers (8%). The hair salon chain store sample of 134 respondents was comprised of 54% men and 46% women. Age ranged from 16 to 80. Table 4 presents the age distribution by group. Mean education was some college (3 on a 7 point scale). Occupations of the sample population were 36% professional such as teachers (2%), engineers (12%), midlevel managers (14%), executive managers (2%), law (2%), finance (1%), and nurses (3%). A large segment of the sample population was comprised of salespeople (18%). Some respondents were students (4%), soldiers (2%), artists (5%), administrators (4%), constructors (5%), and homemakers (6%). Table 4 Age by Group Age Group Percent In the Travel Sample Percent in the Hair Salon Sample 18-20 0% 5% 21-30 15% 18% 31-40 12% 18% 41-50 23% 26% 51-60 34% 12% 61-70 9% 1% 71-80 5% 5% 81-90 2% 0% Procedures The following procedures were applied in both companies. Customer Recruitment. To encourage customer response, customers were pre-contacted. A letter, signed by both the company owner/manager and a representative from Portland State University, was mailed to customers. The letter explained the survey importance and requested participation. Gillpatrick, Harmon, and Tseng (1994) found that different contacting approaches effect the mail survey response rate. Direct pre-contact of subjects increased the response rate. The response rate of respondents that were pre-contacted was 2.8 times higher than that of those not pre-contacted. The hair salon chain store thanked respondents with coupons. Again, Gillpatrick, Harmon, and Tseng (1994) found that the inclusion of a monetary gift has a strong impact on mail survey response. Among those who were not pre-contacted, a monetary gift doubled the response rate. Among those pre-contacted, the monetary gift increased the response rate by 71% relative to those pre-contacted and not receiving a monetary gift. Both companies provided this author with a database of regular customers. Mailing labels were prepared and questionnaires were mailed to 734 randomly chosen customers at a time, until there were at least three customers per stylist. At the travel agency setting, 300 customers were mailed on the first wave and 30 on the second wave. The second wave mail-out targeted respondents of agents who had a low response in the first wave. After two weeks 600 reminder post cards were sent to the hair salons customers and a 125 reminder post cards were sent to the travel agency customers. Stylists/Agents Recruitment. To encourage employee participation, owners and managers assured stylists and agents that surveys are confidential. After customers returned their surveys, stylists/agents who were identified by customers as their main contact were asked to fill out surveys addressing applied incentives. All customer questionnaires were sent by this author and returned to this author’s home address via enclosed prepaid envelopes. All SAs surveys were distributed to the 11 salons that participated in the study by this author and picked up during the three weeks following the distribution. To better understand the data, this author conducted one hour observations in five out of the eleven salons and interviewed stylists and customers. Since most of the communication in the travel agency is managed through the internet or phone/fax, interviews with only 3 agents were conducted in the travel agency. Materials Two separate surveys were used in this study. Stylists and agents received a survey that addressed demographics, customer orientation of the stylist/agent, applied incentives, and adaptability of the sales style. Customers received a survey that addressed demographics, applied relationship selling (RS) levels, perceived value, perceived relative price, perceived quality of service and merchandise, trust in the SA, desire for service relationships, purchase frequency, intentional loyalty to the SA, and behavioral loyalty to the store. Cover letters were attached to each questionnaire presenting the purposes of the study. For customers the declared purpose was to establish a better understanding of customer wants in order to improve the service. For employees the declared purpose was to better understand how to enhance sales through establishing trustful relationships with consumers. An informed consent form was attached to all surveys, for customers and stylists/agents to sign and return with the surveys. In addition, the number of purchases the consumer made in the past six months, store average dollar purchase amount, consumer average dollar purchase amount, data concerning sales volume per SA by dollars, and overall store sales, which are regularly accumulated, were collected from each company. Items in all questionnaires were grouped by scale. To analyze the customer-SA dyad, questionnaires were coded to accord information from a SA with that of her or his customers. To capture both perspectives, that of the customer and that of the SA regarding incentives provided to customers, both parties stated the extent to which each incentive group is provided. Also, to test the extent to which SAs adjust their sales style to accommodate customer desires from service relationships was tested. A scale that measures adaptability of the sales style was included in the survey. Table 5 presents the study variables by the source of data collection. Table 5 Data Collection by Source DATA COLLECTED SOURCE Identified SA Consumer Relationship Duration Consumer Purchase Frequency Consumer Customer Orientation of Sales People Stylist/Agent Customer Desire for Service Relationships Consumer Relative Perceived Price Consumer Relative Perceived Value Consumer Perceived Service and Merchandise Quality Consumer Trust Consumer Loyalty to the SA Consumer Intentional Loyalty to the Store Consumer Behavioral Loyalty to the Store Consumer Applied levels of RS Stylist, Agent, Consumer Demographics (optional): age, tenure, time spent per service encounter Stylist, Agent Sales Style Adaptability Stylist, Agent Stylist/Agent Sales Performance by Dollars Firm Store Overall Sales Firm Number of Purchases per Consumer Firm Store Average Dollar Purchase Firm Consumer Average Dollar Purchase Firm Measures Table 6 summarizes all scales of the study by source, number of items, and psychometrics. Table 6. Summary Table of Scales by Source, Number of Items, Psychometrics Scale Measures Source Items Alphas In Original Scales Alpha in Study Consumer Desire for Service Relationships Based on literature review (Berry, 1995) and developed by author: Structural solutions, social incentives, pricing incentives 9 Travel: .82, .92, .82Hair:.88, .87, .85 Customer Orientation of Sales People Saxe & Wietz (1982) 12/24 .83 .82, .70 Adapts Spiro & Wietz (1990) 10/16 .85 .90, .79 Relative Perceived Price Sirohi et al. (1998) 5/5 .86 .87, .86 Relative Perceived Value Sirohi et al. (1998) 2/2 RS Incentives Applied Based on literature review (Berry, 1995) and developed by the author 9 .81 .83, .87.92, .88.83, .85 Perceived Service Quality ServiceMerchandise Sirohi et al., (1998) 3/11 .97.97 .84, .87.88, .82. Trust Butler (1991) 22/44 .88 .92, .95 Loyalty to SA Ramsey & Sohi (1997) based on Crosby et al., (1990) 4/4 .97 .87, .87 Behavioral Loyalty to Store Taylor (1998) 4/4 Some measures that were used are self explanatory (i.e., age, education, gender, frequency of providing service to a specific customer, frequency of purchase at the store, time spent with the customer, frequency of interaction with a SA, relationship duration, sales by dollars). Items by scale are outlined in Appendix E. All measures that were used, excluding measures of applied incentives and the desire for service relationships, (RS Levels), were adapted from established scales that are shown to possess good psychometric properties (Appendix E). For all constructs, measures were self-reported by either the consumer or the SA. Responses of SAs have been found to be consistent with those of managers and peers in sales contexts (Churchill et al., 1985; Dubinski et al., 1986; Oliver & Anderson, 1994). In addition, self-reports were found to reveal more information than did other sources of information. To accommodate the companies’ concern regarding the length of the questionnaire, measures contained only positively worded items. For the purpose of this study, two scales were reduced, the sales people customer orientation (SOCO) scale (Saxe & Weitz, 1982) and the conditions of trust inventory (CTI) scale (Butler, 1992). The original scales examine the same attributes using negative and positive statements. The negatively worded items were removed to reduce the scale. Reliability coefficients were pre-tested for these scales and were almost identical to the original reliability coefficients. Control Variables. Consumers responded to optional demographic questions as to their gender, age, and education by years. SA responded to optional demographic questions as to their gender, age, tenure. Other control variables are the service orientation of the stylist/agent (SOCO), and adaptability of the sales style (ADAPTS). Customer Orientation of Salespeople (SOCO). This scale was developed by Saxe and Weitz (1982) and was designed to measure the degree to which a salesperson engages in customer-oriented selling, in helping customers to make purchase decisions that will satisfy customer needs. The scale consists of 22 items related to specific actions that salespeople take when interacting with customers. Only the twelve positive items were used in this study. A sample item was “I try to influence my customers with information rather than by pressure” (Appendix E,12e). These items were scored on a 7 point scale ranging from “1” (true for none of my customers/never) to “7” (true for all of my customers/always). Saxe and Weitz (1982) found the alpha to be .83. Adaptability of Sales Style (ADAPTS). This scale was developed by Spiro and Weitz (1990) and was designed to measure the degree to which salespeople alter their sales behaviors across customer interactions based on perceived information about the nature of the selling situation. The scale consists of sixteen items but only the ten positively worded items were used. A sample item was “ When I feel that my sales approach is not working, I can easily change to another approach” (Appendix E, 13). Items were scored on a 7 point scale ranging from 1 (strongly disagree) to 7 (strogly agree). Cronbach Alpha was .85. Relationship Variables. Consumers indicated how often they contact their SA, how much time, on average, the SA invests in serving them per interaction, how long they have known the SA, and how frequently they shop at the store. Consumer Desire for Service Relationships. Measures incentives (pricing, social, structural) that consumers want from the interaction with a SA. This scale was developed by this author using definitions of RS levels developed by Berry (1995). A list of sixteen items (Appendix C) was generated to assess the construct of RS levels as it was theoretically articulated (Berry, 1995). The list included five items of pricing incentives, seven items of social incentives, and four items of structural solutions. The goal was to choose four items for each RS level (Churchill, 1979; Hinkin, 1998; Thurstone, 1947). The list was independently assessed for content validity by seven colleagues: 4 professors and 3 doctoral candidates. Five items that were found to be conceptually inconsistent were deleted from the list. All other items were assigned by all respondents to their intended construct indicating content adequacy (Schriesheim et al., 1993). After deletion of inconsistent items, there were four items that were 100% conceptually consistent for Levels 1 and 2 of RS. Since pricing incentives are related to the consumer’s perception of added value (Sirohi et al., 1998), 7 items that measure perceived value and relative price were incorporated into Level 1 of RS and will be discussed later. For Level 3 of RS two additional items were generated and reassessed for their conceptual consistency. Level 1 of RS, pricing incentives, included items such as: “It is important to me that my stylist/agent will send me coupons directed at my preference” (Appendix E, 1a). Level 2 of RS, social incentives, included items such as: “It is important to me that my stylist/agent will be attentive to me as an individual” (Appendix E, 2b). Level 3 of RS, structural solutions included one item from the original list and two items that were found consistent with the construct. A sample item for this level is: “It is important to me that my stylist/agent will provide services that are not readily available elsewhere” (Appendix E, 3b). To generate sufficient variance among respondents in the subsequent statistical analysis (Stone, 1978), a five point Likert type scale with a neutral midpoint was used (Harrison & McLaughlin, 1991). The Likert type scale has been found to be the most useful in behavioral science (Kerlinger, 1986), and the most suitable for factor analysis (Hinkin, 1998). The scale ranges from “1” (strong disagreement) to “5” (strong agreement). The questionnaire with the twelve conceptually consistent items was administered to a sample of 67 Business Administration students and again to 200 business students (Appendix D). Once this author identified the study sample, items were adjusted to the study sample population and piloted with 127 marketing students in regard to their hair stylists. Since variance may be low on items on the desired incentives scale and bias the factor analysis, respondents were asked about incentives with which they are provided by their Stylist. Factor analysis showed three factors explaining 60% of the total variance in applied incentives. A simple structure was evident with each cluster loading only on one factor. The first factor explained most of the variance of applied incentives (26%). The second factor explained 23% of the variance, and the third factor explained 20% of the variance. The three clusters reflected three levels or incentive groups: Structural solutions (RS Level 3), pricing incentives (RS Level 1), and social incentives (RS Level 2). Three items loaded on each one of the factors with significant weights ranging from .68 to .83. Thus, the final scale includes nine items. For a full factor analysis please turn to Appendix G. Perceived Relative Price. Sirohi’s et al.’s (1998) Perceived Price Scale was used to measure this construct. Respondents compared the price charged by the focal store to prices charged by alternative stores for similar products. Five items were used. A sample item was “Compared to charges made by alternative stores for similar products this provider is {X}” where X refers to one possible response (Appendix E, 4a). Responses ranged from “1” (much lower priced) to “5” much higher priced. Cronbach Alpha for the relative price scale was .86. Perceived Value. Sirohi’s et al.’s (1998) Perceived Value Scale was used to measured this construct. Respondents compared what they get for what they pay in a focal store to what they get and what they pay in a competitor’s store. Two items were used. A sample items was “comparing what I pay to what I get here to that at the competitor’s store the value here is {X} (Appendix E, 5b). X refers to one possible response. Responses ranged from “1” (excellent) to “5” (poor). Applied Incentives (RS Levels). This construct measures incentives (pricing, social, structural) that stylists/agents provide to their customers. The scale that was used to measure desired incentives in the consumer desire for service relationships scale was adjusted to measure incentives that are actually provided. Thus, this scale was developed by this author using definitions of RS levels developed by Berry (1995). For the process of scale development see Desired Incentives Scale above. Respondents indicated which incentives in each RS level are provided by their SA and to what extent. Nine items were used. A sample item was “My SA is always attentive to me” (Appendix E, 2b). Responses ranged from “1” (strong disagreement) to “5” (strong agreement). Psychometric properties of the scale are reported in Appendix G. Trust. Butler’s (1991) Trust Conditions Inventory was used to measure trust. Respondents responded to statements describing conditions that lead to trust. There were 44 items in the original scale. For this study, negative items were excluded resulting in 22 items that measure 11 conditions of trust. In the original scale respondents indicated their agreement with various statements about a specific manager at work. These statements, however, are worded so they could pertain to individuals in all types of relationships except family or close friendships. A sample item was “If I give confidential information my SA keeps it confidential” (Appendix E, 5h). Responses in the original scale ranged from “1” (strongly disagree) to “7” (strongly agree). In congruence to Hinkin (1993) in this study responses ranged from “1” (strong disagreement) to “5” (strong agreement). Cronbach Alpha of the original scale was .88. Cronbach Alphas were tested to examine the construct validity and inter-item reliability of the reduced scale. Cronbach Alpha was found to be .86. Relative Perceived Quality of Service. Sirohi et al.’s (1998) measure of perceived quality of service was used to measure this construct. Respondents rated the quality of different services that are provided by the store. Five items were used. A sample item was “Overall quality of services provided by personnel I interact with is superior” (Appendix E, 7a). Responses ranged from “1” (strongly disagree) to “5” (strongly agree). Cronbach Alpha of the original scale was .97. Alpha Cronbachs were tested to examine the construct validity and inter-item reliability of the reduced scale. Cronbach Alpha was found to be .81. Relative Perceived Quality of Merchandise. Sirohi et al.’s (1998) measure of perceived quality of merchandise was used to measure this construct. Respondents rated the quality of different products that are provided by the store. Five items were used. A sample item was: “Rotation of items so they are always displayed appropriately” (Appendix E, 8b). Responses ranged from “1” (excellent) to “5” (poor). Cronbach Alpha of the original scale was .97. Alpha Cronbachs were tested to examine the construct validity and inter-item reliability of the reduced scale. Cronbach Alpha was found to be .83. Intentional Customer loyalty to the SA. Ramsey and Sohi’s (1990) scale was used to measure the anticipation of future customer-SA interaction. Respondents indicated their agreement with sentences that reflect a bond with the SA. Four items were used. A sample item was “it is probable that I will contact this SA again” (Appendix E, 9a). Responses ranged from “1” (strong disagreement) to “5” (strong agreement). Cronbach Alpha was .97. Intentional Customer Loyalty to the Store. Sirohi et al.’s (1998) scale measuring loyalty to a supermarket chain was used to measure loyalty. Respondents indicated their agreement with sentences that reflect a bond to the store. Three items were used. A sample item was “I plan to continue traveling with this agency” or “I plan to continue treating my hair at this salon” (Appendix E, 10a). Responses ranged from “1” (extremely likely) to “5” (not at all likely). Cronbach Alpha was .87. Since companies realized that they were losing customers despite the high ratings of satisfaction, the measure of satisfaction has been supplanted with the measure of loyalty. This is primarily because loyalty is seen as being a better predictor of actual behavior. The move to measure loyalty is based on a desire to better understand customer retention which has a direct link to the bottom line. Although the measurement of customer loyalty is closer in its focus to what companies need to know about retention, it is still stated in terms of intentions and thus is still one step removed from actual behavior. Taylor (1998) claims that the concept of loyalty needs to be expanded and made more predictive with additional metrics that contain some element of actual behavior. To apply a more comprehensive and integrated approach, this author used both a traditional, intention- based measure of loyalty, and a new behavior-based measure of loyalty. Behavioral Customer Loyalty to the Store. Taylor’s (1998) behavioral items were used. Respondents indicated the frequency of behaviors that reflect a bond to the store. Three items were used. A sample item was “Have you ever waited to purchase a product or a service even if there was a comparable one available from a competitor?” (Appendix E, 11b). Number of Purchases. This variable was measured by the number of purchases the consumer made in the past six months. Spending Commitment. This variable is measured using the percentage of needs that one fulfills with a single provider. Sales Performance of the SA. This variable was measured by a single indicator of dollars for which the SA is personally responsible. CHAPTER EIGHT Analysis This chapter presents findings of the study. Data screening, factor analysis, reliabilities, were used to examine the data and measures. Multiple regressions, hierarchical regressions, and correlations and structural modeling were used to examine the research questions Prior to the analysis, all variables were examined for accuracy of data entry, missing values, and fit between their distributions and assumptions of multivariate statistics. The variables were examined separately for the 121 cases in the travel industry sample and for the 134 cases in the hair salon sample. Findings are presented separately for each sample. Next, findings from the travel agency sample are described. The Travel Agency Sample Data Screening Missing values were scattered randomly through the data matrix. Nine percent of the missing data was found for the items “the agent keeps secrets I tell her/him, and “the agent would keep confidential information confidential.” Fifteen percent of the missing data was found for the item “Dollar purchase amount” and that variable was therefore dropped from the analysis. All other missing data were lower ranging between 0.09% and 6%. Since overall there was no pattern of missing data, these were replaced with the mean at the item level (Tabachnick & Fidell,1996). One outlier value was identified and corrected with the correct value as it appeared on the survey. All variables were examined for normality, linearity, and homoscedasticity. Normality was assessed by both a statistical method and a graphical method. Many variables were not found to have a symmetric distribution. Skewness ranged between .03 and -18.3. Kurtosis ranged between -0.8 and 48. Since the sample is comprised of customers who have long-term relationships with their agents, it makes sense that most scores are on the high end of the scale. For positively worded items most cases were found to be left of the mean and for negatively worded items most cases were found to be right of the mean. Thus, kurtosis and skewness are high but seem to reflect distributions of long term relationships between customers and their sales associates. A probability plot for each variable showed that all variables were linear. Variances were similar across items in the same scale. Table 7 presents statistics by scale. Table 7 Statistics by Scale Variable Min Max Mean St. Deviation Skewness Kurtosis Age 23 80 49 14 .03 -1.2 Education 2.0 7.0 5.0 1.4 -2.1 -1.2 Perceived Service Quality 3 5 4.5 .63 -.3.8 -1.3 Perceived MerchandiseQuality 3 5 4.3 .66 -2.3 -2.4 Perceived Value 1.5 5 3.8 .83 .36 -1.9 Structural Incentives 1 5 1.8 .97 5.5 1.9 Social Incentives 1 5 4.7 .67 -15 30 Pricing Incentives 1 5 3.2 1.05 -.07 -1.7 Trust SA 3 5 4.5 .50 -2.7 - .08 Loyalty SA 1.5 5 4.7 .59 -11.3 18.1 Loyalty Store 0 3 1.9 .90 -1.3 1.9 Desire for Structural 1 5 1.7 .84 9.0 8.4 Desire for Social 1 5 4.7 .58 - 18.3 47.7 Pricing Desire 1 5 4.3 .95 6.4 4.5 Factor Analysis Factor analysis of the applied incentives scale was performed using the Principal Axis Factoring technique. This technique was chosen over the principal components technique since it does not assume perfect measures but rather that any measure is prone to measurement errors. Communalities, produced by the principal components method, showed three factors with Eigenvalues greater than 1 ranging from 1.7 to 3.6. Three factors explained 67% of the variance in incentives. A direct Oblimin rotation was performed. An Oblimin rotation is a special case of an Oblique rotation. If a rotation is oblique, factors can be correlated and the loading matrix from the orthogonal rotation splits into two matrices: a structure matrix describing relationships between factors and variables, and a pattern matrix of unique relationships which excludes the overlap among factors. Rotation results showed a simple structure with three items loading on each factor, reflecting distinct RS levels that each provide different incentive. Items loaded on one of the factors with weights ranging from .70 to .90. The first factor explained 28.7% of the variance after rotation. The second factor explained 21.1% of the variance in incentives. The third factor explained 24.4% of the variance in incentives. Table 8 presents the factors and their definitions. Table 8 Factors and Definitions Factor 1 Structural Solutions Level 3 of RS: The relationship is based primarily on structural solutions to important customer problems. The marketer can offer target customers value-adding benefits that are difficult or expensive for customers to provide and are not readily available elsewhere. Factor 2 Pricing Incentives Level 1 of RS: The relationship relies primarily on pricing incentives to secure customer loyalty. Factor 3 Social Incentives Level 2 of RS: The relationship relies primarily on social bonds focusing on the human aspects of service. The first factor included the items: ‘The agent does not inform me about the industry’, ‘the agent does not present options and advice about what to do’, and ‘the agent does not offer me value added services’. These items were congruent to Berry’s (1995) description of Level 3 of RS and therefore factor 1 was named Structural Solutions. The second factor included the items: ‘The agent reminds me of relevant benefits’, ‘the agent sends me coupons directed at my preferences’, and ‘the agent invites me to entertainment events in the agency’. One item, ‘the agent helps me save money’ cross loaded on all factors and was deleted from the scale. Items loading on the second factor were congruent to Berry’s (1995) definition of Level 1 of RS and this factor was therefore named Pricing Incentives. The third factor included the items: ‘the agent uses my name during transactions’, ‘the agent is attentive to me as an individual’, and ‘the agent is always friendly’. These items matched the definition of Level 2 of RS (Berry, 1995) and therefore, the third factor was named Social Incentives. Tables 9 presents eigenvalues, Table 10 presents the pattern matrix, and Table 11 presents factor inter-correlations. Table 9 Results of Factor Analysis on Applied Incentives Scale Eigenvalue Variance Pct. Cum Pct. Total Rotated 3.6 35.6 35.6 2.87 2.4 23.9 59.5 2.10 1.7 16.8 76.3 2.44 .60 5.9 82.2 .44 4.3 86.6 .39 3.9 90.5 .32 3.2 93.7 .27 2.7 96.4 .21 2.1 98.4 .15 1.5 100 Table 10 Pattern Matrix Variable 1 2 3 Informs about industry .55 -.12 .73 Presents options -.02 .11 .82 Offers added-value -.01 .08 .85 Attentive to the individual .88 .09 -.03 Friendly .94 -.09 -.08 Uses customer’s name .88 .08 -.02 Sends coupons directed at preferences .03 .87 -.06 Reminds customer of benefits .06 .76 -.14 Helps customer save money .36 .37 -.24 Invites customer to entertainment -.14 .73 .14 Table 11 Inter-Factor Corrrelation Factor 1 2 3 1 1.00 .048 -.37 2 .048 1.00 -.13 3 -.37 -.13 1.00 Study Scales Means were used to compute scores on scales. Scales were created by adding items that belong to each scale and dividing the sum by the number of items. The scales that were created are: perceived value, perceived quality of service, perceived quality of merchandise, structural solutions, social incentives, pricing incentives, trust, loyalty to the SA, loyalty to the store, consumer desire for structural solutions, consumer desire for social incentives, and consumer desire for pricing incentives. Loyalty to the store was comprised of three categorical items. A negative response would be coded as zero and an affirmative response was coded as 1. The items were: Have you waited to purchase an item in the specific store while one was available from a competitor, the number of referrals made in the past six months, the number of family members that use the agency out of total number of family members. Loyalty was defined as a wait for the item, as more than two referrals, and more than two family members. To produce a score on this scale items were summed up and divided to three. Reliabilities of study scales were adequate with Cronbach Alphas ranging from .77 to .95. The original perceived value scale included two items. The scale was expanded for this study to refer to both products and services resulting in four items. These items were: ‘Charges more for similar products’, ‘charges more for similar services’, ‘provides high value’, and ‘provides higher value than that provided by competitors’. Inter- item correlation of items on the perceived value scale were low (a=.19). Using the ‘reliability if item deleted’ coefficient, the scale was reduced to two items: ‘Provides high value’ and ‘provides higher value than that of competitors’. These two items yielded a Cronbach Alpha of .77. Reliabilities of the scales consumer desire for social incentives and consumer desire for pricing incentives were also found to be higher if one item was deleted. Thus, ‘using my name during transactions’ was deleted from the desire for social incentives scale, and ‘helping me save money’ was deleted from the desired pricing incentives scale. Table 12 presents reliabilities and number of items in each scale. Table 12 Reliabilities and Number of Items by Scale (For Cronbach Alphas of the original scales as they appear in the literature please turn to Table 6) Scale Number of Items a in Study Quality of Service 3 .84 Merchandise Quality 3 .77 Perceived Value 2 .78 Structural Incentives 3 .84 Social Incentives 3 .92 Pricing Incentives 3 .82 Trust* 22, 10 .95, .92 Loyalty to the SA 4 .93 Desired Structural Incentives 3 .76 Desired Social Incentives 2 .85 Desired Pricing Incentives 2 .76 *A second trust scale was recreated for this sample due to missing data in the hair care sample. Hypothesis Testing To test Hypothesis 1a, stating that the relationship between incentive groups (RS Levels) and trust is moderated by the consumer desire for service relationships, regression analyses were performed. Each RS level was entered into a different model as an independent variable. To determine unique effects of each RS level on trust in the agent, a hierarchical multiple regression was performed with trust as the dependent variable. To test the effect of structural solutions on trust the control variable perceived value was entered in the first block. The variable structural solutions was entered in the second block. The variable desire for structural solutions in service relationships was entered in the third block. To test the hypothesized moderating effect of consumer desire for structural incentives in service relationships on the relationship between incentives and trust, an interaction term was created (consumer desire for structural solutions ´ structural solution incentives) and entered in the last block of the hierarchical regression analysis. In the first block the overall model was significant (F(1,118)=5.23, R-square=.042, Adjusted R- square =.034, p<.05). Perceived value explained 4.2% of the variance in trust. In the second block the variable structural solutions was added to the equation. The model remained significant. Structural solutions was found to be the strongest predictor of trust uniquely explaining 18% of the variance of trust above and beyond that explained by perceived value (DR=.18, F(2,117)=27.3, p<.001). In the third block the variable consumer desire for structural solutions was added to the model. The change in R-square was not found to be significant (DR=.005, F(1,116)=.78, p=.38). Thus, consumer desire for service relationships did not account for a significant amount of variance above and beyond that explained by perceived value and structural solutions. In the fourth and last step the interaction variable was entered and was not found to be significant (RD=.021, F(3,115)=3.1, p=.077). In the final model only the variable structural solutions was found to be significant. To test the effect of social incentives on trust, social incentives (RS Level 2) was entered as an independent variable into a hierarchical regression model. The variable perceived value was entered in the first block. The overall model was significant (F(1,118) =5.2, R- square =.042, Adjusted R-square=.034, p=.024). Again, perceived value explained 4.2% of the variance in trust. In the second block the variable social incentives was added to the equation. The model remained significant. The variable social incentives added a unique contribution of 12% to the explained variance of trust above and beyond that explained by perceived value (DR=.12, F(1,117)=17.1, p<.001). In the third step the variable consumer desire for social incentives was added to the model. The change in R-square was not found to be significant (DR=.021, F (1,116)=3.0, p=.083). In the fourth and last step, again, the interaction variable was not found to be significant (DR=.022, F(1,115)=3.1, p=.079). In the final model only the variable social incentives was found to be significant. To test the effect of pricing incentives on trust, a third hierarchical regression was performed. The variable perceived value was entered into the equation. In the first block the overall model was significant (F(1,118)=5.2, R-square =.042, Adjusted R- square=.034, p<.05). Perceived value explained 4.2% of the variance in trust. In the second block the variable pricing incentives was added to the model. The model remained significant. Pricing incentives added a unique contribution of 11% to the explained variance of trust above and beyond that explained by perceived value (DR=.011, F(1,117)=15.4, p<.001). In the third step the variable consumer desire for pricing incentives was entered. The change in R-square was not found to be significant (DR=.000 F(1,116)=.027, p=.87). In the fourth and last step, again, the interaction variable was not found to be significant (DR=.001, F(1,115)=.11, p=.74). In the final model only the variable pricing incentives was found to be significant. To compare among RS levels in their effect on trust, all RS levels were entered after perceived value into a hierarchical regression model. Perceived was entered first. The model was found to be significant (F (1,118)= 5.2). Perceived value explained 4.2% of the variance in trust (R-square=.042, Adjusted R-square =.034, p< .05). In the second step all incentives were entered into the model. The model remained significant. Overall all RS levels explained 39.5% of the variance in trust in the SA (F(1,3)=25.2, R-square=.395, Adjusted R-square=.38, p=.000). Structural solutions (Level 3 of RS), negatively worded, was the strongest predictor closely followed by pricing incentives and social incentives. Thus, hypothesis 1 was partially supported. All incentives were positively related to trust but across RS levels, the moderation effect of the variable consumer desire for service relationships was not evident. Thus, the main effect was significant thereby supporting hypothesis 1b. Table 13 summarizes regression coefficients of RS levels on trust. Table 13 Regression of RS Levels 1,2, and 3 on Trust in the SA ____________________________________________________________ ______________________ Variable (s) Beta R-square DR-square DF F Predicting Trust by RS Level 1 Step 1: Perceived Value .07** .04 5.23* Step 2: Structural Solutions -.42 ** .22 .18 27.3 Step 3: Consumer Desire for - .08 .22 .01 .78 Structural Solutions Step 4: Interaction .52 .22 .02 3.18 Overall Equation 11.5** Table 13 Continued ____________________________________________________________ ______________________ Variable (s) Beta R-square DR- square DF F Predicting Trust by RS Level 2 Step 1: Perceived Value .17** .04 5.23* Step 2: Social Incentives .35** .17 .12 17.1 Step 3: Consumer Desire for Social Incentives .15 .17 .02 3.1 Step 4: Interaction - 2.7 .17 .02 3.2 Overall Equation 11.5** Table 13 Continued ____________________________________________________________ _______________________ Variable (s) Beta R-square DR- square DF F Predicting Trust by RS Level 3 Step 1: Perceived Value .15** .04 5.23* Step 2: Pricing Incentives .34** .15 .11 15.4 Step 3: Consumer Desire for Pricing Incentives .01 .15 .00 .027 Step4: Interaction - .17 .15 .00 .110 Overall Equation 10.7* Table 13 Continued ____________________________________________________________ ______________________ Variable (s) Beta R-square R-square D F D F Predicting Trust by All RS Levels Step 1: Perceived Value .02** .042 5.23* Step 2: Structural Solutions - .36** .396 .353 18.8 Social Incentives .27** Pricing Incentives .34** Overall Equation 22.4** Note. Coefficients are with all predictors in the equation n=121. *p<.05 (one tailed). **p<.01 (one tailed). To test the second hypothesis stating that perceived quality of service and perceived quality of merchandise moderate the relationship between trust and loyalty to the SA, a hierarchical regression was performed. The variable loyalty to the SA was entered into the regression equation as the dependent variable, and the variable trust in the SA was entered as the independent variable. To test the moderating effects two interaction terms were created (trust ´ perceived quality of service and trust ´ perceived quality of merchandise). Testing for a moderating effect of perceived quality of service, in the first block the variable perceived quality of service was entered into the model. The overall model was significant (F(1,118)=33.7, R-square=.22, Adjusted R- square=.216, p<.001). In the second block the variable trust was added to the model. The model remained significant with trust adding 33% to the explained variance in loyalty beyond that explained by perceived quality (DR=.33, F(1,117)=86.2, p<.001). In the third block the interaction term was entered into the model. Again, the model remained significant. The interaction uniquely contributed 9% to the explained variance in loyalty (DR=.089, F(1,116)=28.6, p<.001). In the final model all predictors were found to be significant. To analyze the effect of trust on loyalty to the SA by group, data was recoded 1 for the group with low perceived quality of service, 3 for the group with moderate perceived quality of service, and 5 for the group with high perceived quality of service and graphed. Graphing results showed that the relationship between trust and loyalty to the SA was stronger when perceived quality of service was either low or high. Table 14 summarizes regression coefficients. Figure 13 presents regression lines by levels of perceived quality of service. Figure 13: Moderation of perceived quality of service on the relationship between loyalty to the SA and trust. To test for a moderation effect of perceived quality of merchandise on the relationship between trust and loyalty, the control variable perceived quality of merchandise was entered in the first block. The overall model was significant. (F(1,118)=22, R-square=.157, Adjusted R- square=.150, p<.001). In the second block, the variable trust was added to the model. The model remained significant (DR=.396, F(1,117)=103.6, p<.001). Trust added 40% to the explained variance of loyalty above and beyond the contribution of perceived quality of merchandise. In the third block the interaction term with quality of merchandise was entered into the model. Again, the model remained significant (DR=.046, F(1,116)=13.2, p<.001). In the final model all predictors were significant. In order to graph the interaction by group, data was recoded 1 for the group with low perceived quality of merchandise, 3 for the group with moderate perceived quality of service and 5 for the group with high perceived quality of merchandise. The graph showed that the higher the perception of quality the stronger the relationship between trust and loyalty. Thus, hypothesis 2a was not supported for perceived quality of service but was supported for perceived quality of merchandise. The hypothesis stated that the relationship between trust and loyalty to the SA was moderated by perceived quality of service and merchandise such that, only when perceived quality is high the relationship is stronger. The moderation effect of perceived quality of service, however, was evident when perceived quality of service was both low and high. The moderation of perceived quality of merchandise was I the hypothesized direction. Figure 14 presents regression lines for the moderation of perceived quality of merchandise. Figure 14: Moderation of perceived quality of merchandise on the relationship between trust and loyalty Table 14 summarizes regression coefficients of all blocks. Table 14 Regression of Trust in the SA, Interaction of Trust in the SA and Perceived Quality on Loyalty to the SA ____________________________________________________________ ___________ Variable (s) Beta R-square DR- square DF F Predicting Loyalty by Trust Step 1: Perceived Quality of Service .47** .22 33.7** Step 2: Trust in SA .74** .55 .33 86.2 Step 3: Interaction - 4.4** .64 .09 28.6 Overall Equation 69.0** Table 14 continued ____________________________________________________________ ________________________ Variable (s) Beta R-square DR- square DF F Predicting Loyalty by Trust Step 1: Perceived Quality of Merchandise 2.0** .16 22.0** Step 2: Trust in SA 2.1** .56 .40 103.6 Step 3: Interaction - 3.0** .61 .05 13.1 Overall Equation 57.7** Note. Coefficients are with all predictors in the equation n=121. *p<.05 (one tailed). **p<.01 (one tailed). Next, the third hypothesis stating that perceived quality of service and merchandise moderate the relationship between trust and loyalty to the store was tested. Loyalty to the store was entered into the regression equation as the dependent variable, and trust in the SA was entered as the independent variable. To test for moderating effects two interaction terms were created. In the first hierarchical regression the moderation effect of perceived quality of service was tested. In the first block the variable perceived quality of service was entered. The overall model was significant explaining 5% of the variance in loyalty to the store (F(1,118)=5.8, R- square =.047, Adjusted R-square=.039, p=.017). In the second block the variable trust was entered into the model. The model remained significant (DR=.040, F(1,117)=5.1, p=.026). Trust explained 4% of the variance in loyalty to the store beyond that explained by perceived quality of service. In the third block the interaction term was added into the model. The change in R-square was not found to be significant (DR=.000, F(1,116)=.009, p=.92). In the final model trust was the sole predictor. To test for the moderation effect of perceived quality of merchandise on the relationship between trust and loyalty to the store, another regression analysis was performed. The variable perceived quality of merchandise was entered in the first block. The overall model was not found to be significant (F(1,118)=2.5, R- square=.021, Adjusted R-square=.013, p=.113). In the second block, the variable trust was added to the equation. The model was significant. Trust explained 6.4% of the variance of loyalty to the store (DR=.064, F(1,117) =8.2, p<.01). In the third block, the interaction term was added to the model. The change in R-square was not found to be significant (DR=.000, F(1,116)=.014, p=.91), indicating that the moderating effect is not significant. In the final model only trust was significant. Thus, hypothesis 3b was supported. In both cases trust was related to loyalty to the store but the relationship between trust and loyalty to the store was not moderated by perceived quality. Table 15 summarizes regression coefficients of trust on loyalty to the store. Table 15 Regression of Trust, Perceived Quality of Service and Merchandise, and Interaction, on Loyalty to the Store. Variable (s) Beta R-square DR- square DF F Predicting Loyalty by Trust Step 1: Perceived Quality of Service .58** . 05 .05 5.9* Step 2: Trust in SA .26** .09 .04 5.1 Step 3: Interaction - .12 .09 .00 .00 Overall Equation 5.6* Table 15 continued ____________________________________________________________ _________________________ Variable (s) Beta R-square DR- square DF F Predicting Loyalty by Trust Step 1: Perceived Quality of Merchandise - .02 .02 .02 2.5 Step 2: Trust in SA .31** .09 .07 8.3 Step 3: Interaction .15 .09 .00 .01 Overall Equation 5.5* Note. Coefficients are with all predictors in the equation n=121. *p<.05 (one tailed). **p<.01 (one tailed). Hypotheses 4 stated that the relationship between social incentives and loyalty to the SA is moderated by perceived quality. To test hypothesis 4, a regression analysis was performed. Loyalty to the SA was entered as the dependent variable and the variable social solutions as the independent variable. To test the moderating effect of perceived quality of service and merchandise on the relationship between social incentives and loyalty to the SA, again, two interaction terms were created social incentives ´ perceived quality of service, and social incentives ´ perceived quality of merchandise) and tested separately. To test the moderation effect of perceived quality of service, a regression analysis was performed. In first block the variable perceived quality of service was entered. The overall model was found to be significant (F (1,118)=35.7, R-square=.23, Adjusted R-square=.225, p<.001). Perceived quality of service explained 23% of the variance in loyalty to the SA. In the second block the variable social incentives were entered into the model. The change in R-square was not found to be significant (DR=.007, F(1,117)=1.1, p=.295). In the third block the interaction variable was entered and was also not found to be significant (RD=.004, F(1,116) =.58, p=.448). In the final model only perceived quality of service predicted loyalty. To test the moderation effect of perceived quality of merchandise, an additional regression analysis was performed. In first block the variable perceived quality of merchandise was entered and was found to be significant (F(1,118)=23.4, R-square=.164, Adjusted R-square=.157, p<.001). Perceived quality of service explained 16% of the variance in loyalty to the SA. In the second block the variable social incentives were entered into the model and found to be significant (DR=.028, F(1,117)=4.0, p=.047). Social incentives explained 3% of the variance in loyalty to the SA beyond that explained by perceived quality of merchandise. In the third block the interaction effect was tested. The interaction was not found to be significant (DR=.005, F(1,116)=.70, p=.402). In the final model perceived quality of merchandise and social incentives predicted loyalty. Thus, hypothesis 4 was partially supported. Social incentives were related to loyalty when perceived quality of merchandise was in the model, but the relationship was not moderated by perceived quality. Results of the regression of social incentives on loyalty to the SA are summarized in Table 16. Table 16 Regression of Social Incentives, Perceived Quality of Service and Merchandise, and Interaction, on Loyalty to the Sales Associate. ____________________________________________________________ ________ Variable (s) Beta R-square DR-square DF F Predicting Loyalty by Social Incentives Step 1: Perceived Quality of Service .44** .23 35.7** Step 2: Social Incentives .10 .24 .01 1.1 Step 3: Interaction .54 .24 .00 .57 Overall Equation 18.4** Table 16 continued Variable (s) Beta R-square DR-square DF F Predicting Loyalty by Social Incentives Step 1: Perceived Quality of Merchandise .35** .16 23.4** Step 2: Social Incentives .18** .19 .03 4.0 Step 3: Interaction .71 .20 .01 .71 Overall Equation 14.0** Note. Coefficients are with all predictors in the equation n=121. *p<.05 (one tailed). **p<.01 (one tailed). Hypothesis 5a stated that the relationship between structural solutions and loyalty to the SA is moderated by perceived quality of service and merchandise. Again, two interaction terms were created. To test hypothesis 5 the variable structural solutions was entered into an equation as an independent variable predicting loyalty to the SA. In the first block the variable perceived quality of service was entered into the model. The overall model was significant (F(1,119)=35.7, R-square =.23, Adjusted R- square = .225, p<001). In the second block the variable structural solutions was added to the equation. The change in R-square was not found to be significant (DR=.015, F(1,118)=2.3, p=.13) indicating that the variable structural solutions does not account for a significant amount of the variance in loyalty to the SA beyond that explained by perceived quality of service. In the third block the moderation effect was tested. The interaction was not found to be significant (DR=.006, F (1,117)=.97, p=.325). Next, the moderation effect of perceived quality of merchandise on the relationship of structural solutions and loyalty to the SA was tested. The variable perceived quality of merchandise was entered first. Perceived quality of merchandise added 16.4% to the explained variance in loyalty to the SA (F(1,119)=23.4, R-square =.164, Adjusted R-square =.157, p<.001). In the second block the variable structural solutions was entered. The model remained significant with structural solutions uniquely accounting for 4.6% of the explained variance in loyalty to the SA beyond that explained by perceived quality of merchandise (DR=.046, F(1,118)=6.9, p=.10). In the third block, the moderation effect was tested. The model remained significant (DR=.062, F(1,117)=9.9, p<.05) indicating a moderation effect of perceived quality of merchandise. In the final model all predictors were significant. To test the direction of the interaction, regression lines were graphed by group. The variable perceived quality was recoded 1 for low perceived quality of merchandise, 3 for the moderate perceived quality of merchandise, and 5 for the high perceived quality of merchandise. Results of the graphing showed that the relationship between structural solutions and loyalty to the SA is stronger when perceived quality of merchandise is high. Thus, hypothesis 5a was supported. The relationship between structural solutions and loyalty is moderated by perceived quality. Figure 15 presents regression lines for the moderation of perceived quality of merchandise. Figure 15: Moderation effect of perceived quality of merchandise on the relationship between structural solutions and loyalty Table 17 summarizes results of the regression of trust on loyalty. Table 17 Regression of Structural Solutions, Perceived Quality of Service and Merchandise, and Interaction, on Loyalty to the SA ____________________________________________________________ _________________________ Variable (s) Beta R-square DR-square DF F Predicting Loyalty by Structural Solutions Step 1: Perceived Quality of Service .42** .23 35.7** Step 2: Structural Incentives - .14 .25 .02 2.3 Step 3: Interaction .4 7 .25 .01 .97 Overall Equation 19.3** Table 17 continued ____________________________________________________________ ________________________ Variable (s) Beta R-square DR-square DF F Predicting Loyalty by Structural Solutions Step 1: Perceived Quality of Merchandise - .16** .16 23.4** Step 2: Structural Solutions - 1.8** .21 .05 6.9 Step 3: Interaction 1.6** .27 .06 9.9 Overall Equation 14.6** ____________________________________________________________ ___________ Note. Coefficients are with all predictors in the equation n=121. *p<.05 (one tailed). **p<.01 (one tailed). To test Hypothesis 6a stating that the relationship between structural solutions and loyalty to the store is moderated by perceived quality, a hierarchical regression was performed. The variable loyalty to the store was entered as the dependent variable and the variable structural solutions as an independent variable. Again, two interaction terms were created and tested separately. In the first block, the variable perceived quality of service was entered. The overall model was significant (F(1,119)=5.6, R-square=.045, Adjusted R- square=.037, p=.019). Next, the variable structural solutions were entered into the equation. The change in R- square was not found to be significant (DR=.000, F(1,117) =.051, p=.82). In the third block, interaction was added to the model. Again, the change in R-square was not found to be significant (DR=.003, F(1,117)=.37, p=.55). To test the moderation effect of perceived quality of merchandise, another regression analysis was performed. The variable perceived quality of merchandise was entered first and was not found to be significant (F(1,119)=2.5, R- square=.020, Adjusted R-square=.012, p=.119). In the second block the variable structural solutions was entered into the model. The change in R-square was also not found to be significant (DR=.006, F(1,118)=.68, p=.412). In the third block, the interaction variable was added to the model. Again, the model was not found to be significant (DR=.015, F(1,117)=1.7, p=.183). Hypothesis 6 was not supported and thus, the null hypothesis was not rejected. There was no relationship between structural solutions and loyalty to the store. Table 18 presents regression coefficients Table 18 Regression of Structural Solutions, Perceived Quality of Service and Merchandise, and Interaction on Loyalty to the Store. ____________________________________________________________ _______________________ Variable (s) Beta R-square DR-square DF F Predicting Loyalty by Structural Solutions Step 1: Perceived Quality of Service .20* .05 5.6* Step 2: Structural Incentives - .02 .05 .00 .05 Step 3: Interaction .32 .05 .00 .36 Overall Equation 5.6* Table 18 continued ____________________________________________________________ ______________________ Variable (s) Beta R-square DR-square DF F Predicting Loyalty by Structural Solutions Step 1: Perceived Quality .14 .02 2.5 of Merchandise Step 2: Structural Solutions - .08 .03 .01 .68 Step 3: Interaction .75 .04 .02 1.8 Overall Equation 2.5 Note. Coefficients are with all predictors in the equation n=121. *p<.05 (one tailed). **p<.01 (one tailed). To test hypothesis 7a stating that the relationship between loyalty and SA sales performance is mediated by spending commitment, Barron and Kenny’s method (1986) was used in three steps: a) the relationship between loyalty and sales performance was tested b) the relationship between loyalty and spending commitment was tested c) the relationship between loyalty and sales was tested when spending commitment was in the model. In the first step loyalty to the store was not found to be related to sales (F(1,119)=.339, p=ns). In the second step the relationship between spending commitment and loyalty was tested. Loyalty to the store explained 5% of the variance in spending commitment (F (1,119)=6.1, R-square=.049, Adjusted R-square=.041, p<.05). In the third step the relationship between loyalty to the store and sales was tested when spending commitment was in the model (F(1,118)= 2.3, p=ns). Thus, no mediation effect was evident. Loyalty to the SA explained 5% of the variance in spending commitment (F(1,119)=6.6, R-square=.052, Adjusted R-square=.044, p<.05). Loyalty to the SA was not related to sales neither when spending commitment was missing from the model (F(1,119)=.186, p=ns), nor when spending commitment was in the model (F(1,118)=2.8, p=ns). Thus, hypothesis 7a was not supported. Hypothesis 7b stated that if no mediation effect is evident the relationship between spending commitment and sales is significant. To test hypothesis 7b the relationship between spending commitment and sales was examined. Spending commitment explained 4% of the variance in sales (F(1,119)=4.7, p<.05). Thus, Loyalty to the SA and loyalty to the store together explained 9% of the variance in spending commitment which in turn affected the sales. Table 19 presents regression coefficients of spending commitment on sales and of loyalty on spending commitment and sales. Table 19 Regression of Spending commitment and Loyalty to the SA on Sales and of Loyalty to the SA on Spending Commitment ____________________________________________________________ ________ Variable (s) Beta R-square T Sig Predicting Spending Commitment by Loyalty Step 1: Loyalty to the SA .23** .052 2.6 .01 Predicting Sales by Loyalty Step 2: Loyalty to the SA - .04 .002 - .43 .67 Predicting Sales by Loyalty to SA When Spending Commitment is in Model Step 3: Loyalty to SA - .089 .046 - .96 .34 Predicting Sales by Spending .20** .038 2.2 .03 Commitment Table 19 continued ____________________________________________________________ _____________________ Variable (s) Beta R-square T Sig Predicting Spending Commitment by Loyalty to Store Step 1: Loyalty to the Store .60** .36 8.1 .00 Predicting Sales by Loyalty to Store Step 2: Loyalty to the Store .08 .01 . 90 .37 Predicting Sales by Loyalty to Store When Spending Commitment is in Model Step 3: Loyalty to Store with Spending Commitment - .05 .04 - .47 .64 Note. Coefficients are with all predictors in the equation n=121. *p<.05 (one tailed). **p<.01 (one tailed). Structural Equation Modeling This author considered the use of results from the regression analyses to build a CSM model. Since regression analysis is designed to test variables with single indicators, and since regression is designed to test prediction equations with a single dependent variable, CSM was performed. CSM combines factor analysis and regression to examine relationships between one or more independent variables and one or more dependent variables. When some variables in the model are both independent and dependent, CSM is the only method that estimates all relationships simultaneously while estimating and removing errors of measurement and prediction. Moderating and mediating effects can also be tested using the structural model. (Tabachnick & Fidell, 1996). Multivariate normality and linearity were assumed. Prior to the analysis modification indices were requested. Circles in the final model (See Appendix I) represent latent variables (factors) and rectangles represent measured variables. Lines with one arrow are directional paths that imply the direction of the effect of one variable over another. Lines with arrows at both ends are covariate paths that imply covariance between variables with no implied direction of the effect. The Measurement Model The measurement model is derived from a confirmatory factor analysis and relates measured variables to factors. Perceived value had two indicators: High value and higher value. Pricing incentives had three indicators: Reminds, helps, and sends. Social incentives had three indicators: Friendly, attentive, and uses name. Structural solutions, the last incentive group, had three indicators as well: Informs, presents, and offers. Perceived quality of service had three indicators: Response, quality, and customer orientation. Perceived quality of merchandise had three indicators: Merchandise quality, sell high qquality, and brand selection. To simplify the structural model, only one indicator, trust, measured the factor Trusting. Loyalty to the SA had four indicators: Willing to stay with agent, plan to continue doing business with agent, plan to purchase in the future, and plan to contact agent again. Loyalty to the store had three indicators: Waiting for service or merchandise at this agency despite its availability at competitors, number of family members that purchase at the agency out of total number of family members, and number of referrals made in the past six months. Finally, sales had a single indicator: Sales associate annual sales. For the model picture the reader is advised to turn to Appendix I. Table 20 presents the factor loadings for each measured variable with the latent variable on which it loads. Table 20 Loadings of Measured variables Factor Measured Var. Weight Communalities Perceived Value Highvalue 0.79 0.62 Highervalue 0.80 0.64 Pricing Incentives Remind 0.71 0.49 Help 0.74 0.55 Send 0.90 0.81 Social Incentives Friendly 0.91 0.82 Attentive 0.88 0.77 Usesname 0.87 0.76 Structural Solutions Inform 0.70 0.49 Present 0.85 0.71 Offer 0.83 0.69 Service Quality Response 0.76 0.58 Qualitys 0.74 0.55 Customor 0.91 0.85 Merchandise Quality Qualitym 0.79 0.48 Sellhq 0.74 0.72 Brandslc 0.65 0.42 Trusting Trust 1.00 1.00 Loyalty SA Willsty 0.79 0.64 Contbus 0.98 0.96 Purchagi 0.95 0.90 Contacag 0.74 0.55 Loyalty Store Waitserv 0.46 0.01 Famloyal 0.31 0.02 Referral 0.60 0.17 Spending Commitment Percent 1.00 0.70 Sales Sasales 1.00 1.00 All loadings excluding that of one of the variables clustering around the factor loyalty to the store were above .40 and indicate significant relationships between factors and their measured variables. Variables loading on the factor loyalty to the store were dummy coded and then averaged into a scale score ranging from 0 to 1. Thus, the score is in fact a composite score which may explain this low loading on the measured variable ‘family loyalty’. All other communalities indicating the reliability across measures of a latent variable showed that reliabilities ranged between moderate to high. The Structural Model Based on the literature, the hypotheses development process, and results of the regression analyses the model viewed the variable Sales as predicted by three variables: Loyalty to the SA, loyalty to the store, and spending commitment (For a picture of the model the reader is advised to turn to Appendix I). Other dependent variables were loyalty to the SA, loyalty to the store, and trust. Loyalty to the SA was predicted by trust, social incentives, structural solutions, perceived quality of service, and perceived quality of merchandise. Loyalty to the store was predicted by trust, structural solutions, perceived quality of service, and perceived quality of merchandise. Finally, trust was predicted by perceived value, structural solutions, pricing incentives, and social incentives. Table 21 presents squared multiple correlations among latent variables. Table 21 Squared Multiple Correlations Among Latent Variables Factor Explained Independent latent Variables SMC Sales Spending commitment, loyalty to the SA, and loyalty to the store 3% Spending Commitment Loyalty to the SA and loyalty to the store 21% Loyalty to the SA Trust, social incentives, structural solutions, perceived quality of service, and perceived quality of merchandise. 48% Loyalty to the Store Perceived quality of service, perceived quality of merchandise, structural solutions, and loyalty to the SA. 14% Trust Perceived value, pricing incentives, social incentives, and structural solutions. 34% Model Estimation Based upon a maximum likelihood estimation a Chi square was computed when the solution converged. The null hypothesis stating that the model fits the data was rejected (Chi square (70, 295)=442.5, Chi sig<.001). Chi square values depend on the sample size. In models with large samples trivial differences often cause the Chi square to be significant solely because of the sample size. This is supported by Hoelter’s indice for the sample size that would enable analysis of factors without rejecting the null hypothesis (n=96). Additional measures are therefore necessary to estimate the model. The model explained 3% of the variance in sales. Table 22 presents goodness of fit measures across model A, the best possible model (Saturated model) and the worse possible model (Independence Model). Measures based on population discrepancy were not included in the comparison. Table 22 Goodness of Fit Measures Across Models Model Cmin Cmin/df RMR AGFI PNFI RFI CFI A 442.5 1.5 2781.7 0.7 0.7 0.7 0.9 Sat. 0.0 0.000 0.000 1.0 0.0 1.0 1.0 Ind. 2065 5.9 28000 0.26 0.0 0.0 0.0 Table 22 presents four types of measures for examining goodness of fit of a single hypothesized model: Discrepancy measures, baseline measures, and parsimony measures. Sample Discrepancy Measures (CMIN): Are final values of discrepancy which AMOS minimizes. AMOS compares correlations predicted by the model to the observed covariances. CMIN is measured by a Chi Square function. Based on this measure, CMIN/df should yield a ratio of less than 3 to 1 when the model fits the data well. The model shows a ratio smaller than of 1.5 which is within the accepted range. Deviation Measures: Are discrepancy measures that use values of the squared rot of the amount by which the sample variances and covariances differ from their estimates under the model. This value is expressed by Root mean Square residual (RMR) and should be lower than 0.05. Model A shows a huge discrepancy (67021696). Adjusted Goodness of Fit Measures (AGFI): Are discrepancy measures that adjust for degrees of freedom and should be greater than .90 when the model fits the data. Model A is 0.7. All Discrepancy measures show that the covariance matrix of Model A is not close to that of the estimated population. Comparison to Baseline Measures (RFI): Compare the hypothesized model to the independence model. RFI, a normed measure adjusted for degrees of freedom, indicates the location of the model in terms of its distance from the independence model. Model A is 71% away from a worse possible model. Parsimony Adjusted Measures (PNFI, PCFI): Are measures that are adjusted by the degrees of freedom. The value represents the number of distinct parameters being estimated divided by the number of parameters in the independence model. The closer the value of PNFI and PCFI to 0, the better the model. Model A was closer to the Saturated model than it was to the Independence model but was 0.63, and 0.76 respectfully. Based on the analysis off all measures Model A does not fit the data well. Table 23 presents the standardized regression weights. Table 23 Standardized Regression Weights Path Regression Weights T value Pricing to Trust 0.34 3.9 Social to Trust 0.22 2.6 Structural to Trust -0.36 -3.9 Perceived Quality of Service to Loyalty SA 0.44 1.6 Perceived Quality of Merchandise to Loyalty to SA - 0.18 -0.7 Social Incentives to SA Loyalty 0.03 0.4 Structural Solutions to SA Loyalty 0.04 0.4 Trust to Loyalty to SA 0.54 6.7 Perceived Quality of Service to Loyalty to Store 0.50 1.0 Perceived Quality of Merchandise to Loyalty to the Store -0.31 -0.7 Trust to Loyalty to Store 0.09 0.5 Structural Solutions to Loyalty to the Store -0.2 Loyalty to the SA to Spending Commitment 0.25 2.6 Loyalty to the store to Spending Commitment 0.33 1.8 Loyalty SA to Sales -0.07 -0.7 Loyalty to store to Sales 0.02 0.1 Spending Commitment to Sales 0.18 1.7 Table 23 shows that only paths that predicted Trust, Loyalty to the SA and Spending Commitment were found to be significant. The null hypothesis stating that no relationships among the variables in the model exist is therefore, not rejected. Most regression coefficients were not found to be significantly different from zero. Non of the paths explaining sales were found to be significant. As for the prediction of trust, similar to regression results, structural solutions were the strongest predictor of trust (Gamma=-0.36, t=3.6, p<.05). In contrast to the regression results, structural solutions was followed by pricing incentives (Gamma=0.34, t=3.9, p<.05), and only then social incentives (Gamma=0.22, t=2.6, p<.05). As for the prediction of loyalty to the SA, similar to the regression results, trust was the strongest predictor of loyalty to the SA (Beta=0.54, t=6.7, p<.05). Spending commitment was best explained by loyalty to the SA (Beta=0.25, t=2.6, p<.05). Exploratory Analysis At this phase of the analysis the major question was how to modify the model. Modifications were performed in an attempt to develop a better fitting model. All modifications suggest covariate paths between measurement errors and between latent variables (Trusting) and measured variables (attentive) which were not warranted in this recursive model. Thus, the model did not predict loyalty to the store, or sales. This result may be explained by the small sample size versus the large number of variables in the model (70). CSM is based on covariances. Covariances, like correlations, are less stable when estimated from small samples. Moreover, parameter estimates and Chi-Square tests of fit are also very sensitive to sample size. Since in CSM there is no linear relationship between the number of variables and the number of parameters (Bentler, 1995), it is possible to use the number of subjects per estimated variable as the rule of thumb for sample size (Tabachnick & Fidell, 1996). Thus, this sample is comprised of 121 subjects while the adequate number of subjects per estimated parameter for a small to medium size model is 200 (Boomsma, 1983). Thus, the sample size pertains a limitation to the modeling and explains the poor fit of the above model. Findings Summary To sum the findings, six out of seven hypotheses were supported. Hypothesis 1a stated that the relationship between incentives and trust is be moderated by consumer desire for service relationships. Hypothesis 1b stated that if no moderation effect is evident, the relationship between incentives and trust is significant. Hypothesis 1b was supported. All incentives were found to be related to trust. Hypothesis 2a stated that the relationship between trust and loyalty to the SA is be moderated by perceived quality of service and merchandise. Although the moderation effect was evident it was in the opposite direction than that hypothesized when perceived quality of service was in the model. Therefore, hypothesis 2a is partially supported. The relationship between trust and loyalty to the SA was stronger when perceived quality of merchandise was high and when perceived quality of service was both high and low. Hypothesis 3a stated that the relationship between trust and loyalty to the store is moderated by perceived quality. Hypothesis 3b stated that if no moderation effect is evident, the relationship between trust and loyalty to the store is significant. Trust explained loyalty to the store but the relationship between trust and loyalty to the store was not moderated by perceived quality. Therefore, hypothesis 3b was supported. Hypothesis 4a stated that the relationship between social incentives and loyalty to the SA is moderated by perceived quality of service and merchandise. Hypothesis 4b stated that if no moderation effect is evident, the relationship between social incentives and loyalty is significant. The relationship was found to be significant only in case of perceived quality of merchandise but not significant in the case of perceived quality of service. In both cases, the moderation effect was not found to be significant. Thus, hypothesis 4b was partially supported. Hypothesis 5a stated that the relationship between structural solutions and loyalty is moderated by perceived quality of service and merchandise. The relationship between structural solutions and loyalty to the SA was stronger when perceived quality of merchandise was high. Therefore, hypothesis 5a was partially supported. Hypothesis 5b stated that if no moderation effect is evident, structural solutions is related to loyalty to the SA. Hypothesis 5b was supported when perceived quality of service was in the model. Hypothesis 6a stated that the relationship between structural solutions and loyalty to the store is moderated by perceived quality of service and merchandise. Hypothesis 6b stated that if no moderation effect is evident, the relationship between structural solutions and loyalty to the store is significant. Hypothesis 6a and 6b were not supported. Hypothesis 7a stated that the relationship between loyalty and sales will be mediated by spending commitment. Loyalty was found to be related to spending commitment but not to sales. Loyalty to the SA was also found to predict spending commitment but was not found to be related to sales. Thus, no mediation effect was found and therefore, hypothesis 7a was not supported. Hypothesis 7b stated that if no mediation effect is evident the relationship between spending commitment and sales is significant. Spending commitment was found to be related to sales, thereby supporting hypothesis 7b. Table 24 summarizes results of the study. Table 24 Results of the Study by Hypothesis Hypothesis Result 1a Not supported: The relationship between incentives and trust is not moderated by consumer desire for service relationships. 1b Supported: The relationship between RS levels and trust exists 2a Supported: The relationship between trust and loyalty to the SA is moderated by perceived quality. 3a Not supported: The relationship between trust and loyalty to the store is not moderated by perceive quality. 3b Supported: Trust is related to loyalty to the store. 4a Not supported: The relationship between social incentives and loyalty to the SA is not moderated by perceived quality. 4b Supported: The relationship between social incentives and loyalty to the SA exists when perceived quality of merchandise is in the model. 5a Partially supported: The relationship between structural solutions and loyalty to the SA is moderated by perceived quality of merchandise. 5b Partially supported: The relationship between structural solutions and loyalty to the SA exists. 6a Not supported: The relationship between structural solutions ad loyalty to the store is not moderated by perceived quality. 6b Not supported: No relationship between structural solutions and loyalty to the store is evident. 7a Not supported: The relationship between loyalty and sales was not mediated by spending commitment. 7b Supported: The relationship between spending commitment and sales is significant. The Professional Hair Care Sample Data Screening Missing values were scattered through the data matrix for most variables. Across most of the study scales missing values ranged from 0% to 6%. These missing values were replaces with the mean on the item level. On the intentional loyalty scale missing values ranged from 12% to 17% and on the trust scale a pattern of missing values was evident. Twelve variables had missing values ranging between 5% to 38%. The trust scale for the hair-salon sample was therefore, recreated using ten variables with up to 5% missing values. These variables are outlined in Table 25 (For the purpose of comparing between the samples, analyses in the travel sample were performed using both trust scales and will be reported later). Missing values for items on the reduced trust scale were replaced with the mean. All analyses were performed using the Listwise technique. No outliers were identified. Table 25 Variables in the New Trust Scale My SA gives me a fair deal My SA is around when I need her/him My SA performs her/his tasks with skill My SA does things in a consistent manner My SA does things competently My SA deals honestly with me My SA would not do anything to make me look bad My SA makes an effort to understand what I have to say My SA tells me what she/he thinks I trust my agent Study variables were examined for normality, linearity, and homoscedasticity. Normality was assessed by both a statistical method and a graphical method. Many variables were not found to have a symmetric distribution. Skewness ranged between –6.4 and 0.05. Kurtosis ranged between -0.78 and 4.7. A probability plot for each variable showed that all variables were linear. Variances were similar across items in the same scale. Table 26 presents statistics by scale. Table 26 Statistics by Scale Variable Min Max Mean St. Deviation Skewness Kurtosis Perceived Service Quality 1.33 5 3.89 .76 -1.54 0.24 Perceived MerchandiseQuality 2.67 5 3.83 .70 1.14 2.06 Perceived Value 1 5 3.06 .55 0.05 4.6 Structural Incentives 1 5 2.95 1.23 0.28 2.14 Social Incentives 1 5 3.9 1.07 -3.6 0.98 Pricing Incentives 1 5 2.3 1.09 2.25 -0.78 Trust SA 2.6 5 4.06 .81 -.97 2.38 Loyalty SA 1 5 4.08 .99 -3.8 1.27 Desire for Structural 1 5 2.04 1.01 3.1 0.53 Desire for Social 1 5 4.30 .77 -5.6 4.72 Desire for Pricing 1 5 4.37 .87 -6.4 4.00 Factor Analysis Factor analysis of the applied incentives scale was performed using Principal Axis Factoring technique. Factor analysis in this sample showed the same results as those of the travel agency sample. Communalities, produced by the principal components method, showed three factors with Eigenvalues greater than 1 ranging from 1.1 to 4.2. Three factors explained 71% of the variance in incentives. A Direct Oblimin rotation was performed. Rotation results showed a simple structure with three items loading on each factor, reflecting distinct RS levels that each provide different incentive. Items loaded on one of the factors with weights ranging from .69 to .92. After rotation the first factor explained 32.8% of the variance in incentives. The second factor explained 29.6% of the variance in incentives. The third factor contributed 34.4% of the explained variance. The first factor included the items: ‘The stylist does not inform me about the industry’, ‘the stylist does not present options and advice about what to do’, and ‘the stylist does not offer me value added services’. These items were congruent to Berry’s (1995) description of Level 3 of RS and therefore factor 1 was named Structural Solutions. The second factor included the items: ‘The stylist uses my name during transactions’, ‘the stylist is attentive to me as an individual’, and ‘the stylist is always friendly’. These items are congruent to Berry’s (1995) definition of Level 2 of RS and this factor was therefore named Social Incentives. The third factor included the items: The stylist sends me coupons directed at my preferences’, ‘the stylist reminds me of relevant benefits for regulars’ and ‘the stylist helps me save money’. These items match the definition of Level 1 of RS (Berry, 1995) and therefore, the third factor was named Pricing Incentives. Table 27 presents results of the factor analysis. Table 28 presents the pattern matrix. Due to the fact that items on the structural solutions scale are negatively worded, coefficients are negative. Table 29 presents inter-correlations among factors. Due to the negatively worded items, coefficients are negative. Table 27 Factor Analysis on Applied Incentives Scale (Based on Principal Axis Factoring) Eigenvalue Variance Pct. Cum Pct. Total Rotated 4.2 46.8 46.8 2.9 1.8 20.9 67.7 2.7 1.1 12.4 80.1 3.1 .47 5.2 85.3 .40 4.3 89.7 .29 3.2 92.9 .26 2.9 95.8 .22 2.5 98.3 .16 1.8 100 Table 28 Pattern Matrix Item/factor 1 2 3 Informs about industry -.82 .04 -.04 Presents options -.70 -.01 -.07 Offers added-value -.92 .07 .06 Attentive to the individual -.04 .89 -.01 Friendly -.02 .93 -.09 Uses customer’s name .06 .72 .14 Sends coupons directed at preferences .05 -.05 .67 Reminds customer of benefits -.03 .06 .84 Helps customer save money -.09 -.07 .92 Table 29 Inter-Factor Corrrelations Factor 1 2 3 1 1.0 .28 .59 2 .28 1.0 .35 3 .59 .35 1.0 Study Scales Means were used to compute scales scores. Scales were created by adding up items that belong to each scale and dividing the sum by the number of items. The scales that were created are: perceived value, perceived quality of service, perceived quality of merchandise, structural solutions, social incentives, pricing incentives, trust, loyalty to the SA, loyalty to the store, consumer desire for structural solutions, consumer desire for social incentives, and consumer desire for pricing incentives. Loyalty to the store was comprised of three categorical items: the number of referrals made in the past six months, the number of family members that use the salon out of total number of family members, and the wait for a product although it is available from a competitor. Loyalty was defined as greater than two referrals, greater than one member, and a wait for the product although one is available at a competitor. Scales in both samples were identical. Reliabilities of study scales were adequate with Cronbach Alphas ranging from .82 to .95 (sample size for each scale is outlined in Table 30). Reliabilities in this sample were similar to those found in the travel agency sample (See Appendix H). The original perceived value scale included two items. The scale was expanded for this study to refer to both products and services resulting in four items. These items were: ‘Charges more for similar products’, ‘charges more for similar services’, ‘provides high value’, and ‘provides higher value than that provided by competitors’. Inter-item correlation of items on the perceived value scale were low (a=.26). Using the ‘reliability if item deleted’ coefficient, the scale was reduced to two items: ‘Provides high value’ and ‘provides higher value than that of competitors’. These two items yielded a Cronbach Alpha of .86. Due to a high percentage of missing values (40%) on some items in the trust scale Cronbach Alpha was .63. A deletion of the twelve items with missing variables yielded ten distinct conditions of trust with an alpha of .95. Table 30 presents reliabilities, number of items in each scale, and the sample size. Table 30 Reliabilities and Number of Items by Scale Scale Number of Items a Sample Size Quality of Service 3 .88 133 Merchandise Quality 3 .82 115 Perceived Value 2 .86 129 Structural Incentives 3 .87 127 Social Incentives 3 .88 130 Pricing Incentives 3 .85 121 Trust 22, 10 .95, .92 109 Loyalty to the SA 4 .95 107 Desired Structural Incentives 3 .84 127 Desired Social Incentives 3 .88 129 Desired Pricing Incentives 3 .93 128 Hypothesis Testing To test the research questions the same steps used in the travel agency sample were used in this sample. To test the first hypothesis, stating that the relationship between incentive groups (RS Levels) and trust is moderated by the consumer desire for service relationships, regression analyses were performed (n=80). To determine unique effects of each RS level on trust in the stylist, a hierarchical multiple regression was performed with trust as the dependent variable. Each RS level was entered into a different model as an independent variable. The control variable perceived value was entered in the first block. The variable structural solutions was entered in the second block. The variable desire for structural solutions in service relationships was entered in the third block. To test the hypothesized moderating effect of consumer desire for structural incentives in service relationships on the relationship between incentives and trust, an interaction term was created (consumer desire for structural solutions ´ structural solution incentives) and entered in the last block of the hierarchical regression analysis. In the first block the overall model was significant (F(1,78) = 7.9, R-square=.093, Adjusted R- square =.081, p=.006). The variable perceived value explained 9.3% of the variance in trust. In the second block the variable structural solutions was added to the equation. The model remained significant (DR=.178, F(1,77) =18.8, p<.001). Structural solutions was found to be a strong predictor of trust uniquely explaining 27% of the variance of trust above and beyond that explained by perceived value. In the third block the variable consumer desire for structural solutions was added to the model. The change in R-square was not found to be significant (DR=.016 F(1,76)=1.7, p=.20). Thus, consumer desire for service relationships did not account for a significant amount of variance above and beyond that explained by perceived value and structural solutions. In the fourth and last step the interaction variable was entered and was found to be significant (DR=.182, F(1,75)=25.6, p<.001). In the final model all predictors were found to be significant. To test the direction of the interaction data was recoded 1 for low desire for structural solutions and 3 for moderate desire for structural solutions and 5 for high desire for structural solutions. Results showed that the relationship between structural solutions and trust is stronger when the customer desire for structural solutions is high. Figure 16 presents regression lines for the moderation of consumer desire for structural solutions. Figure 16: The Moderation effect of consumer desire for structural solutions on The relationship between structural solutions and trust. To test the effect of social incentives on trust, the variable social incentives (RS Level 2) was entered as an independent variable into a hierarchical regression model (n=81). The variable perceived value was entered in the first block. The overall model was significant (F(1,79) =7.9, R-square =.093, Adjusted R- square=.081, p=.006). In the second block the variable social incentives was added to the equation. The model remained significant. The variable social incentives added a unique contribution of 36% to the explained variance of trust above and beyond that explained by perceived value (DR=.363, F(1,78)=52.1, p<.001). In the third step the variable consumer desire for social incentives was added to the model. The change in R-square was found to be significant (DR=.118, F(1,77)=21.4, p<.001). In the fourth and last step, again, the interaction variable was tested. The moderation effect was not found to be significant (DR=.004, F(1,76)=.804, p=.373). In the final model only social incentives and the consumer desire for social incentives were found to be significant. To test the effect of pricing incentives on trust, a third hierarchical regression was performed (n=76). The variable perceived value was entered into the equation. In the first block the overall model was significant (F(1,76) =7.02, R-square=.085, Adjusted R- square=.072, p=.010). In the second block the variable pricing incentives was added to the model. The model remained significant. Pricing incentives added a unique contribution of 14% to the explained variance of trust above and beyond that explained by perceived value (DR=.142, F(1,75)=13.8, p<.001). In the third step the variable consumer desire for pricing incentives was entered. The change in R-square was found to be significant (DR=.116 F(1,74)=13.0, p=.001). In the fourth and last step, again, the interaction variable was entered and was not found to be significant (DR=.012 F(1,73)=1.3, p=.248). In the final model only pricing incentives and the consumer desire for pricing incentives were found to be significant. To compare among RS levels in their effect on trust, all levels were entered into a multiple regression model (n=85). Overall RS levels explained 51.6% of the variance in trust in the SA (R-square=.516, Adjusted R-square=.498, F(1,3)=29.1, p<.001). Social incentives (Level 2 of RS) was the strongest predictor followed by structural solutions. Pricing incentives was not found to be significant. However, results of the hierarchical regression analysis show that hypothesis 1 was supported. All incentives were positively related to trust. Also, the relationship between structural incentives and trust was moderated by the consumer desire for service relationships. Table 31 summarizes regression coefficients of RS levels on trust. Table 31 Regression of RS Levels 1,2, and 3 on Trust in the SA ____________________________________________________________ ______________________ Variable (s) Beta R-square DR- square DF F Predicting Trust by RS Level 1 Step 1: Perceived Value .30** .09 7.96* Step 2: Structural Solutions - .42** .27 .18 18.7 Step 3: Consumer Desire for - .14 .29 .02 .1.7 Structural Solutions Step 4: Interaction .1.7 .47 .18 Overall Equation 25.6** Table 31 Continued ____________________________________________________________ _____________________ Variable (s) Beta R-square DR- square DF F Predicting Trust by RS Level 2 Step 1: Perceived Value .30** .09 .07 Step 2: Social Incentives .62** .46 .36 32.7 Step 3: Consumer Desire for Social Incentives .37 .58 .12 34.6 Step 4: Interaction .69 .58 .00 Overall Equation 26.1** Table 31 Continued ____________________________________________________________ ______________________ Variable (s) Beta R-square DR- square DF F Predicting Trust by RS Level 3 Step 1: Perceived Value .30** .09 7.01* Step 2: Pricing Incentives .39** .23 .14 11.0 Step 3: Consumer Desire for Pricing Incentives .34** .35 .12 12.8 Step4: Interaction 1.0 .35 .01 Overall Equation 10.0* Table 31 Continued ____________________________________________________________ ______________________ Variable (s) Beta R-square R-square D F D F Predicting Trust by All RS Levels Step 1: Perceived Value .12** .09 7.01* Step 2: Structural Solutions - .22** .51 .43 19.7 Social Incentives .51** Pricing Incentives .11 Overall Equation 21.2** ____________________________________________________________ ______________________ Note. Coefficients are with the predictor structural solutions n=80. Coefficients with the predictor social incentives n=81. Coefficients with the predictor pricing incentives n=76. Coefficients with all predictors in the equation n=85). *p<.05 (one tailed). **p<.01 (one tailed). To test the second hypothesis stating that perceived quality of service and perceived quality of merchandise moderate the relationship between trust and loyalty to the sales associate, a hierarchical regression was performed. The variable loyalty to the sales associate was entered into the regression equation as the dependent variable, and the variable trust in the SA was entered as the independent variable (n=83). To test moderating effects two interaction terms were created. (trust ´ perceived quality of service and trust ´ perceived quality of merchandise). Testing for a moderating effect of perceived quality of service, in the first block the variable perceived quality of service was entered into the model. The overall model was significant (F(1,83)=56.8, R-square=.406, Adjusted R- square =.39, p<.001). In the second block the variable trust was added to the model. The model remained significant with trust adding 36% to the explained variance in loyalty beyond that explained by perceived quality (DR=.363, F(1,82)=129.4, p<.001). In the third block the interaction term was entered into the model. Again, the model remained significant. The interaction uniquely contributed 2.3% to the explained variance in loyalty (DR=.023, F(1,81)=8.9, p<.05). In the final model all predictors were found to be significant. To Analyze the effect of trust on loyalty to the SA by group, data was recoded 1 for the group with low perceived quality of service and 3 for the group with moderate perceived quality of service and 5 for the group with high perceived quality of service. The analysis showed that the relationship between trust and loyalty to the SA was stronger when perceived quality of service was low. Coefficients summary is presented in Table 32. Figure 17 presents regression lines for the moderation. Figure 17: Moderation of perceied quality of service on the relationship between loyalty to the SA and trust To test for a moderation effect of perceived quality of merchandise on the relationship between trust and loyalty, the control variable perceived quality of merchandise was entered in the first block (n=76). The overall model was significant. (F(1,76)=30.6, R-square=.287, Adjusted R- square=.278, p<.001). In the second block, the variable trust was added to the model. The model remained significant (DR=476, F(1,75)=150.7, p<.001). Trust added 48% to the explained variance of loyalty above and beyond the contribution of perceived quality of merchandise. In the third block the interaction term with quality of merchandise was entered into the model. The model was not found to be significant (DR=.11, F(1,74)=3.6, p=.061). In the final model only trust was significant. Table 32 Regression of Trust in the SA, Interaction of Trust in the SA and Perceived Quality on Loyalty to the SA ____________________________________________________________ ___________ Variable (s) Beta R-square DR- square DF F Predicting Loyalty by Trust Step 1: Perceived Quality of Service .64** .41 56.8** Step 2: Trust in SA .89** .77 .36 129.1 Step 3: Interaction - 1.2 .77 .02 8.9 Overall Equation 103.2** Table 32 continued ____________________________________________________________ _____________________________________________ Variable (s) Beta R-square DR-square DF F Predicting Loyalty by Trust Step 1: Perceived Quality of Merchandise .54** .29 30.6** Step 2: Trust in SA .88** .76 .48 150.7 120.8 Step 3: Interaction - 1.3** .77 .01 3.6 Overall Equation 84.5** Note. Coefficients are with all predictors in the equation n=76. *p<.05 (one tailed). **p<.01 (one tailed). Next, the third hypothesis stating that perceived quality of service and merchandise moderate the relationship between trust and loyalty to the store was tested (n=87). Loyalty to the store was entered into the regression equation as the dependent variable, and trust in the SA was entered as the independent variable. To test for moderating effects two interaction terms were created. In the first hierarchical regression the moderation effect of perceived quality of service was tested. In the first block the variable perceived quality of service was entered. The overall model was significant explaining 5% of the variance in loyalty to the store (F (1,85)= 4.0, R-square =.045, Adjusted R-square=.034, p>.049). In the second block the variable trust was entered into the model. The change in R-square was not found to be significant (DR=.003, F(1,84)=.307, p=.581). In the third block the interaction term was added into the model. Again, the change in R-square was not found to be significant (DR=.001, F(1,83)=.054, p=.816). In the final model perceived quality of service was the sole predictor. To test for the moderation effect of perceived quality of merchandise on the relationship between trust and loyalty to the store, another regression analysis was performed (n=89). The variable perceived quality of merchandise was entered in the first block. The overall model was not found to be significant (F(1,74) =1.18, R-square=.016, Adjusted R-square=.003, p=.28). In the second block, the variable trust was added to the equation. The model was not found to be significant (DR=.006, F(1,73)=.47, p=.51). In the third block, the interaction term was added to the model. The change in R- square was not found to be significant (DR=.031, F(1,72) =2.3, p=.13), indicating that the moderating effect is not significant. In the final model none of the predictors were significant. Thus, hypothesis 3 was not supported. Table 33 summarizes regression coefficients of trust on loyalty to the store. Table 33 Regression of Trust, Perceived Quality of Service and Merchandise, and Interaction, on Loyalty to the Store. Variable (s) Beta R-square DR- square DF F Predicting Loyalty by Trust Step 1: Perceived Quality of Service .21* .05 .045 4.0* Step 2: Trust in SA .09 .05 .003 .31 Step 3: Interaction - .19 .05 .001 .05 Overall Equation 1.4 Table 33 continued ____________________________________________________________ ______________________ Variable (s) Beta R-square DR- square DF F Predicting Loyalty by Trust Step 1: Perceived Quality of Merchandise .13 . 02 1.2 Step 2: Trust in SA .06 .02 .01 .45 Step 3: Interaction .09 .05 .03 2.4 Overall Equation 1.4 ____________________________________________________________ _______ Note. Coefficients are with the predictor perceived quality of service n=87. Coefficients with the predictor perceived quality of merchandise n=89. *p<.05 (one tailed). **p<.01 (one tailed). Hypotheses 4 stated that the relationship between social incentives and loyalty to the SA is moderated by perceived quality. To test hypothesis 4, a regression analysis was performed (n=90). Loyalty to the SA was entered as the dependent variable and the variable social incentives as the independent variable. To test the moderating effect of perceived quality of service and merchandise on the relationship between social incentives and loyalty to the SA, again, two interaction terms were created (social incentives ´ perceived quality of service, and social incentives ´ perceived quality of merchandise) and tested separately. To test the moderation effect of perceived quality of service, a regression analysis was performed. In first block the variable perceived quality of service was entered. The overall model was found to be significant (F (1,87)=62.3, R-square=.419, Adjusted R-square=.412, p<.001). In the second block the variable social incentives was entered into the model. The change in R- square was found to be significant (DR=.070, F(1,88)=11.8, p=.001). Social incentives added 7% to the explanation of the variance of trust beyond that explained by perceived quality of service. In the third block the interaction variable was entered and was not found to be significant (DR=.003, F(1,87)=.43, p=.51). In the final model only perceived quality of service and social incentives predicted loyalty. To test the moderation effect of perceived quality of merchandise, an additional regression analysis was performed (n=83). In first block the variable perceived quality of merchandise was entered and was found to be significant (F(1,81)=24.8, R-square =.235, Adjusted R- square=.225, p<.001). In the second block the variable social incentives were entered into the model. The model remained significant (DR=.136, F(1,80)=17.2, p<.001). Social incentives added 14% to the explanation of the variance of trust beyond that explained by perceived quality of merchandise. In the third block the interaction effect was tested. The interaction was not found to be significant (DR=.001, F(1,79)=.085, p=.771). In the final model perceived quality of merchandise and social incentives predicted loyalty. Thus, hypothesis 4 was partially supported. Social incentives were related to loyalty but the relationship was not moderated by perceived quality. Results of the regression of social incentives on loyalty to the SA are summarized in Table 34. Table 34 Regression of Social Incentives, Perceived Quality of Service and Merchandise, and Interaction, on Loyalty to the Sales Associate. ____________________________________________________________ _________ Variable (s) Beta R-square DR-square DF F Predicting Loyalty by Social Incentives Step 1: Perceived Quality of Service .64** .41 62.3** Step 2: Social Incentives .32** .48 .07 11.8 Step 3: Interaction - .28 .48 .00 .44 Overall Equation 27.2** Table 34 continued ____________________________________________________________ ______________________ Variable (s) Beta R-square DR-square DF F Predicting Loyalty by Social Incentives Step 1: Perceived Quality of Merchandise .48** .24 24.8** Step 2: Social Incentives .42** .37 .14 17.2 Step 3: Interaction .27 .37 .00 .09 Overall Equation 15.5** Note. Coefficients with the predictor perceived quality of service n=90. Coefficients with the predictor perceived quality of merchandise n=83. *p<.05 (one tailed). **p<.01 (one tailed). Hypothesis 5a stated that the relationship between structural solutions and loyalty to the SA is moderated by perceived quality of service and merchandise. Again, two interaction terms were created. To test hypothesis 5 structural solutions were now entered into the equation as an independent variable predicting loyalty to the SA. In the first block the variable perceived quality of service was entered into the model (n=87). The overall model was significant (F(1,87)=62.7, R-square =.419, Adjusted R-square = .412, p=.279). In the second block the variable structural solutions was added to the equation. The change in R-square was found to be significant (DR=.042, F(1,86)=6.6, p=.012) indicating that the variable structural solutions significantly accounted for 4.2% of the variance in loyalty beyond that explained by perceived quality. In the third block the moderation effect was tested. The interaction was not found to be significant (DR=.012, F(1,85)=1.99, p=.16). Next, the moderation effect of perceived quality of merchandise on the relationship of structural solutions and loyalty to the SA was tested (n=79). The variable perceived quality of merchandise was entered first. Perceived quality of merchandise added 23.6% to the explained variance in loyalty to the SA (F(1,79)=24.3, R- square =.236, Adjusted R-square =.226, p<.001). In the second block the variable structural solutions was entered. The model remained significant with structural solutions uniquely accounting for 9.2% of the explained variance in loyalty to the SA beyond that explained by perceived quality of merchandise (DR=.092, F(1,78)=10.7, p<.05). In the third block, the interaction effect was tested. The model was not found to be significant (RD=.007, F(1,77)D=.755, p=.39) indicating no moderation effect of perceived quality of merchandise. In the final model perceived quality of merchandise and structural solutions were predictors of loyalty. Table 35 summarizes results of the regression of trust on loyalty. Table 35 Regression of Structural Solutions, Perceived Quality of Service and Merchandise, and Interaction, on Loyalty to the SA ____________________________________________________________ _______ Variable (s) Beta R-square DR-square DF F Predicting Loyalty by Structural Solutions Step 1: Perceived Quality of Service .65** .42 62.7** Step 2: Structural Incentives - .22** .42 .04 6.6 Step 3: Interaction .5 2 .42 .01 1.9 Overall Equation 25.4** Table 35 continued ____________________________________________________________ ______________________ Variable (s) Beta R-square DR-square DF F Predicting Loyalty by Structural Solutions Step 1: Perceived Quality of Merchandise .49** .23 24.3** Step 2: Structural Solutions - .31** .24 .09 10.7 Step 3: Interaction . 46 .24 .01 .76 Overall Equation 12.9** ____________________________________________________________ __________ Note. Coefficients with the predictor perceived quality of service n=87. Coefficients with the predictor perceived quality of merchandise n=79. *p<.05 (one tailed). **p<.01 (one tailed). To test the sixth hypothesis stating that the relationship between structural solutions and loyalty to the store is moderated by perceived quality, a hierarchical regression was performed (n=98). Loyalty to the store was entered as the dependent variable and structural solutions as an independent variable. Again, two interaction terms were created and tested separately. In the first block, the variable perceived quality of service was entered. The overall model was significant (F(1,98)=7.2, R-square=.069, Adjusted R-square=.060, p=.008). Next, the variable structural solutions was entered into the equation. The change in R-square was not found to be significant (DR=.005, F(1,97)=.535, p=.47). In the third block, interaction was added to the model. Again, the change in R- square was not found to be significant (DR=.010, F(1,96) =1.07, p=.30). In the final model only perceived quality of service predicted loyalty. To test the moderation effect of perceived quality of merchandise, another regression analysis was performed (n=83). The variable perceived quality of merchandise was entered first and was not found to be significant (F(1,83) =.59, R-square=.007, Adjusted R-square=-.0052, p=.44). In the second block the variable structural solutions was entered into the model. The change in R-square was not found to be significant (DR=.006, F(1,82)=.532, p=.468). In the third block, the interaction variable was added to the model. Again, the model was not found to be significant (RD=.019, F(1,81)D=1.6, p=.210). Thus, hypothesis 6 was not supported and the null hypothesis was not rejected. There was no relationship between structural solutions and loyalty to the store. Table 36 presents regression coefficients. Table 36 Regression of Structural solutions, Perceived Quality and Interaction on Loyalty to the Store. ____________________________________________________________ ________ Variable (s) Beta R-square DR-square DF F Predicting Loyalty by Structural Solutions Step 1: Perceived Quality of Service .26** .07 7.2* Step 2: Structural Incentives - .08 .07 .01 .54 Step 3: Interaction .46 .08 .01 1.1 Overall Equation 2.9 Table 36 continued ____________________________________________________________ ______________________ Variable (s) Beta R-square DR-square DF F Predicting Loyalty by Structural Solutions Step 1: Perceived Quality .08 .0 1 .59 of Merchandise Step 2: Structural Solutions - .08 .01 .01 .53 Step 3: Interaction .79 .03 .02 1.6 Overall Equation .90 Note. Coefficients with the predictor perceived quality of service n=98. Coefficients with the predictor perceived quality of merchandise n=83. *p<.05 (one tailed). **p<.01 (one tailed). To test the seventh hypothesis stating that the relationship between loyalty and SA sales performance is mediated by the spending commitment, Barron and Kenny’s method (1986) was used (n=114). This method is used in three steps: a) the relationship between spending commitment and sales was tested b) the relationship between loyalty and sales was tested c) the relationship between loyalty and sales was tested when spending commitment was in the model. The relationship between spending commitment and sales was not found to be significant (F(1,125)=1.22, p=ns). The relationship between loyalty to the SA and spending commitment was not found to be significant (F(1,89) =.023, p=ns). The relationship between loyalty to the SA and sales was also not found to be significant neither when spending commitment was not in the model (F(1,91) = .020, p=ns), nor when spending commitment was in the model. (F (1,89)=.037, p=ns). Table 32 presents regression results. The relationship between loyalty to the store and spending commitment was not found to be significant (F (1,116) =3.2, p=ns). The relationship between loyalty to the store and sales was not found to be significant neither when spending commitment was missing from the model (F (1,117)=2.25, p=ns), nor when spending commitment was in the model (F(1,114)=1.8, p=ns). Table 37 presents regression coefficients. Table 37 Regression of Spending commitment and Loyalty to the SA on Sales and of Loyalty to the SA on Spending Commitment ____________________________________________________________ ______ Variable (s) Beta R-square T Sig ____________________________________________________________ ____________________ Predicting Spending Commitment by Loyalty to the SA Step 1: Loyalty to the SA .02 .00 .15 .88 Predicting Sales by Loyalty Step 2: Loyalty to the SA - .09 .01 - .93 .35 Predicting Sales by Loyalty to SA When Spending Commitment is in Model Step 3: Loyalty to SA - .01 .00 - .11 .91 Predicting Sales by Spending Commitment .11 .01 1.2 .21 Table 37 continued ____________________________________________________________ ______________________ Variable (s) Beta R-square T Sig Predicting Spending Commitment by Loyalty to Store Step 1: Loyalty to the Store .16 .03 1.8 .07 Predicting Sales by Loyalty to Store Step 2: Loyalty to the Store .14 .02 1.5 .14 Predicting Sales by Loyalty to Store When Spending Commitment is in Model Step 3: Loyalty to Store with .12 .03 1.2 .21 Spending Commitment Since the percentage of missing data in this sample indicated a pattern of missing data the replacement of missing values by the mean was inapplicable. Covariate Structural Modeling (CSM) requires values for all items and therefore, was not be performed for this sample. To further test the possibility to perform a CSM analysis this author considered the combination of samples. Since regression results differed between samples this author did not combine samples for that purpose. Findings Summary To sum the findings, four out of the seven hypotheses were supported. Hypothesis 1a stated that the relationship between incentives and trust is moderated by consumer desire for service relationships was supported for structural solutions. Hypothesis 1b stated that all RS levels are related to trust and was supported for social incentives (RS Level 2) and pricing incentives (RS Level 1). Hypothesis 2a stated that the relationship between trust and loyalty to the SA is moderated by perceived quality of service and merchandise. The moderation of perceived quality of service was evident but the direction of the moderation effect was opposite to that hypothesized and, therefore, hypothesis 2a was not supported. Hypothesis 2b stated that if no moderation effect is evident trust is related to loyalty. When perceived quality of merchandise was in the model hypothesis 2b was supported. Hypothesis 3a stated that the relationship between trust and loyalty to the store is moderated by perceived quality. This hypothesis was not supported. Hypothesis 3b stated that if no moderation effect is evident, trust will be related to loyalty to the store. This hypothesis was not supported. Hypothesis 4a stated that the relationship between social incentives and loyalty is moderated by perceived quality of service and merchandise. This hypothesis was not supported. Hypothesis 4b stated that if no moderation effect is evident social incentives are related to loyalty. Hypothesis 4b was supported. . Hypothesis 5a stated that the relationship between structural solutions and loyalty to the SA is moderated by perceived quality of service and merchandise. Hypothesis 5a was not supported. Hypothesis 5b stated that if no moderation is evident structural solutions are related to loyalty to the SA. Since the relationship between structural solutions and loyalty to the SA was found to be significant but no moderation effect was evident hypothesis 5b was supported. Hypothesis 6a stated that the relationship between structural solutions and loyalty to the store is moderated by perceived quality of service and merchandise. No support was found for hypothesis 6a. Hypothesis 6b stated that if no moderation effect is evident, structural solutions will be related to loyalty to the store. Hypothesis 6b was not supported. Hypothesis 7a stated that the relationship between loyalty and sales is mediated by spending commitment. This hypothesis was not supported. Hypothesis 7b stated that if no mediation is evident spending commitment is related to sales. Hypothesis 7b was not supported. Table 38 summarizes study results by hypothesis. Table 38 Study Results by Hypothesis Hypothesis Results 1a Supported for RS Level 3: The relationship between structural solutions and trust is stronger when customer desire for structural solutions is low. 1b This hypothesis was supported for RS levels 1 and 2. There is a relationship between social incentives and trust and between pricing incentives and trust. 2a Not supported: The relationship between trust and loyalty to the SA is moderated by perceived quality of service but in the opposite direction than that hypothesized. 2b Partially Supported: There is a relationship between trust and loyalty when perceived quality of merchandise is in the model. 3a Not supported: The relationship between trust and loyalty to the store is not moderated by perceived quality. 3b Not supported: There is no evident relationship between trust and loyalty to the store. 4a Not supported: The relationship between social incentives and loyalty to the SA is not moderated by perceived quality. 4b Supported: There is a relationship between social incentives and loyalty to the SA. 5a Not supported: The relationship between structural solutions and loyalty to the SA is not moderated by perceived quality. 5b Supported: There is a relationship between structural solutions and loyalty to the SA. 6a Not supported: The relationship between structural solutions and loyalty to the store is not moderated by perceived quality. 6b Supported: There is a relationship between structural solutions and loyalty to the store. 7a Not supported: The relationship between loyalty and sales is not mediated by spending commitment. 7b Not supported: There is no relationship between spending commitment and sales. Comparison of Findings Between Samples For the purposes of comparison between studies the trust scale which was reformed for the hair care industry was recreated for the travel agency sample as well. Thus, both samples used identical scales. Results using the reduced trust scale in the travel agency sample are presented below. Sample size for all analysis was 121. Testing Hypotheses with the Reduced Trust Scale in the Travel Sample In the first block the overall model was significant (F(1,118)=7.3, R-square=.059, Adjusted R- square =.051, p=.008). Perceived value explained 6% of the variance in trust. In the second block the variable structural solutions was added to the equation. The model remained significant. Structural solutions was found to be the strongest predictor of trust uniquely explaining 14% of the variance of trust above and beyond that explained by perceived value (RD=.14, F(1,117)D=20.9, p<.001). In the third block the variable consumer desire for structural solutions was added to the model. The change in R-square was not found to be significant (RD=.001 F(1,116)D=.14, p=.70). Thus, consumer desire for service relationships did not account for a significant amount of variance above and beyond that explained by perceived value and structural solutions. In the fourth and last step the interaction variable was entered and was not found to be significant (DR=.015, F(1,115)=2.1, p=.14). In the final model only the variable structural solutions was found to be significant. To test the effect of social incentives on trust, social incentives (RS Level 2) was entered as an independent variable into a hierarchical regression model. The variable perceived value was entered in the first block. The overall model was significant (F(1,118) =7.3, R- square =.059, Adjusted R-square=.051, p=.008). Again, perceived value explained 6% of the variance in trust. In the second block the variable social incentives was added to the equation. The model remained significant. The variable social incentives added a unique contribution of 11% to the explained variance of trust above and beyond that explained by perceived value (DR=.11, F(1,117)=15.2, p<.001). In the third step the variable consumer desire for social incentives was added to the model. The change in R-square was not found to be significant (DR=.031, F (1,116)=4.4, p=.037). In the fourth and last step, again, the interaction variable was not found to be significant (DR=.020, F(1,115)=2.9, p=.090). In the final model only the variables perceived value and social incentives were found to be significant. To test the effect of pricing incentives on trust, a third hierarchical regression was performed. The variable perceived value was entered into the equation. In the first block the overall model was significant (F(1,118) =. Fsig. = 7.3, R-square=.059, Adjusted R-square=.051, p=.008). Perceived value explained 4.2% of the variance in trust. In the second block the variable pricing incentives was added to the model. The model remained significant. Pricing incentives added a unique contribution of 11% to the explained variance of trust above and beyond that explained by perceived value (DR=.084, F(1,117)=11.4, p=.001). In the third step the variable consumer desire for pricing incentives was entered. The change in R-square was not found to be significant (DR=.004 F(1,116) =.051, p=.47). In the fourth and last step, again, the interaction variable was not found to be significant (DR=.003, F(1,115)=.36, p=.55). In the final model only the variables perceived value and pricing incentives were found to be significant. To compare among RS levels in their effect on trust, all RS levels were entered after perceived value into a hierarchical regression model. Perceived value was entered first. The model was found to be significant (F(1, 118)= 7.3, Fsig=.008). Perceived value explained 6% of the variance in trust (R-square=.059, Adjusted R-square =.051). In the second step all incentives were entered into the model. The model remained significant (F (3,115) =16.5, FDSig=.000). Overall all RS levels explained 39.5% of the variance in trust in the SA (R-square=.34, Adjusted R-square=.32, p=.000). Structural solutions (Level 3 of RS), negatively worded, was the strongest predictor closely followed by pricing incentives and social incentives. Thus, hypothesis 1b was supported. All incentives were positively related to trust but across RS levels, the moderation effect of the variable consumer desire for service relationships was not evident. Thus, the main effect was significant thereby supporting hypothesis 1b. Table 39 summarizes regression coefficients of RS levels on trust. Table 39 Regression of RS Levels 1,2, and 3 on Trust in the SA ____________________________________________________________ ______________________ Variable (s) Beta R-square DR- square DF F Predicting Trust by RS Level 1 Step 1: Perceived Value .24** .06 7.36* Step 2: Structural Solutions - .39** .20 .14 20.9 Step 3: Consumer Desire for - .03 .20 .00 .14 Structural Solutions Step 4: Interaction .44 .22 .02 2.13 Overall Equation 7.96* Table 39 Continued ____________________________________________________________ ______________________Variable (s) Beta R-square DR-square DF F Predicting Trust by RS Level 2 Step 1: Perceived Value .24** .06 7.36* Step 2: Social Incentives .33** .17 11 15.2 Step 3: Consumer Desire for Social Incentives .18 .20 .03 4.4 Step 4: Interaction - 2.6 .22 .02 2..9 Overall Equation 8.00* Table 39 Continued ____________________________________________________________ ______________________ Variable (s) Beta R-square DR- square DF F Predicting Trust by RS Level 3 Step 1: Perceived Value .24** .06 7.36* Step 2: Pricing Incentives .29** .14 .08 11.4 Step 3: Consumer Desire for Pricing Incentives .18 .15 .00 .512 Step4: Interaction - .32 .15 .00 .368 Overall Equation 5.03* Table 39 Continued ____________________________________________________________ ______________________ Variable (s) Beta R-square R-square D F D F Predicting Trust by All RS Levels Step 1: Perceived Value .24** .06 7.36* Step 2: Structural Solutions - .32** .34 .28 16.5 Social Incentives .26** Pricing Incentives .29** Overall Equation 14..9* ____________________________________________________________ ______________________ Note. Coefficients with all predictors in the equation n=121. *p<.05 (one tailed). **p<.01 (one tailed). To test the second hypothesis stating that perceived quality of service and perceived quality of merchandise moderate the relationship between trust and loyalty to the SA, a hierarchical regression was performed. The variable loyalty to the SA was entered into the regression equation as the dependent variable, and the variable trust in the SA was entered as the independent variable. To test the moderating effects two interaction terms were created (trust ´ perceived quality of service and trust ´ perceived quality of merchandise). Testing for a moderating effect of perceived quality of service, in the first block the variable perceived quality of service was entered into the model. The overall model was significant (F(1,118)=33.7, R-square=.22, Adjusted R- square=.216, p<.001). In the second block the variable trust was added to the model. The model remained significant with trust adding 32% to the explained variance in loyalty beyond that explained by perceived quality (DR=.32, F(1,117)=81.2, p<.001). In the third block the interaction term was entered into the model. The interaction was not found to be significant. (DR=.001, F (1,116)=.19, p=.66). In the final model only trust was found to be significant. To test for a moderation effect of perceived quality of merchandise on the relationship between trust and loyalty, the control variable perceived quality of merchandise was entered in the first block. The overall model was significant. (F(1,118)=22, R-square=.157, Adjusted R-square=.150, p<.001). In the second block, the variable trust was added to the model. The model remained significant (DR=.38, F(1,117)=98.1, p<.001). Trust added 38% to the explained variance of loyalty above and beyond the contribution of perceived quality of merchandise. In the third block the interaction term with quality of merchandise was entered into the model. The interaction was not found to be significant. (DR=.003, F(1,116)=.85, p=.36). In the final model only trust was found to be significant. Table 40 summarizes regression coefficients of all blocks. Table 40 Regression of Trust in the SA, Interaction of Trust in the SA and Perceived Quality On Loyalty to the SA ____________________________________________________________ _______________________ Variable (s) Beta R-square DR- square DF F Predicting Loyalty by Trust Step 1: Perceived Quality of Service .47** .22 33.7** Step 2: Trust in SA .72** .54 .32 81.2 Step 3: Interaction .15 .54 .00 .190 Overall Equation 45.7** Table 40 continued ____________________________________________________________ ______________________ Variable (s) Beta R-square DR- square DF F Predicting Loyalty by Trust Step 1: Perceived Quality of Merchandise .40** .16 22.0** Step 2: Trust in SA .76** .54 .38 98.1 Step 3: Interaction .33 .55 .00 .853 Overall Equation 46.7** Note. Coefficients with all predictors in the equation n=121. *p<.05 (one tailed). **p<.01 (one tailed). Next, the third hypothesis stating that perceived quality of service and merchandise moderate the relationship between trust and loyalty to the store was tested. Loyalty to the store was entered into the regression equation as the dependent variable, and trust in the SA was entered as the independent variable. To test for moderating effects two interaction terms were created. In the first hierarchical regression the moderation effect of perceived quality of service was tested. In the first block the variable perceived quality of service was entered. The overall model was significant explaining 9% of the variance in loyalty to the store (F (1,118)=5.9, R-square = .047, Adjusted R-square=.039, p=.017). In the second block the variable trust was entered into the model. The model remained significant (DR=.042, F(1,117)=5.4, p=.022). Trust explained 4.2% of the variance in loyalty to the store beyond that explained by perceived quality of service. In the third block the interaction term was added into the model. The change in R- square was not found to be significant (DR=.001, F(1,116) =.106, p=.74). In the final model trust was the sole predictor. To test for the moderation effect of perceived quality of merchandise on the relationship between trust and loyalty to the store, another regression analysis was performed. The variable perceived quality of merchandise was entered in the first block. The overall model was not found to be significant (F(1,118)=2.5, R- square=.021, Adjusted R-square=.013, p=.113). In the second block, the variable trust was added to the equation. The model was significant. Trust explained 6.7% of the variance of loyalty to the store (DR=.067, F(1,117) =8.6, p=.004). In the third block, the interaction term was added to the model. The change in R-square was not found to be significant (DR=.002, F(1,116)=.26, p=.61), indicating that the moderating effect is not significant. In the final model only trust was significant. Thus, hypothesis 3 was partially supported. In both cases trust was related to loyalty to the store but the relationship between trust and loyalty to the store was not moderated by perceived quality. Table 41 summarizes regression coefficients of trust on loyalty to the store. Table 41 Regression of Trust, Perceived Quality of Service and Merchandise, and Interaction, on Loyalty to the Store. Variable (s) Beta R-square DR- square DF F Predicting Loyalty by Trust Step 1: Perceived Quality of Service .21** .0 5 5.9* Step 2: Trust in SA .26** .09 .04 5.4 Step 3: Interaction .16 .09 .00 .10 Overall Equation 3.8 Table 41 continued ____________________________________________________________ ______________________ Variable (s) Beta R-square DR- square DF F Predicting Loyalty by Trust Step 1: Perceived Quality of Merchandise .15** .02 .02 2.5 Step 2: Trust in SA .32** .09 .07 8.6 Step 3: Interaction .26 .09 .00 .25 Overall Equation 3.6 Note. Coefficients with all predictors in the equation n=121. *p<.05 (one tailed). **p<.01 (one tailed). Thus, the comparison of the results between samples using identical scales shows that in both samples all RS levels predicted trust. While consumer desire moderated the relationship in the hair salon sample, no moderation effect was evident in the travel agency sample. In both studies trust predicted loyalty and in neither sample was this relationship moderated by perceived quality of merchandise. The relationship between trust and loyalty to the store was found to be significant only in the travel agency sample. No moderation effect was evident. In both samples the relationship between social incentives and loyalty to the SA was significant. In the travel agency sample, however, this relationship was significant only when perceived quality of service was in the model. As for the relationship between structural solutions and loyalty to the SA, in the travel agency sample the relationship was found to be moderated by perceived quality of merchandise. In the hair salons sample, the relationship was found to be significant but no moderation effect was evident. In both samples the relationship between structural solutions and loyalty to the store was not found to be significant. Finally, the relationship between loyalty to the SA or to the store and sales was not found to be significant in either sample. In the travel agency sample however, both loyalty to the SA and loyalty to the store predicted spending commitment and spending commitment uniquely predicted 4% of the sales. Table 42 presents a summary of findings across samples. Table 42 Summary of Findings Across Samples Hypothesis Travel Industry Hair Care Industry 1a: The relationship between incentive groups (RS Levels 1, 2, and 3) is moderated by consumer desire for service relationships. Not Supported Partially supported: the relationship is moderated by consumer desire for structural solutions (RS Level 3). 1b: If no moderation is evident, all incentives groups (RS Levels 1, 2, and 3) are related to trust. Supported for all incentive groups Partially supported: Social Incentives (RS level 2) and Pricing Incentives (RS Level 1) are related to trust. 2a: The relationship between trust and loyalty to the SA is moderated by perceived quality of service and merchandise. Supported. The relationship between trust and loyalty was found to be stronger when perceived quality of service and perceived quality of merchandise were low Partially supported: The relationship between trust and loyalty to the SA is moderated by perceived quality of service. 2b: If no moderation is evident trust will be related to loyalty to the SA. Not supported Partially supported: There is a relationship between trust and loyalty to the SA when perceived quality of merchandise is in the model. 3a: The relationship between trust and loyalty to the store is moderated by perceived quality of service and merchandise. Not supported Not supported 3b: If no moderation is evident trust will be related to loyalty to the store. Supported Not supported 4a: The relationship between social incentives and loyalty to the SA is moderated by perceived quality of service and merchandise. Not supported Not supported 4b: If no moderation is evident social incentives will be related to loyalty to the SA. Supported Supported 5a: The relationship between structural solutions and loyalty to the SA is moderated by perceived quality of service and merchandise. Partially supported: the relationship is moderated by perceived quality of merchandise. The relationship is stronger when perceived quality of merchandise is high. Not supported 5b: If no moderation is evident, structural solutions will be related to loyalty to the SA. Partially supported: The relationship between structural solutions and loyalty to the SA is significant when perceived quality of service is in the model. Supported: 6a: The relationship between structural solutions and loyalty to the store is moderated by perceived quality of service and merchandise. Not supported Not supported 6b: If no moderation effect is evident structural solutions will be related to loyalty to the store. Not supported Not supported 7a: The relationship between loyalty and sales is mediated by spending commitment. Not supported Not supported 7b: The relationship between spending commitment and sales is significant Supported Not supported A comparison of the results between samples shows that in both samples all RS levels predicted trust. While consumer desire for service relationships moderated the relationship in the hair care industry sample, no moderation effect was evident in the travel industry sample. In both samples trust predicted loyalty but only in the travel industry sample this relationship was moderated by perceived quality of merchandise. The relationship between trust and loyalty to the store was found to be significant only in the travel agency sample. No moderation effect was evident. In both samples the relationship between social incentives and loyalty tot he SA was found to be significant. In the travel industry sample, however, this relationship was significant only when perceived quality of merchandise was in the model. As for the relationship between structural solutions and loyalty to the SA, in the travel industry sample the relationships was moderated by perceived quality of merchandise. In the hair care industry sample, the relationship was found to be significant but no moderation effect is evident. In both samples the relationship between structural solutions and loyalty to the store was not found to be significant. Finally, the relationship between loyalty to the SA or to the store and sales was not found to be significant in either sample. In the travel industry sample, however, both loyalty to the SA and loyalty to the store predicted spending commitment and spending commitment uniquely predicted 4% of the sales. CHAPTER TEN Discussion This chapter discusses the study findings, limitations, future research, and managerial implications. This chapter closes with a systemic view of the findings. Figure 18 presents findings discussed in this chapter. Figure 18: Study Findings * Reflects findings for the travel industry sample only. ** Reflects findings for the hair care industry sample only Relationship Antecedents Customer relationships have received considerable attention from both academicians and practitioners (Reynolds & Beatty, 1999). This study contributes to the existing literature of relationship marketing in that it tested theorized yet unexplored relationships between incentives provided to relationship customers and customer loyalty. This study also extends the existing literature of trust. Over the years trust has emerged as a central construct in the study of relationship marketing (Dwyer, Schurr, & Oh, 1987; Morgan & Hunt, 1994). Trust has been examined in the context of bargaining (Schurr & Ozanne, 1985) buyer–seller relationships (Doney & Cannon, 1997; Ganesan, 1994), distribution channels (Anderson & Weitz, 1989; Dwyer & Oh, 1987), and the use of market research (Morman, Desphande, & Zaltman, 1993; Morman, Zaltman, & Desphande, 1992). This study examined groups of incentives as antecedents of trust and the effect of these different incentives groups on loyalty to the SA, loyalty to store, and sales. Findings show that in both samples all RS levels predicted trust. Trust explained loyalty to the SA. Loyalty to the SA, which was also predicted by structural solutions and social incentives, led to greater loyalty to the store. In the travel agency sample loyalty to the SA, loyalty to the store, predicted by trust, perceived quality, and loyalty to the SA also resulted in spending commitment. Finally, spending commitment resulted in higher SA sales. Findings show that in both companies all RS levels predicted trust in the SA. This finding supports Bendapudi and Berry (1997) who proposed that bonding, that may occur in the context of customer relationships, can serve to increase the dependence on the service provider and to build customer trust. Findings also demonstrate, however, that in each company the same incentives were more strongly related to trust. In both samples structural solutions and social incentives (Levels 3 and 2 of RS) were stronger predictors of trust than were pricing incentives (RS Level 1). Predictors of Trust In the travel agency sample structural solutions (RS Level 3) were the strongest predictor uniquely explaining 18% of the variance in trust above and beyond that explained by perceived value. In the hair salons sample social incentives were the strongest predictor uniquely explaining 45.6% of the variance in trust above beyond that explained by perceived value. These findings were reflected in interviews with customers in the hair salons sample who tended to describe their relationship with their stylists in social terms. Furthermore, friendships between long term customers and their stylists were common and thus support Beatty et al.’s findings (1996). Differences between samples regarding the strongest predictor of trust may be explained by relationship duration and market features. In the travel agency sample mean duration of the relationship was 5.2 years with a standard deviation of 5 years, while in the hair salons sample mean duration was 7.5 months with a standard deviation of 5 months. Also, in the travel agency sample maximum duration was twenty-five years while in the hair salon sample maximum duration was two years. Since customer relationships usually develop over time, it is possible that RS levels differ in importance across stages of the relationship development. It is possible that in an early stage of the relationship social incentives are the most appealing for relationship customers and in later stages, when a relationship is established, customer needs are known to the SA and structural solutions become the underlying drive for relationships. As for market features, the travel agency sample represents a high-end market segment. In the travel agency sample mean education was a Bachelor degree (5 on a 1-7 scale) compared to mean education of some college (3 on a 1- 7 scale) in the hair salons sample. In the travel agency sample 58% percent of respondents were professionals compared to only 35% professionals in the hair salons sample. Differences in the level of education may reflect differences in consumer perceptions of service and in desires from service relationships. Thus, in each sample a different RS level was found to be the strongest predictor of trust. This finding supports McAdams (1988), Reynolds and Beatty (1999), and Sheth and Parvatiyar (1995) who theorized that relational customers can receive benefits that serve to fill many human needs. In both samples the relationship between pricing incentives and trust was the weakest among RS levels (11%). Although a moderation effect of relationship duration on the relationship between social incentives and trust as well as on the relationship between structural solutions and trust was not found to be significant, again, it is possible that across stages of a relationship development, incentives differ in their importance. As for the relationship among RS levels, as evident from the inter-correlation among factors, in both the travel agency sample and the hair salon sample, social incentives were positively related to structural solutions which were negatively worded (r=.28;-.36 respectively). Thus, the more the SA is attentive, friendly, and knows the customer by name, the more the SA presents options for the customer, informs the customer of relevant tips, and offers services that would add value for that customer. Likewise, the more a SA recommends options to a particular customer, the more a SA informs the customer of tips that are relevant to her/his situation, and the more the SA offers services that would add value for that customer, the more the SA knows the customer as an individual. This correlation may be explained by the need to communicate with the consumer and to be familiar with consumer lifestyle, preferences, and family members, in order for the SA to design a suitable customized solution for the customer. Thus, when Level 2 of RS is strong, Level 3 of RS is feasible. Likewise, when Level 3 of RS is strong, Level 2 is subsumed in providing structural solutions. Structural solutions were also positively related to pricing incentives (-.071, -.58 respectively). The more the SA reminded the customer of benefits, helped the customer save money, and sent coupons targeted at the customer’s preferences, the more the SA offered value added services, informed the customer of products benefits and presented option for the customer. This correlation may reflect the link between Levels 1 and 3 of RS as Levels that are facilitated by the store policy and are, therefore, distinct from Level 2 of RS which is facilitated by individual SAs and reflects the social aspect of service encounters. This correlation, however, is weaker in the travel agency sample because while in the hair salons setting coupons are offered on a regular basis, in the travel agency sample coupons are occasionally offered reducing the variance and therefore decreasing the correlation. Pricing incentives were positively related to social incentives in the hair salon sample (r=.39) but negatively related to social incentives in the travel agency sample (-.018). The relationship between social incentives and pricing incentives seems to differ across settings and may also be affected by the developmental stage of the relationship. In the hair salon sample the more the stylist knows the customer, and is attentive and friendly to the customer, the higher the probability that the customer might receive coupons for the next purchase. Likewise, the more the customer receives coupons and is reminded of benefits and money savings, the higher the probability that the customer will visit the store more frequently. Frequent visits increase the probability that the stylist will recognize the customer, be friendly, attentive, and use the customer’s name. In the travel agency sample, the more the relationship is based on social incentives the less the relationship is based on pricing incentives and reversed. In this setting pricing incentives may be very important as a drive to engage in relationships but less important as an underlying drive for maintaining an existing relationship. This may be especially true for high-end market segments, where trustful relationships shift the basis of the relationship from a pricing level to a structural solutions level. This may indicate that in high-end segments once a trustful relationship exists, customers are less price sensitive and more solutions oriented. As for the moderation effect, as mentioned above, in the travel agency sample the relationship between all incentives and trust was significant but there was no evidence of a moderation effect. In the hair salons sample hypothesis 1 was partially supported for social and pricing incentives and fully supported for structural solutions. The moderation of the consumer desire for service relationships was significant only for structural solutions. The relationship between structural solutions and trust was stronger when the consumer desire for structural solutions was high. This finding may indicate a phenomenon of exceeding customer expectations leading to greater trust. This supports previous work of Gwinner et al. (1998) and Reynolds and Beatty (1999) who demonstrated that customers not only expect to receive satisfactory core service, but are also likely to receive additional incentives leading to greater trust. The hair salon chain store is positioned as a cost leadership and therefore customers may be surprised when structural solutions are provided. Sheth and Parvatiyar (1995) define the process of exceeding customer expectations as “customer’s delight”. The existence of a moderation effect only in the hair salon chain store may be attributed to the difference in relationship duration and strength. In the travel agency sample most relationships are long-term with customers following their agents from one agency to another purchasing service from a particular agent. In the hair salons sample relationships exist yet, they extend for shorter periods of time and different stylists occasionally provide service. Moreover, due to a high turnover rate at the hair salons sample, the customer never knows if the stylist will still be there the next time that she or he needs service. Thus, there is a selectivity effect as for the desire for service relationships in the travel agency sample. The relationships between all RS levels and trust support Darden and Dorsch (1990), Haley (1968), Gutman (1982), Reynolds and Gutman (1984) and Reynolds and Beatty (1999) who suggested that customers select products and services on the basis of incentives they desire. These are significant findings that may illustrate the importance of the contribution of social incentives and structural solutions on maintaining customer relationships. According to the results of this study all incentives are important in establishing relationships but levels 2 and 3 of RS are stronger predictors of customer trust. RS Levels and the Multiple Perspective View The importance of these findings supports Linstone’s (1984) multiple-perspectives view. The three RS levels reflect different perspectives, ways to look and analyze service environments. Each RS level contributes to building and maintaining relationships. Analyzing a service environment from only one perspective leads to reductionism and negatively affects long term relationships. To view the whole all three perspectives are to be used. Level 1 of RS, pricing incentives reflects a technical, quantifiable, aspect of service. This aspect focuses on a cost-benefit analysis. Level 2 of RS, social incentives involves the human aspect of service. This level reflects the personal perspective and focuses on charisma, intuition, and self-interests. This level is influenced by experience, background, roles, personality structures of players, and the changing environment. Level 3 of RS reflects the organizational perspective and focuses on values, organizational culture, organizational service orientation, politics, interests, communication issues, and organizational responsibilities. Findings of this study highlight the interdependency of all RS levels as a key success factor of relationship selling. The existence of a close SA-consumer relationship, regardless of the RS level on which the relationship is based, reflects the emergence of a new property in the system (Lendaris, 1986). From an A stance of the system the company is a unit and the sales department is a sub- unit. The Supersystem is the retail industry and the supra system is service. The focus of the observer is relationship selling. When sub-units operate together they manifest a whole. The whole is a supersystem that was comprised of unrelated parts that are turning into a network of players. These previously unrelated systems of suppliers, competitors, etc., are becoming inseparable and thus supporting Senge’s (1994) argument that seeing the world as comprised of unrelated parts is only an illusion. From an A stance of the system ‘networking’ is an emergent property of the system. From a B stance of the system, the unit is the sales department and the subunit is the SA-consumer dyad. Sub-units, when jointly operating according to trust principles, do something that is greater than a simple collection of their individual uncoordinated operation. When dyads work in a coordinated manner they contribute together to the emergence of a new whole, to a trust building environment which makes the whole greater than the sum of its parts (Lendaris, 1986). These findings are significant since building customer relationships based on the same incentives without understanding how trust, across RS levels, factors into the relationship from the consumer’s perspective, may lead to relationship formation with all consumers regardless of the consumer desire in a relationship or in a particular incentive group. Findings show that structural solutions, for example, should be used to form trust even when consumers desire for structural solutions is high. Agreement between Perspectives of Parties to the Dyad Linstone’s multiple perspectives view also calls to view incentives from perspectives of both parties of the dyad. Thus, agreement between parties to the dyad on the extent to which incentives are provided may indicate the quality of the relationship (Leuthesser, 1997) and was examined by analyzing data that were collected from both customers and their agents or stylists. In the travel agency sample the correlation between consumer perspective and SA perspectives on applied incentives in Levels 2 (social incentives) and Level 3 of RS (structural solutions) was not significant. The correlation between perspectives of both parties regarding incentives provided in RS Level 1 (pricing incentives) was low (.21, p<.05). Thus, the agreement between the consumer and the SA on which incentives are provided and how much of pricing incentive is provided to the customer was low. This finding is congruent to previous findings of this study in this setting. Pricing incentives were not a drive for maintaining a relationship and were found to be the weakest predictor of trust. Likewise, the correlation measuring the extent of agreement between desired incentives and applied incentives was not significant across RS levels. In the hair salons sample the correlation indicating agreement between parties on their perspective of applied incentives across RS levels was not significant. The correlation indicating the agreement between applied incentives versus desired incentives was not significant in Level 1 of RS (pricing incentives) but was significant for Level 2 (social incentives) and Level 3 of RS (structural solutions). In Level 2 of RS the correlation was high (.34, p<.01) and in Level 3 of RS the correlation was high as well (.39, p<.01). Given multiple sources of data, these correlations are high and reflect an agreement between perspectives of customers and their service providers on applied incentives versus desired incentives. An examination of the extent of agreement between parties to the dyad was also performed by analyzing the ability of travel agents to adjust their sales style according to the customer desires. The analysis showed that adjustment scale mean was 3.54 on a scale of 1 to 7 in the travel agency sample. In the hair salons sample adjustment scores were higher. Scale mean was 4.1. In both samples a few elements in the sales style require improvement and are discussed later. Relationship Consequences Building relationships with customers was thought to increase customer loyalty (Beatty, 1996; Berry & Parasuraman, 1995; Berry, 1995; Czepiel, 1990: Reynolds & Beatty, 1999). Findings of this study show that the second hypothesis stating that the relationship between trust in the SA and loyalty to the SA is moderated by perceived quality was supported. In both samples the relationship between trust and loyalty to the SA was very strong. In the travel agency sample trust uniquely explained 33% of the variance in loyalty beyond that explained by perceived quality of service and 40% of the variance in loyalty beyond that explained by perceived quality of merchandise. Service was defined as how the travel agent treats the customer (responsive, creative, responsible) while merchandise was defined as what the agent sells to the customer as part of the travel package (e.g., hotels, airfare, car rental). In the hair salons sample trust explained 36% of the variance in loyalty beyond that explained by perceived quality of service and 48% beyond that explained by perceived quality of merchandise. Service was defined as the services rendered by the stylist (e.g., hair cut, perm) and merchandise was defined as shelf product sold to customers (e.g., shampoo, conditioners). In both samples in a multiple regression analysis with all potential predictors of loyalty (trust, RS Levels 1, 2, 3, perceived quality) trust was the only significant predictor of loyalty. In both companies the moderation effect was significant. The relationship between trust and loyalty was stronger when perceived quality was low. In the hair salon sample, however, only perceived quality of service was found to be a moderator. The three evident moderation effects, however, are in the opposite direction than that hypothesized and highlight the importance of trust as a mechanism that decreases discomfort caused by perceived risk. This illustrates the core definition of trust as the willingness of one party to be vulnerable to the actions of another party (Bitner, 1995; Gwinner et al., 1998; Mayer et al., 1995) and to replace uncertainty (Sheth & Parvatiyar, 1995). Berry (1995) proposed that consequences of RS levels may be different for the SA and the store. Thus, customers can experience loyalty to the SA or to the store or to both. Because customers desire and receive incentives from the SA in addition to the pure service acquisition (Beatty at al., 1996; Bitner, 1995), feelings of delight with the service and trust in the SA may be transferred to the store leading to loyalty to the store. Hypothesis 3a stated that the relationship between trust in the SA and loyalty to the store is moderated by perceived quality of service and merchandise. Hypothesis 3a was not supported as there was no evidence of a moderation effect. Hypothesis 3b was supported. In the travel agency sample trust in the SA was a significant predictor of loyalty to the store. Trust explained 4% of the variance in loyalty to the store beyond that explained by perceived quality of merchandise. When perceived quality of service was in the model trust did not have a significant contribution to the explanation of loyalty. This finding supports Sheth and Parvatiyar (1995) who view trust as a marketing tool in contexts of intangible services that are difficult to evaluate prior to the purchase and to the customer experiencing the product. In the hair salon chain store trust did not predict loyalty to the store. The only predictor of loyalty to the store was perceived quality of service explaining 5% of the variance in loyalty to the store. Berry (1995) theorized that Levels 2 and 3 of RS predict loyalty. Hypothesis 4a stated that the relationship between social incentives and loyalty to the SA is moderated by perceived quality of service and merchandise. The moderation effect was not evident but the main effect was found to be significant therefore supporting hypothesis 4b. In the travel agency sample social incentives explained 3% of the variance in loyalty to the SA beyond that accounted for by perceived quality of merchandise. Perceived quality of merchandise explained 16% of the variance in loyalty to the SA. Social incentives did not account for a significant amount of the variance in loyalty when perceives quality of service was in the model. Perceived quality of service explained 23% of the variance in loyalty to the SA. No moderation effect was evident. Similarly, in the hair salon chain store only the main effect was significant. Social incentives explained 7% of the variance in loyalty to the SA beyond that explained by perceived quality of service and 13% beyond that explained by perceived quality of merchandise. The moderation effect was not found to be significant. Perceived quality of service and social incentives together explained 48% of the variance in loyalty to the SA. Perceived quality of merchandise and social incentives together explained 37% of the variance in loyalty to the SA. This supports Ramsey and Sohi’s (1997) who found that relationship quality had a significant influence on customer anticipation of future interactions with the SA. Although Level 2 of RS (social incentives) cannot overcome a competitive core product (Crosby & Stephens, 1987), findings of this study show that Level 2 of RS can drive customer loyalty when competitive differences are not strong. Hypothesis 5a stated that the relationship between structural solutions and loyalty to the SA is moderated by perceived quality of service and merchandise. This hypothesis was supported in the travel agency sample. In the hair salons sample only the main effect was found to be significant. In the travel agency sample structural solutions accounted for a significant amount of the variance in loyalty to the SA beyond that accounted for by perceived quality of service (2%) and merchandise (5%). The significant moderation effect showed that the relationship between structural solutions and loyalty was stronger when perceived quality of merchandise was high. In the hair salon chain store structural solutions explained 4% of the variance in loyalty beyond that explained by perceived quality of service and 9% of the variance beyond that explained by perceived quality of merchandise. The relationship, however, was not moderated. This supports previous research (Berry, 1995: Lovelock, 1994; Miller, 1994) that showed that when offering added value solutions that are designed into the system rather than depend on relationship building skills of individual SAs, the solution binds the customer to the company. This bind is in addition to the bind with an individual SA who may leave the store in the future. Thus, in contrast to the small effect of Level 2 of RS on loyalty, Level 3 of RS, structural solutions, had a larger effect and was found to be a strong predictor of loyalty to the SA. A regression analysis of perceived quality on loyalty to the SA showed that perceived quality of service was a greater predictor of SA loyalty (Beta=.551, t=5.8, p<.01) than was perceived quality of merchandise (Beta=.198, t=2.1, p=.043). While examining perceived quality of service and merchandise as predictors of loyalty to the SA, the zero order correlation of perceived quality of service and merchandise with loyalty were compared to the partial correlation of these variables with loyalty (Appendix H). Large differences between the zero order correlation and the partial correlation in addition to a strong correlation between perceived quality of service and perceived quality of merchandise (.70; .53, respectively) may be evidence of multicollinearity between perceived quality of merchandise and perceived quality of service. This may reflect a lack of differentiation between the two in the consumer’s perspective. Thus, consumers may expect superior service to be fundamental and unquestionable across their personal service providers and to include a high quality selection of merchandise. Hypothesis six examined the effect of structural solutions (RS Level 3) on loyalty to the store. Although positive feelings towards SAs often transfer to the store (Beatty, 1996), and although the store assumes and determines the extent of structural solutions that is needed to delight customers, the null hypothesis was not rejected. Neither hypothesis 6a nor hypothesis 6b was found to be significant. In both samples the only predictor of loyalty to the store was perceived quality of service which contributed 5% to the explanation of variance of loyalty to the travel agency and 8.4% to the variance of loyalty to the hair salon chain store. These findings support the model presented by Ramsey and Sohi (1997) which found perceived quality to be the strongest predictor of loyalty to the store. Findings of hypotheses five, and six indicate that the consumer primary loyalty that is predicted by this model is loyalty to the SA. This finding supports Oliver’s (1997) and Reynolds and Beatty’s (1999) argument that loyalty to the SA is more substantial than other forms of loyalty such as loyalty to the store or to a brand. Findings of this study may be explained by trust as being the underlying foundation of loyalty to the SA. Trust as an interpersonal phenomenon leads to attachment and commitment (Morgan & Hunt, 1994) and may, therefore, be more deeply exhibited in human relationships than in consumer-store relations. If the SA, however, was to leave the company it is likely that consumers would follow the SA as long as the perception of service and merchandise quality is similar. Thus, loyalty to the store and loyalty to the SA seem to be distinct constructs. Loyalty to the SA, however, was also found to be an important predictor of loyalty to the store. In the travel agency loyalty to the SA explained 9% of the variance in loyalty to the store. In the hair salon chain store loyalty to the SA explained 5% of the variance in loyalty to the store. Final Outcomes Hypothesis 7a stated that spending commitment mediates the relationship between loyalty and sales and that if no moderation is evident spending commitment is related to sales. This hypothesis was not supported as the mediation effect was not found to be significant. Hypothesis 7b stated that the main effect is significant and spending commitment is related to sales. Spending commitment is defined as the percentage of service provided by a single provider, in this case, the percentage of travel needs that the customer buys from a particular travel agency or the percentage of professional hair care needs that a customer buys from a particular hair salon chain store. In the travel agency loyalty to the SA explained 5% of the variance in spending commitment. Loyalty was not related to sales neither when spending commitment was not in the model nor when spending commitment was in the model. Loyalty to the store also explained 5% in spending commitment. Loyalty to the store was not related to sales, neither when spending commitment was missing from the model nor when spending commitment was in the model. Finally, spending commitment did explain 4% of the variance in SA sales. This supports Passingham’s (1998) findings concerning the importance of spending commitment as the next target in consumer marketing. Thus, spending commitment was found to be the strongest predictor of behavioral loyalty. Loyalty to the SA and loyalty to the store together explained 9% of the variance in spending commitment. Thus, focusing on spending commitment as the outcome is superior to the focus on loyalty cards or on other current measures of loyalty in enhancing business advantage. Thus, spending commitment is an essential variable that maintains revenue in the system (Ashby, 1964). Without revenue the system will die (Rubinstein & Firstenberg, 1995). From a Cybernetics perspective, the system must regulate itself. The system must control its interaction with its environment by understanding how the system may operate to assure stability of loyalty leading to spending commitment and sales. To assure stability of the spending commitment the system should use an error-controlled regulation. Thus, feedback should be continuously received from customers and SAs. The set of responses that the regulator uses in response to these inputs should be greater than the number of inputs that flows into the system (Law of Requisite Variety, Ashby, 1964). Thus, management of customer relationships in a system that has error-control regulation may include a broad range of changes. Understanding this concept of feedback is the building block of systems thinking (Richmond, 1991). Feedback can overturn deeply engrained ideas such as causality and facilitate learning to recognize structures that recur again and again. This simplifies complexity and enables to see deeper patterns behind events (Senge, 1994). In the hair salon sample the seventh hypothesis linking loyalty to sales was not supported. In explaining these results this author inferred that the small sample size and little variance ($13K) across stylists on sales performance due to a cost leadership position in the market, may have resulted in an underestimate of the correlation and weakened the statistical power to reject the null hypothesis. Limitations and Directions for Future Research Although this author shed some light on several important issues, there are some limitations associated with this study. First, customers in the travel agency sample may constitute a high end of the travel market. Thus, the construct in this study may play out differently in other retail settings and with other study populations. Further more, in this study this author discovered that the type of retail (travel, professional hair care) has an effect on relationships between customer incentives and outcomes. Therefore, results of this study may not be generalizable across other service retail domains. (Reynolds & Beatty, 1999). Moreover, multiple regression analyses using same data may result in common method variance inflating Type I error (Tabachnick & Fidell, 1996). Thus, in order to increase our understanding of this phenomenon, future studies should focus on examining other service settings in which customer-SA relationships are likely to exist. Second, the design of this study is cross- sectional. Although the study model is logical and consistent with previous work, this author cannot declare cause and effect relationships between variables studied. A longitudinal study would increase the confidence in the causal dynamic. Also, a longitudinal study would enable an examination of which RS level is more important across stages of the relationship development. Future studies should, therefore, adopt a longitudinal research design and study how relationships and trust in relationships evolve and change over time. A longitudinal study would also enable to test whether customers with low levels of trust can become relational customers with the use of relational bonds. In this study the variance in trust was very small. It would be interesting to test differences in trust between long-term relationships and short term relationships. Trust is a precondition to move across developmental stages of a relationship. Trust may be very high in long term relationships due to consumer rationalization processes. It may also be possible, however, that trust is more important in early stages of the relationship when the perception of quality is still in formation and trust functions as a mechanism for reducing perceived risk. Third, the model of this study regards trust as both a dependent and an independent variable. Due to the vital importance of trust, future studies should focus on trust as a moderator and test its effect on the relationship between RS levels and loyalty to the store. It is possible that the relationship will exist only when the degree of trust is high. Fourth, the size of the sample in each company was too small to test the whole model simultaneously using a structural equation modeling technique. The size of the sample was also too small to segment relationship customers by the RS level they prefer most and test the hypothesized relationships by segment. Future studies should focus on testing which segment of customers, those in RS levels 1, 2, or 3 are more trustful, more loyal, and more profitable. Fifth, the incentives scale was developed for the purpose of this study. Future use of this scale will assist in further validating and in improving its psychometric properties. Also, a further development of the study model is needed in order to increase the explained variance. Other antecedents such as communication, shared values, may be tested for their effect on trust, store loyalty, and sales. Future studies should test how these variables relate to RS levels and how they interact in these and other settings. In terms of under-searched areas, there is very little known about how customers characterize the strength of a relationship (Bitner, 1995; Leuthesser, 1997; Reynolds & Beatty, 1999). Future studies should focus on the question what determines closeness in service relationships. There is also very little known about why relationship customers decide to terminate a relationship (Berry, 1995; Sheth & Parvatiyar, 1995). Proposed reasons for termination in the literature are dissatisfaction, conflict, boredom, costs, or superior alternatives (Sheth & Parvatiyar, 1995) which carry differential implications for management. Lastly, since the primary loyalty is to the SA, the use of relationship management to motivate SAs and enhance their commitment is called for. In many retail settings SA turnover is high. The affect of high turnover on both the development and the maintenance of customer relationship should be tested. Future studies should focus on how customer relationships affect SA commitment and turnover. Managerial Implications This research contributes to marketing practice by addressing four issues. First, it aids in understanding what values and work norms should be developed as part of the organizational culture. Second, it outlines which RS levels should be applied in forming and maintaining trust. Third, it identified the key mechanism through which sales should be enhanced. Finally, it highlights important systemic facilitators of RS in retail companies that choose to apply the RS strategy. Organizational Perspective. Trust was found to be an important antecedent of both loyalty to the SA and loyalty to the store. The primacy of this value and of customer orientation in achieving loyalty is strongly indicated by this study. Trust is, therefore, a critical company asset affecting competence. As an asset, trust should be built on a store-consumer level as well as on a company-SA level. Work norms that enhance trust should be both applied and controlled by the organizational culture. As suggested by Deal and Kennedy (1982) cultural control implies that strong shared values and work norms serve as important sources in guiding and controlling SAs behaviors. A trust environment not only assists in forming and maintaining relationships and loyalty but also, enhances job satisfaction and overall performance of sales people (Rich, 1997). Maintaining a good rapport with salespeople, stressing equity, fostering two way communication, collaborating, encouraging creativity and initiative, and approving appropriate risk taking, are a few suggested ways to create trust in the workplace (Rich, 1997). Guaranteeing superior service to customers and employees and assuring a cross-organizational effort to facilitate relationships, will assist in building an environment of trust and in making trust a core value of the organizational culture Marketing Perspective. A focus on profitability calls to apply multiple strategies for each market segment by identifying the spending commitment of that segment. One strategy should target customers who are moderate to heavy spenders. A different strategy should be applied with light spenders and a third may be applied with transactional customers. Some customers may be deal- prone. These customers are receptive to better offers from competitors or seeking out for such. Since relationship selling involves fixed and variable costs across stages of the relationship initiation and development, managers should screen customer ability to be effective relationship partners. Managers should identify customers who are most likely to be loyal and develop the overall relationship selling strategy around delivering superior value to these customers. Managing trustful relationships with these customers is to be the core mission of the sales job, enhancing spending commitment and sales through loyalty to SAs, loyalty to the store, and trust. Managers should design an applicable incentives program for each RS level and categorize customers by their RS level preference. Highly social customers may want a completely different package of incentives than that wanted by customers who desire specific solutions. For successful maintenance of customer relationships it is recommended that managers should strive to clearly understand how customers are segmented by RS level preference and exceed customer expectation. The moderation effect of consumer desire in structural solutions showed that the relationship between RS level 3 and trust was stronger when consumer desire in structural solutions was low. This indicates that companies may have to educate consumers about the potential added-value of structural solutions. Although it is simpler to measure and manage pricing incentives, retailers interested in extending business opportunities in existing markets, should target their promotion and relationship management efforts at structural solutions and social incentives (RS Levels 2 and 3) in addition to pricing incentives (RS Level 1). Since pricing is the most imitated element of the marketing mix (Stephenson & Fox, 1987), the potential for sustainable advantage using Level 1 of RS is low. Also, customers that are interested in pricing incentives such as customers in the hair salon chain store are vulnerable to competitor promotion and thus, are less profitable than those interested in higher levels of RS. Since RS level preferences are dynamic, managers should use feedback loops and open communication with customers while giving different emphases to different RS levels across stages of the relationship. Findings of this study show that the primary loyalty of the customer is to the SA rather than to the store. To reduce customer defection, loss of expenditures in relationship building, and loss of future revenues when SAs leave the company, managers are to create service teams that are responsible for a customer rather than a setting of single service providers. Thus, customers will be served by their personal SA but when the latter leaves they would be able to turn to other SAs that they knew in the process. Also, data regarding relationship management and customer files should be recorded so that other SAs may use them to learn the customer profile and maintain an existing relationship rather than establish a new relationship from scratch. Management Perspective. Successful management of relationships also requires SA retention. Employee retention is viewed as an antecedent to customer retention. High turnover creates a cycle of failure (Schlesinger & Heskett, 1991). High turnover rates discourage management from investing in hiring, training, and other commitment building activities. This, in turn leads to ineffective performance and /or the perception of dull or dead-end work, which feeds employee turnover. High employee turnover negatively affects service quality and customer retention, hurting profitability, and reducing resources available to invest in employees’ success. This leads to an inability to form new relationships and to difficulties in maintaining existing relationships. Management is to identify salespeople that can independently form and manage customer relationships. Using adequate selection processes should be a first step to SA retention and customer satisfaction. Job enrichment may also result in higher SA retention. Providing SAs with greater autonomy in relationship management and with an exceptional latitude of decision making may enhance performance (Fried, 1993). Lack of sufficient empowerment is one among a few situational impediments of systems thinking (Richmond, 1991; Senge, 1994). Linking SA performance and incentives to customer satisfaction may greatly enhance SA commitment leading to reduced turnover as well as to increased organizational citizenship behaviors (Netemeyer, Boles, McKee, & McMurrian, 1997). Validation of processes of selection, training, performance appraisal, is to be performed on a continuos basis. These behaviors are believed to directly promote the effectiveness of the organization and to influence salespeople’s objective sales productivity (Podsakoff & Mackenzie, 1994) and customer service (George, 1991). An antecedent of organizational citizenship behaviors is trust (Organ, 1988). SAs are to be trained to recognize desires of consumers from service relationships, to adjust their sales style accordingly so that, the level of agreement between consumer desired incentives and applied incentives will rise. Also, a better understanding of consumer desires on an individual level will enhance the agreement on applied incentives between parties to the dyad. This understanding is fundamental in delivering beyond expectations service. Since customers differ on their expectations and preferences of RS levels, SAs are required to adjust their sales approach in accordance to the customer needs. An examination of the extent of agreement between parties to the dyad in the travel agency sample was done by examining the ability of travel agents to adjust their sales style according to the customer desires. The analysis showed that adjustment scale mean was 3.54 on a scale of 1 to 7. A few items that require further effort of adjustment are: experiment with different sales approaches, flexibility in the sales approach the SA uses, an easy, wide use of various sales approaches, modification of sales presentation when the situation calls for it, and varying the sales style across situations. These items ranged from 2.4 to 3.6 on a scale of 1-7. In the hair salons sample adjustment scores were higher. Scale mean was 4.1 with four items requiring improvement: The ability to change an approach easily, the ability to experiment with various approaches, the ability to use of a wide variety of approaches, and the ability to modify the sales presentation when the situation calls for it. Training programs will assist SAs in opening themselves to deeper mental models and behavior patterns. This would lead to a better understanding of forces that affect their decision making concerning customer relationships (Linstone, 1984; Richmond, 1991; Senge, 1994). Examining deep mental models SAs will be able to also understand what are their difficulties in sales style adjustment. Training should also assist in designing a shared vision that inspires SAs, compels them to be role models in the process of building commitment to relationships, and connects them to the organization. Conclusions: A Systems Perspective In order to be a learning organization (Richmond, 1991; Senge, 1994) all managerial levels are to let people continually expand their capacity to create results they truly desire. Managers should be committed to nurture new patterns of thinking and to set aspirations free so that people learn how to learn together. A learning organization can learn faster from its competitors, encouraging all levels within the organization to learn to engage in learning and to turn everyday practice into organizational knowledge bases. Trust as well as the formation, maintenance, and management of relationships rely on tacit knowledge that the organization develops overtime and uses as a commercial asset in the marketplace. Tacit knowledge was found to be the most difficult strategy to be duplicated by competitors (Taylor, 1999). Relationship selling is a successful strategy that depends on a high level of trust as a precondition that enables companies to extract business in existing markets, to enhances loyalty, market share, business share, image, and sales. This study is an interdisciplinary study with holistic thinking as its main concept. The inquiry of the topic of relationship selling should take a Sinergian approach (Mitroff & Turoff , 1973) that combines values, ethics, psychology, with the problem that is being put forth: a marketing strategy. The analysis of the topic of relationship selling using a Sinergian inquiry system, is an evolving iterative process. This process targets both future and present objectives of a company without attributing priority to one aspect over another. This study shows that all RS levels, trust, and loyalty are inter-dependent leading to the maintaining of dynamic systems such as retail companies. REFERENCES Aaker, D. Kumar, V. & Day, G. (1995) Marketing research, 5th ed. John Wiley and Sons Inc. ch. 13, 14 Aijo, T. S. (1996). The theoretical and philosophical underpinning of relationship marketing: Environmental factors behind the changing marketing paradigm. European Journal of Marketing, 30, 8-19. Albrecht, K. (1992). The only thing that matters. Bringing the power of the customer into the center of your business. New York, HarperBusiiness. Anderson, J.C., & Narus, J. A. (1990). A model of distributor firm and manufacturer firm working relationship. Journal of Marketing, 54, 42-48. Anderson, E. & Wietz B. (1989). Determinants of continuity in conventional industrial channel dyads. Marketing Science, 8, 310-323. Anderson, F. S. (1992). Customers must be pleased not just satisfied. News Week, 3, 52 Ashby, W. R. (1964). An introduction to Cybernetics. Ch 1, 10-12. London: Mathuen & Company. Axelrod, R. (1984). The evolution of cooperation. New York: Basic Books. Baber, F. (1997). Fulfilling needs. Progressive Grocer, 12, 18-21. Barber, B. (1983). The logic and limits of trust. New Brunswick, NJ: Rutgers University Press. Barnes, J. G. (1994). The issues of establishing relationships with customers in service companies: When are relationships feasible and what form should they take? Paper presented at frontiers in service conference. American Marketing Association and Vanderbilt University Center for Service Marketing. Barlow, R. G. (1992). Relationship marketing-the ultimate customer services. Retail Control, 60, 29-37. Barrons, R. M. & Kenny, D. A. (1986). The moderator- mediator variable distinction in social psychological research: The conceptual, strategic, and statistical consideration. Journal of Personality and Social Psychology, 51, 1173-1182. Barney, J. B. & Hansen, M. H. (1994). Trustworthiness as a source of competitive advantage. Strategic Management Journal, 15, 175-190. Bauer, T. N. & Green, S. G. (1996). Development of leader- member exchange (LMX): A longitudinal test. Academy of Management Journal, 39, 1538-1567. Berg, J., Dickhout, J., & McCabe, K. (1995). Trust, reciprocity, and social history. Unpublished working paper, University off Minnesota, Minneapolis. Beatty, S. E., Mayer, M. Coleman, J. E. Reynolds, K. E. & Lee, J. (1996). Customer sales associate retail relationship. Journal of Retailing, 72, 223-247. Bendapudi, N. & Berry, L. L. ( 1997). Customers motivations for maintaining relationships with service providers. Journal of Retailing, 73, 15-37. Berry, L. L. (1983). Relationship marketing. Emerging perspectives on service marketing. Eds. L. L. Berry, G. L. Upah. Chicago: American Marketing Association, 25-38. Berry, L. L. (1995). Relationship marketing of services: Growing interests, emerging perspectives. Journal of the Academy of Marketing Science, 23, 236-246. Berry, L. L. & Parasuraman, A. (1991). Building a new academic field: The case of services marketing. Journal of Retailing, 69, 13-60. Berry, L. L., Sieder, K., & Gresham, L. G. (1997). For love and money: the common traits of successful retailers. Organizational Dynamics, 7 ,23. Bhattacharya, R., Devinney, T. M., & Pillutla, M. M. (1998). A formal model of trust based on outcomes. Academy of Management Journal 23, 459-472. Bies, R. J. & Tripp, T. M. (1996). Beyond distrust: “getting even and the need for revenge”. In R. M. Kramer & T. R. Tyler (Eds.) Trust in organizations: Frontiers of theory and research. 246-260. Thousand Oaks, CA: Sage. Bigley, G. A. & Pearce, J. L. (1998). Straining for shared meaning in organizational science: problems of trust and distrust. Academy of Management Review, 23, 405-422. Bitner, J. M. (1996). Building service relationships: It’s all about promises. Journal of the Academy of Marketing Science, 23, 246-251. Bitter, C. (1998). Lazy thinking will not be rewarded. Progressive Grocer, 6, 88. Bjorkman, B. (1997). What’s ahead for grocery retailing. Oregon Food Journal,17, 25-32. Blau, P. M. (1964). Exchange and power in social life. New York, Wiley. Bolton, N. R. (1998). A dynamic model of the duration of the customer’s relationship with a continuous service provider: the role of satisfaction. Marketing Science, 17, 1, 45-65. Bonoma, T. V. & Clark, B. H. (1988). Marketing performance assessment. Boston, Harvard Business School Press. Borch, F. (1957). The marketing philosophy as a way of business life: the marketing concept, its meaning to management. Marketing Series, 99, 3-5. Bower, A. G., Garber, S., & Watson, J. C. (1996). Learning about a population of agents and the evolution of trust and cooperation. International Journal of Industrial Organization, 15, 165-190. Boyers, K. (1998). Travel Industry outlook remains strong. Association Management, 50, 64-65. Bradack, J. & Eccles, R. (1989). Price, authority, and trust: from ideal types to plural forms. Annual Review of Sociology, 15, 97-118. Brockner, J., Siegel, P. A., Tyler, T., & Martin, C., (1997). When trust matters: the moderating effect of outcome favorability. Administrative Science Quarterly, 42, 558-583. Bromiley, P. & Cummings, L. L. (1995). Transaction costs in organizations with trust. Research on Negotiation in Organizations 5, 219-247. Burt, R. S. & Knez, M. (1995). Kinds of third party effects on trust. Rationality and Society, 7, 255-292. Burt, R. S. & Knez, M. (1996). Trust and thirdd party gossip. In R. M. Kramer & T. R. Tyler (Eds.), Trust in organizations: Frontiers of theory and research. 68-89. Thousand Oaks, CA: Sage. Butler, J. K. Jr. (1983). Reciprocity of trust between professionals and their secretaries. Psychological Reports, 53, 411-416. Butler, J. K. Jr. (1991). Toward understanding and measuring conditions of trust: Evolution of a condition of trust inventory. Journal of Management, 17. Butler, J. K. Jr. & Cantrell, R. S. (1984). A behavioral decision theory approach to modeling dyadic trust in superiors and subordinates. Psychological Reports, 55, 19- 28. Casti, J. (1994). Complexification: Explaining a paradoxical world through the science of surprise. New York, HarperCollins Publishers. Calton, J. M. & Lad, L. J. (1995). Social contracting as a trust building process of network governance. Business Ethics Quarterly, 5, 271-295. Chiles, T. H. & McMacklin, J. F. (1996). Integrating variable risk preferences, trust, and transaction cost economics. Academy of Management Review, 21, 73-99. Chipkin, H. (1997). An upstick in the upfront: Travel. Brandweek, 38, S28-S33. Churchill, G. A. Jr., Ford, N. M. Steven, H. W. & Orville, C. W. Jr. (1985). The determinants of salespeople performance: A Meta analysis. Journal of Marketing Research, 22, 103-118. Churchill, G. A. (1979). A paradigm for developing better measures of marketing constructs. Journal of Marketing Research, 16, 64-73. Clarck, M. S. & Millis, J. (1979). Interpersonal attraction in exchange and communal relationships. Journal of Personality and Social Psychology, 37, 1, 12-24. Clarck, M. & Payne, A. (1994). Achieving long term customer loyalty: A strategic approach. Working paper. Center for services management, Cranfeld School of Management. Coleman, J. S. (1990). Foundations of social theory. Cambridge, MA: Belknap Press of Harvard University. Collins, G. (1995). Ben & Jerry’s talent hunting. The New York Times, 2. Cook, J. & Wall, T. (1980). New York attitude measures of trust, organizational commitment, and personal need non- fulfillment. Journal of Occupational Psychology, 53, 39- 52. Corstjens, J. & Corstjens, M. (1995) Store Wars: The battle for Mindspace and Shelfspace. John Wiley and Sons. Creed, W. E. D. & Miles, R. E. (1996). Trust in organizations: a conceptual framework linking organizational forms managerial philosophies, and the opportunity costs of controls. In R. M. Kramer & T. R. Tyler (Eds.) , Trust in organizations : Frontiers of theory and research. 302-330. Thousand Oaks, CA : Sage. Crosby, L. Evans, K. R. & Cowles, D. (1990). Relationship quality in services selling: An interpersonal influence perspective. Journal of Marketing, 54, 68-81. Cross, R. & Smith, J. (1995). Customer bonding. Chicago, IL: NTC Business Books. Crosby, L. A. & Stephens, N. (1987) Effects of relationship marketing on satisfaction, retention, and prices of life insurance. Journal of Marketing Research, 24, 404-411. Cummings, L. L. & Bromiley, P. (1996) The organizational trust inventory (OTI): development and validation. In R. M. Kramer & T. R. Tyler (Eds.), Trust in organizations: Frontiers of theory and research, 302-330, Thousand Oaks, CA: Sage. Curral, S. C. & Judge, T. A. (1995). Measuring trust between organizational boundary role persons. Organizational Behavior and Human Decision Processes, 64, 151-170. Czepiel, J. A. (1990). Service encounters and service relationships, implications for research. Journal of Business Research, 20, 13-21. Darden W. R. & Dorsch M. J. (1990). An action strategy approach to examining shopping behavior. Journal of Business Research, 21, 289-308. Day, S. G. (1990). Market driven strategy. processes of creating value, The Free Press. Dasgupta, P. (1988). Trust as a commodity. In D. G. Gambetta (Ed.), Trust. 49-72. New York, Basil Blackwell. Das, T. K. & Teng, B. S. (1998). Between trust and control: developing confidence in partner cooperation in alliances. The Academy of Management Review, 23, 491-512 Deutsch, M. (1958). Trust and suspicion. Journal of Conflict Resolution, 2, 265-279. Deutsch, M. (1960). The effect of motivational orientation upon trust and suspicion. Human Relations, 13, 123-139. Deutsch, M. & Krauss, R. M. (1962). Studies of interpersonal bargaining. Journal of Conflict Resolution, 6, 52-75. Dienesch, R. M. & Liden, R. C. (1986) Leader member exchange model (LMX) of leadership: A critique and further development. Academy of Management Review, V 11, 618-634. Dodgson, M. (1993). Learning, trust, and technological collaboration. Human Relations, 46, 77-95. Doggett, L. R. (1998). Tourism’s role in a changing economy. Business America, 119, 15. Donney, P. M., Cannon, J. P., & Miellen, M. R. (1998). Understanding the influence of national culture on the development of trust. Academy of Management Review, 23, 601- 621. Dodds W. B. (1991). In search of value: How price and store name information influence buyers product perceptions. Journal of Service Management, 5, 3, 27-36. Dodds, W. B. & Monroe, K. B. (1985). The effects of brand and price information In E. C. Hirschman & M. B. Holbrook (Eds.) Advances in consumer research, Provo, UT The Association for consumer reseach. Dorgan, T. (1997). Belief in the future. Progressive Grocer, 76, 67-68. Driscoll, J. W. (1978). Trust and participation in organizational decision making as predictors of satisfaction. Academy of Management Journal, 21, 44-56. Dubinski, A. J., Roy, D. H., Thomas, N. I., & Bellenger, D. (1986). Sales force socialization. Journal of Marketing, 50, 192-207. Dwyer, F. R., Schurr, P. H., & Oh, S. (1987). Developing buyer seller relationships. Journal of Marketing, 51, 11-27. Earley, P. C. (1986). Trust, perceived importance of praise and criticism, and work performance: an examination of feedback in the United States and England. Journal of Management, 12, 457-473. Eisenberger, R., Fasolo, P., & Davis La Mastro, V. (1990). Perceived organizational support and employee diligence, commitment, and innovation. Journal of Applied Psychology, 75, 51-59. Elangoven, A. R. & Shapiro, D. L. (1998). Betrayal and trust in organizations. Academy of Management Review, 23, 547-567. Elliot, G. R. (1990). The marketing concept-necessary but sufficient? An environmental view. European Journal of Marketing, 24, 20-30. Etzioni, A. (1968). The active society. New York Press. Evans, J. R. & Laskin, R. L. (1994). The relationship marketing process: a conceptualization and application. Industrial Marketing Management, 23, 439-452. Evans, K., Fraser, P., & Walkilate, S. (1996). Whom can you trust. The politics of grassing on an inner city housing estate. The Sociological Review, 44, 361-380. Egolf, K. (1997). Keeping the focus on a one on one marketing. Advertising Age Business Marketing, 82, 16. Eldridge, D. (1993). Non store retail planning for a big future, Chain Store Age Executives, 34-35 Endicott, R. C. (1997). Non traditional segments spur 14% billings growth. Advertising Age Business Marketing, 68, 16. Fein, S. & Hilton, J. L. (1994). Judging others in the shadow of suspicion. Motivation and Emotion, 118, 167-198. Feldman, J. M. (1998). Pricing and cyber sales. Air Transport World, 35, 64-66. Fellman, M. W. (1998). Travel industry aims for bull’s-eye with niche marketing. Marketing News, 32, 1, 15. Foa, U.G., Converse, J. Jr., Turnblum, K. Y., & Foa, E. B. (1993). In J., Berg, K., Piner, & S. Frank (Eds.), Resource theory exploration and applications. Academic Press, Inc. Harcourt Brace Javanovich Publishers. 13-30, 169-195.. Fine, G. A & Holyfield, L. (1996). Secrecy, trust, and dangerous leisure: generating group cohesion in voluntary organizations. Social Psychology Quarterly, 59, 22-38. Fox, A. (1974). Beyond contract: Work, power and trust relations. London, Faber and Faber. Frey, B. S. (1993). Does monitoring increase work effort? The rivalry with trust and loyalty. Economic Inquiry, 31, 663-670. Freietag, H. (1997). Relationship marketing works only when practiced. Hotel and Motel Management, 212, 18. Fukuyama, F. (1995). Trust: The social virtues and the creation of prosperity. New York: The Free Press. Gabarro, J. (1990). The development of working relationships. In J. Galagher, R. E. Kraut, & C. Edigo (Eds.), Intellectual teamwork: Social and technological foundations of cooperative work. 79-110. Hillsdale, NJ: Erlbaum. Gabarro, J. (1978). The development of trust, influence, and expectations. In A. G. Athos and J. J. Gabarro (Eds.), Interpersonal behavior: Communication and understanding in relationships. 290-303. Galper, J. (1998). Population Update. American Demographics, 20, 32. Gambetta, D. (1985). Can we trust. In D. Gambetta (Eds.), Trust making and breaking cooperative relations. 213-238, New York, Basil Blackwell. Gengler, C. E., Leszczyc, P., Popkowski, T. L. (1997). Using customer satisfaction research for relationship marketing: a direct marketing approach. Journal of Direct Marketing, 11, 1, 23-29. George, J. M. (1991) State or trait: Effects of positive mood on prosocial behaviorsat work. Journal of Applied Psychology, 76, 2, 299-307. Gillpatrick, T. R., Harmon, R. R., & Tseng, D. L. P. (1994). The effect of a nominal monetary gift and different contacting approaches on mail survey response among engineers. IEEE Transactions on Engineering Management, 41, 285. Goffman, E. (1971). Relations In Public. New York. Basic Books. Goldman, S. L., Nagel, R. N., & Preiss, K. (1994). Agile Competitors and Virtual Organizations: Strategies for Enriching the Customer. Library of Congress Cataloging in Publication data. Govier, T. (1994). Is it a jungle out there? Trust, distrust, and the construction of social reality. Dialogue, XXXIII, 237-252. Good, D. (1988). Individuals, interpersonal relations, and trust. In D. Gambetta (Eds.), Trust: Making and breaking cooperative relations: 31-48, New York, Basil Blackwell. Goodman, P. S., & Fichman, M. F., Lerch, J. F., & Snyder, P. (1995). Customer firm relationships, involvement, and customer satisfaction. Academy of Management Journal, 38, 1310-1324. Goodwin, C. & Gremler, D. D. (1996). Friendship over the counter: how social aspects of service encounters influence consumer service loyalty. In T. Swartz, D. Bowen, & S. Brown (Eds.), Advances in service marketing and management. Graen, G. B. & Uhl-Bien, M. (1995). Development of leader member exchange theory of leadership over 25 years: Applying a multi-level, multi-domain perspective. Leadership Quarterly, 6, 219-247. Graen, G. B. & Scandura, T. A. (1987). Toward a psychology of dyadic organizations In L. L. Cummings & B. M. Staw (Eds). Research in Organizational Behavior, 9, 175-208, Greenwich, CT: JAI press. Granovetter, M. (1985). Economic action and the social structure: the problem of embeddedness. American Journal of Sociology, 91, 481-510. Gronoos, C. (1994). From marketing mix to relationship marketing: Toward a paradigm shift in marketing. Management Decision, 32, 4-21. Gronoos, C. (1996). Relationship marketing strategic and tactical implications. Management Decisions, 34, 5. Gummesson, E. (1996). Relationship marketing and imaginary organizations: A synthesis. European Journal of Marketing, 30, 31-45. Gummesson, E. (1997). Relationship marketing: The emperor’s new clothes or a paradigm shift? Marketing & Research Today, 25, 53-60. Gulati, R. (1995). Does familiarity breed trust? The implications of repeated ties for contractual choice in alliances. Academy of Management Journal, 38, 85-112. Gurtman, M. B. (1992).Trust, distrust, and interpersonal problems: a circumplex analysis. Journal of Personality and Social Psychology, 62, 989-1102. Guth, W., Ockenfeld, P., & Wendel, M. (1997). Cooperation based on trust: an experimental investigation. Journal of Economic Psychology, 18, 15-43. Gutman, J. (1982) A means –end chain model based on consumer categorization processes. Journal of Marketing, 46, 60-72 Gwinner, K. P., Gremeler, D. D., & Bitner, M. J. (1998). Relational benefits in services industries: the customer’s perspective. Journal of the Academy of Marketing Science, 26, 101-114. Hagen, J. M. & Choe, S. (1998). Trust in Japanese interfirm relations. Institutional sanctions matter. Academy of Management Review, 23, 589-601. Haley, R. I. (1968). Benefit segmentation: A decision oriented research tool. Journal of Marketing, 32, 30- 35. Harmon, B. (1997). Personal Communication. Hardwig, J. (1991). The role of trust in knowledge. The Journal of Philosophy, LXXXVIII, 693-708. Hamel, G. & Prahalad, C. K. (1994). Competing for the future. Harvard Business School Press. Harrison, D. A., & McLaughlin, M. E. (1993). Cognitive processes in self-report responses: Tests of item context effect on work attitude measures. Journal of Applied Psychology, 78, 129-140. Han, S. L., Wilson, D. T., & Dant, S. P.(1993). Buyer supplier relationship today. Industrial Marketing Management, 22, 331-338. Hardin, R. (1993). The street level epistemology of trust. Politics and Society, 21, 505-529. Hart, P. & Saunder, C. (1997). Power and trust: critical factors in the adoption and the use of electronic data interchange. Organizational Science, 8, 23-42. Harrison, D. A. & McLaughlin, M. E. (1991). Exploring the cognitive processes underlying reponses to self report instruments. Proceedings of the 1991 Academy of Management Annual Meeting, 310-314. Hass, D. F. & Deseran, F. A. (1981). Trust and symbolic exchange. Social Psychological Quarterly, 44, 3-13. Haslam, N. & Fiske, A. P. (1992)..Implicit relationship prototypes: investigating five theories of the cognitive organization of social relationships. Journal of Experimental Social Psychology,28, 441-474. Herzberg, F. (1966). Work and the nature of man. New York. World Publishers. Hinkin, T. R. (1998) A brief tutorial on the development of measures for use in survey questionnaires. Organizational Research Methods, 1, 104-121. Hoffman, G. (1998). Understanding the first priority. Progressive Grocer, 12 88. Holmes, J. G. (19910. Trust and the appraisal process. In W. H. Jones and D. Perlman (Eds.), Advances in personal relationships, 2, 57-104. London, jessica Kingsley. Homans, G. C. (1961). Social behavior and its elementary forms. New York, Hartcourt Brace and World. Hollon, C. J. & Gemmil, G. R. (1977). Interpersonal trust and personal effectiveness in the work environment. Psychological Reports, 40, 454. Hwamg, P. & Burgers, W. P. (1997). Properties of trust: an analytical view. Organizational Behavior and Human Decision Processes, 69, 67-73. Jarillo, J. C. (1988). On strategic networks. Strategic Management Journal, 9, 31-41. Jackson, D. (1993) The seven deadly sins of financial marketing and the road to redemption. Direct Marketing, 79, 43-48. Jennings, E. E. (1971). Routes to the executive suite. New York, McGraw Hill. Johnson-George C. E., & Swap, W. C. (1982). Measurement of specific interpersonal trust: construction and validation of a scale to assess trust in a specific other. Journal of Personality and Social Psychology, 43, 1306-1317. Johnson, D. W. & Noonan, M. P. (1972). Effects of acceptance and reciprocation of self disclosures on the development of trust. Journal of Counseling Psychology, 9, 411-416 Kahn, E. B. (1998). Dynamic relationships with customers: high variety strategies. Journal of the Academy of Marketing Science, 26, 1, 45-53. Kawelski, M. (1997). Relationship marketing: Turning customers into friends. Agriculture Marketing, 35, 30-32. Keith, R. J. (1960). The marketing revolution. Journal of Marketing, 35-38. Kegan, D. L. & Rubinstien, A. H. (1973). Trust, effectiveness, and organizational development: a field study in R & D. The Journal of Applied Behavioral Science, 9, 498-513. Kerlinger, F. N. (1986). Foundations of Behavioral Research. (3rd Ed) new York. Kipnis, D. (1995). Trust and technology. In R. M. Kramer & T.R. Tyler (Eds.), Trust in Organizations: Frontiers of Theory and Research, 39-50. Thousand Oaks, CA: Sage. Kirchler, E., Fehr, E., & Evans, R. (1996). Social exchanges in the labor market: reciprocity and trust versus egoistic money maximization. Journal of Economic Psychology, 17, 313-341. Koller, M. (1988). Risk as a determinant of trust. Basic and Applied Social Psychology, 9, 265-276.- Kogut, B. Shan, W., & Walker, G. (1993). Ompetitive cooperation in biotechnology: Learning through networks? In N. Nohria & R. Eccles (Eds.) Networks aand organizations: Structure, form, and action. Boston, Harvard Business School Press. Koselka, R. (1992). Distribution revolution. Forbes, 54- 62. Kotler, P. (1997). Marketing management. Analysis, planning, implementation, and control. Prentice Hall. Klomonski, R. J. & Karol, B. L. (1976). The impact of trust on creative problem solving groups. Journal of Applied Psychology, 61, 630-633. Kramer, R. M. (1994). The sinister attribution error: paranoid cognition and collective distrust in organizations. Motivation and Emotion, 18, 199-230. Kramer, R. M. & Isen, A. M. (1994) Trust and distrust: its psychological and social dimensions. Motivation and Emotion, 18, 105-107. Kramer, R. M., Brewer, M. B., & Hanna, B. A. (1994). Collective trust and collective action: the decision to trust as a social decision. In R. M. Kramer, & T.R. Tyler (Eds.), Trust in organizations: Frontiers of theory and research, 357-389, Thousand Oaks, CA; Sage. Krecker, M. L. (1995). “From the instinct of workmanship” to “gift exchange”: employment contracts, social relations of trust, ad the meaning of work. Research in the Sociology of Work, 5, 105- 133. Kreps, D. (19900. Corporate culture and economic theory. In J. Alt and K. Shepsle (Eds.), Perspectives on positive political economy. 90-143. Cambridge, England. Cambridge University Press. Lahno, B. (1995a) Trust and strategic rationality. Rationality and Society, 7, 442-464. Lagace, R. R., Dahlstorm, R., & Gassenheimer, D. B. (1991). The relevance of ethical salesperson behavior on relationship quality: the pharmaceutical industry. Journal of Personal Selling and Sales Management, 11, 39-47. Lane, C. & Buchman, R. (1996). The social construction of trust: supplier relationships in Britain and Germany. Organization Studies, 17, 365-395. Larson, M. (1996). In pursuit of a lasting relationship. Journal of Business Strategy, 6, 31. Larzerele, R. E. & Huston, T. L. (1980). The dyadic trust scale: toward understanding interpersonal trust in close relationships. Marriage and the Family, 42, 595-604. Lendaris, G. G. (1986). On systemness and the problem solver. Tutorial comments. IEEE Transition on Systems, Man, and Cybernetics, July, 603-610. Levine, J. (1993). Relationship marketing. Forbes, 152, 232- 234. Leuthesser, L. (1997). Supplier relational behavior: An empirical assessment. Industrial Marketing Management, 26, 245-254. Lewis, G. (1997). Designing for a more perfect world, 12, 24. Lewis, I., Semeijn, J., & Talalayevski, A. (1998). The impact of information technology on travel agents. Transportation Journal,37, 20-25. Lewis, D. & Weigert, A. (1985). Trust as a social reality. Social Forces, 63, 967-985. Lewicki, R. J. & Bunker, B. B. (1995). Trust in relationships. A model of development and decline. In Bunker, B. B., Rubin, J. Z., & Associates: Conflict cooperation and justice. Jossey Bass, 133-173. Lewicki, J. R., McAllister, J. D., & Bies, J. R. (1998). Trust and distrust: new relationships and realities. The Academy of Management Review, 23, 438-459. Liden, R. C., Sparrow, R. T., & Wayne, J. S. (1997) Leader member exchange theory: The past and potential for the future. Research in Personnel and Human Resource Management, 15. Linstone, H. A. (1984). Multiple perspectives for decision making. Elsevier, N.Y. Longabough, R. (1963). Sociometry, 26, 319. Loomis, J. L. (1959). Communication and the development of trust. Human Relations, 12, 304-315. Lorenz, E. H. (1992). Trust and the flexible firm: international comparisons. Industrial Relations, 32, 455- 472. Lorenz, E. H. (1993). Flexible production systems and the construction of trust. Politics and Society, 21, 307-324. Lovelock, C. H. (1983). Classifying services to gain strategic marketing insights. Journal of Marketing, 47, 9- 20. Macaulay, S. (1963). Non contractual relationships in business. A preliminary study. American Sociological Review, 28, 55-67. MacCrimmon, K. R. & Wehrung, D. A. (1986). Taking risks: The management of uncertainty. New York, Free Press. Madhok, A. (1995). Revisiting multinational firms’ tolerance for joint ventures: a trust based approach. Journal of International Business Studies, 26, 117-137. Marshall, G. W. (1996). The folly of rewarding for A while waiting for B. Marketing News, 30, 23, 6-9. Magazine, S. L. & Williams, L. J. (1996). A confirmatory factor analysis examination of reverse coding effects on Meyer & Allen’s affective and commitment scales. Journal of Educational and Psychological Measurement, 56, 241-250. Maslow, A. (1954). Motivation and personality. Harper and Row. 80-106. Mathews, R. (1997). Welcome to the logistics Museum. Progressive grocer, 12, 6-12. center Mathews, R. (1998). The merchant of the future. Progressive Grocer, 4, 6-8. Matthews, B. A., Kordonski, W. M., Shimoff, E. (1983). Temptation and maintenance of trust: Effects of bilateral punishment capability. Journal of Conflict Resolution, 27, 255-277. Mayer, R. C., Davis J. H., & Schoorman, F. D. (1995). An integrative model of organizational trust. Academy of Management Review, 20, 709-734. Mazurki, D. & Jacoby, J. Forming impressions of merchandise and service quality In J. Jacoby & L. Olson (Eds.) Perceived quality. Lexington Books. McAdams, D. P. (1988) Personal Needs and Personal Relationships. In The Handbook of Personal Relationships. S. N. Duck (Ed). New York, John Wiley & Sons 7-22. McAllister, D. J. (1995). Affects and cognition based trust as foundations for interpersonal cooperation in organizations. Academy of Management Journal, 38, 24-59. McArdle, S. (1997). Overview of the U.S. cosmetics and toiletries market, Drug and Cosmetics Industry, 160, 28-36. McCallum, R. J. & Harrison, W. (1985). Interdependence in the service encounter. In J. A., Czepiel, M. R. Solomon, & C. F. Superanant (Eds.), The service encounter: Managing player customer interaction in service businesses. Lexington, MA: Lexington Books. McKenna, R. (1991). Relationship marketing: Successful strategies for the age of the customer. Reading: MA Addison- Wesley. McKnight, D. H., Cummings, L. L., & Chrvany, N. L. (1998). Initial trust formation in new organizational relationships. Academy of Management Review, 23, 473-491. McKitterick, J. B. (1957). What is the marketing management concept? Frontiers of marketing thought and actions. Chicago, American Marketing Association, 71-82. Melvin, M. K. (1996). WWA speaker advice: Get to know customers. Amusement Business, 100, 5. Meyerson, D., Weick K. E., & Kramer, R M. (1996). Swift trust and temporary groups. In R. M. Kramer & T. R. Tyler (Eds.), Trust in organizations: Frontiers of theory and research, 166-195, Thousand Oaks, CA: Sage. Mishra, A. K. (1996). Organizational responses to crisis: the centrality of trust. In R. M. Kramer & T. R. Tyler (Eds.), Trust in organizations: Frontiers of theory and research, 166-195, Thousand Oaks, CA: Sage. Mishra. A. K. & Spreitzer, G. M. (1998). Explaining how survivors respond to downsizing: the roles of trust empowerment, justice, and work redesign. Academy of Management Review, 23, 567-589. Mishra, J. & Morrisey, M. A. (1990) Trust in employee/employer relationship: a survey of West Michigan. Public Personnel Management, 19, 443-485. Mishra, A. .K. & Mishra, K. (1994). The role of mutual trust in effective downsizing strategies. Human Resource Management, 19, 261-279. Mitchelle W. & Singh, K. (1996). Survival of business using collaborative relationships to commercialize complex goods. Strategic Management Journal, 17, 169-195. Mittal, B. & Lasser, W. L. (1996). The role of personalization in service encounters. Journal of Retailing, 72, 95-109. Mitroff, I. I. & Turoff, M.. (1973). Technological forecasting and assessment: Science and or mythology. Technological Forecasting and Social Change, 14, 329-351. Moorman, C., Zaltman, G., & Deshpande, R. (1992). Relationships between providers and users of market research. The dynamics of trust within and between organizations. Journal of Marketing Research, 29, 314-328. Morns, J. H. & Moberg, D. J. (1994). Work organizations as context for trust and betrayal. In T. R. Sabin, R. M. Carney, & C. Eoyand (Eds.), Citizen espionage: Studies in trust and betrayal. Westport, CT: Praeger. Morris, D. (1996). Behavioral science invades marketing. Marketing News, 30, 4. Morgan, R. M. & Hunt, S. D. (1994) The commitment-trust theory of relationship marketing. Journal of Marketing, 58, 20-39. Mowen, J. & Miner, M. (1998). Customer behavior. Prentice Hall. Munns, A. K. (1995). Potential influence of trust on the successful completion of a project. International Journal of Project Management, 13, 19-24. Nelson, (1974) Nelson-Forsyth, J. (1993). Department stores and industry restructure for the 90’s, Chain Store Age Executive, 34-35. Netemeyer, R. G., Boles, J. S., McKee, D. O., & McMurrian, R. (1997). An investigation into the antecedents of organizational citizenship behaviors in a personal selling context. Journal of Marketing, 61, 85-98. Neu, D. (1991). Trust, contracting and the prospectus process. Accounting, Organizations and Society, 16, 243- 256. Nicholls, J. A. F. , Gilbert, G. R. & Roslow, S. (1998). Parsimonious measurement of customer satisfaction with personal service and the service provider. Journal of Consumer Marketing, 15, 3, 239-253. Nootenboom, B. (1966). Trust, opportunism, and governance: a process and control model. Organization Studies, 17, 985- 1010. Nootenboom, B., Berger, H., & Nooderhaven, N. G. (1997). Effects of trust and governance on relational risk. Academy of Management Journal, 40, 308-338. O’Brien, R. C. (1995). Employee involvement in performance improvement: a consideration of tacit knowledge, commitment and trust. Employee Relations, 17, 110-120. Oliver, R. L. (1980). A cognitive model of the antecedents and consequences of satisfaction decisions, Journal of Marketing Research, 17, 460-469. Oliver, R. L. & Anderson, E. (1994). An empirical test of the consequences of behavior and outcome based sales control systems. Journal of Marketing, 58, 53-67. Olshavski, R. W. (1985). Perceived quality in consumer decision making: An integrated theoretical perspective In J. Jacoby & L. Olson (Eds.) Perceived quality. Lexington Books. Organ, D. W. (1990). The motivational basis of organizational citizenship behavior. In B. M. Staw & L. L. Cummings (Eds.), Research In Organizational Behavior, 43, 72, Greenwich, CT: JAI Press. Orbell, J., Dawes, R., & Scchwarz-Shea, P. (1994). Trust, social categories and individuals: the case of gender. Motivation and Emotion, 18, 109-128. Passingham, J. (1998). Grocery retailing and the loyalty card. Journal of the Market Research Society, 40, 55-63. Paley, N. (1994). Cut out for success. Sales and Marketing Management, 43-44. Parasuraman, A., Berry, L. L. & Ziethaml, V. A. (1991). Understanding customer expectations of service. Sloan Management Review, 39- 48. Pearce, J. L., Bigley, G. A., & Branyiczki I. (In Press). Procedural justice as modernism: placing industrial/organizational psychology in context. Applied Psychology: An International Review. Peterson, R. A. (1995). Relationship marketing and the consumer. Journal of the Academy of Marketing Science, 23, 278-281. Pilotte, W. J. & Gamble, R. K. (1990). The impact of positive and negative worded items on validity of a computer anxiety test. Educational and Psychological Measurement, 603-610. Pollack, J. (1997). New player enters couponing fray. Advertising Age Business Marketing, 82, 16. Podsakoff, P. M., MacKenzie, S. B., & Bommer, W. H. (1996). Transformational leadership behavior and substitutes for leadership as determinants of employee satisfaction, commitment, trust and organizational citizenship behaviors. Journal of Management, 22, 259-298. Porter, M. (1995). Competitive advantage: Creating and Sustaining Superior Performance. New York, Free Press. Powel, W. W. (1996). Trust based forms of governance. In R.M. Kramer & T. R. Tyler (Eds.), Trust in organizations: Frontiers of theory and research. 51-67, Thousand Oaks, CA: Sage. Prince, R. A. (1995). Marketing estate planning to sell the affluent. Trust & Estates, 1334, 43-46. Pruit, D. G. & Kimmel, M. J. (1976). Twenty years of experimental gaming: critique, synthesis, and suggestions for the future. Annual Review of Psychology, 28, 363-392. Ramsey, G. P. & Sohi, R. S. (1997). Listening to your customers: the impact of perceived salesperson listening behavior on relationship outcomes. The Journal of the Academy of Marketing Science, 25, 127-137. Rao, A. R. & Monroe, K. B. (1989). The effect of price, brand name, and store name on buyer’s perception of product quality: An integrative review. Journal of Marketing Research, 26, 351-357. Rapoport, C. & Martin, J. (1995). Retailers go global, Fortune, 62-108. Ravald, A. & Gronoos, C. (1996). The value concept and relationship marketing. European Journal of Marketing, 30, 19-31. Reichheld, F. F. & Sasser, E. W. (1990). Zero defections: Quality comes to services. Harvard Business Review, 68, 105- 111. Reichheld, F. F. (1993). Loyalty based management. Harvard Business Review, 71, 64-73. Remple, J. K., Holmes J. G., & Zanna, M. P. (1985). Trust in close relationships. Journal of Personality and Social Psychology, 49, 95-112. Reynolds, K. E. & Beatty, S. E. (1999). Customer benefits and company consequences of customer-salesperson relationships in retailing. Journal of Retailing, 79. Reynolds, K. E. & Beatty, S. E. (1999). A relationship customer typology. In press. Reynolds, T. J. & Gutman, J. (1984). Laddering: Extending the repertory grid methodology to construct attribute- consequence value hierarchies, In Personal Values and Consumer Psychology. Pitts, R. E. & Woodside, A. G. (Eds.) Lexington, MA. Lexington Books. Riechers, F. F. (1994). Loyalty based management. Harvard Business Review, 71, 64-73. Rich, G. A. ((1997). The sales manager as a role model: Effects on trust, job satisfaction, and performance of salespeople. Journal of the Academy of Marketing Science, 25, 4, 319-128. Richmond, B. (1991). System thinking: a critical set of critical thinking skills for the 90’s and beyond. Systems Dynamics, 190. Rickards, L. (1995). Supercenters entice shoppers, Advertising Age, 1-10. Ring, P. S. (1996). Fragile and resilient trust and their roles in economic exchange. Business & Society, 35, 148- 175. Ring, P. S. & Van De Ven, A. (1992). Structuring cooperative relations between organizations. Strategic Management Journal, 13, 483-498. Roberts, K. H. & O’Rielly, C. A., III (1976). Failures in upward communication in organizations: three possible culprits. Academy of Management Journal, 17, 205-215. Robb, K. (1997). The media as a message. Progressive Grocer, 12, 28-30. Robinson, S. L. (1996). Trust and breach of the psychological contract. Administrative Science Quarterly, 41, 574-599. Ross, W. & La Croix, J. (1996). Multiple meanings of trust in negotiation theory and research: a literature review and integrative model. The International Journal of Conflict Management, 7, 314-360. Ross, W. H. & Wieland, C. (1996). Effects of interpersonal trust and time pressure on managerial mediation strategy in a stimulated organizational dispute. Journal of Applied Psychology, 81, 1-21. Rotter, J. B. (1971). General expectancies for the interpersonal trust. American Psychologist, 26, 443-452 Rousseau, M. D., Sitkin, B. S., Burt, S. R., & Camerer, C. (1998). Not so different after all: a cross discipline view of trust. Academy of Management Review, 23, 393-405. Rudnitski, H. (1992). Battle of the malls, Forbes. Rubinstien, M. F. & Firstenberg, I. R. (1995). Patterns of Problem Solving. Prentice Hall, Englewood Cliffs, New Jersey. Sable, C. F. (1993). Studied trust: building new forms of cooperation in a volatile economy. Human Relations, 46, 1133-1170. Sahlins, M. (1972). Stone age economics. New York.Aldine De Grutyer. Saxe, R. & Weitz, B. A. (1982). The SOCO scale: A measure of the customer orientation of sales people. Journal of Marketing Research, 19, 34-351. Schijns, J. M. C. & Schroder, G. J. (1996). Segment selection by relationship strength. Journal of Direct Marketing, 10, 69-79. Schindler, P. L. & Thomas, C. C. (1993). The structure of interpersonal trust in the workplace. Psychological Reports, 73, 563-573. Schrieshiem, C. A., Powers, K. J., Scandura, T. A., Gardiner, C. C. & Lankau, M. J. (1993). Improving construct measurement in management research: Comments and quantitative approach to assessing the theoretical content adequacy of paper and pencil survey type instruments. Journal of Management, 19, 358-417. Schrieshiem, C. A., Eisenbach, R. J., & Hill, K. D. (1991). The effects of negative items on questionnaire reliability and validity: An experimental investigation. Journal of Educational and Psychological Measurement, 51, 67-78. Schrieshiem, C. A. & Hill, K. D. (1981). Controlling acquiescence response by item reversal: The effect on questionaire validity. Journal of Educational and Psychological Measurement, 41, 1101-1114. Schultz, D. (1993). Marketing from outside in. Journal of Business Strategy, 14, 4. Sheaves, D. E. & Barnes, J. G. (1996). The fundamentals of relationships: an exploration of the concept to guide marketing implementation. Advances in Services Marketing and Management, 5, 215-245. Scott, M.. (1995). Relationship selling. Executive Excellence, 12, 18. Scott, C. L. III. (1980).Interpersonal trust : a comparison of attitudinal and situational factors. Human Relations, 33, 805-812. Schlesinger, L. & Heskett, J. (1991). Breaking the cycle of failures in services. Sloan Management Review, 32, 17-28. Seal, K. (1998). Economic impact study dissects dollars by district, Hotel and Motel Management, 213, 4, 81. Sellers, P. (1989). Getting customers to love you. Fortune, 38-49. Senge, P. M. (1994). The fifth discipline. New York, NY: Doubleday. Shapiro, S. P. (1987). The social control of impersonal trust. American Journal of Sociology, 93, 623-658. Shapiro, D. Sheppard, B. H. & Cheraskin, L. (1992). Business as a handshake . Negotiation Journal, 8, 365-377. Shell, G. R. (1991). Opportunism and trust in the negotiation of commercial contracts: toward a new cause of action. Vanderbilt Law Review, 44, 221-282. Sheppard B. H. & Sherman, D M. (1998). The grammars of trust : a model and general implications. Academy of Management Review, 23, 422-438. Sheth, J. & Parvatiyar, A. (1995). Relationship marketing in consumer markets: Antecedents and consequences. Journal of Academy of Marketing Science, 23, 4. Sheth, J. & Sharma, A. (1996). Supplier relationship. Emerging issues and challenges. Industrial Marketing Management Sirohi, N., McLaughlin, E. W., & Wittick, D. R. (1998). A model of consumer perceptions and store loyalty intentions for a supermarket retailers. Journal of Retailing, 74, 223- 245. Sitkin, S. B. (1995). On the positive effects of legalization on trust. Research on Negotiation in Organizations, 5, 185-217. Sitkin, S. B. & Bies, R. J. (1994). The legalistic organization. Thousand Oaks, CA: Sage. Sitkin, S. B. & Roth, N. L. (1993). Explaining the limited effectiveness of legalistic remedies for trust/distrust. Organization Science, 4, 367-392. Sitkin, S. B. & Strickel, D. (1996). The road to hell: the dynamics of distrust in an era of quality. In R. M. Kramer & T. R. Tyler (Eds.), Trust in organizations: Frontiers of theory and research. 196-215, Thousand Oaks, CA; Sage. Smith, J. B. (1997). Selling alliances: issues and insights. Industrial Marketing Management, 26, 149-161. Smith, J. B. & Barclay, D. W. (1997). The effects of organizational differences and trust on the effectiveness of selling partner relationships. Journal of Marketing, 61, 3-21. Smitka, M. (1994). Contracting without contracts: hoe do japanese manage organizational transactions. In S. B. Sitkin & R. J. Bies (Eds.), The legalistic organization, 91- 108. Thousand Oaks, CA: Sage. Sorrentino, R. M., Hanna, S. E., Holmes, J, G., & Sharp, A. (1995). Uncertainty orientation and trust in close relationships: individual differences in cognitive styles. Journal of Personality and Social Psychology, 68, 314-327. Stacey, R. (1994). The many benefits of relationship marketing. Direct Marketing, 5, 68-70. Stephenson, F. J. & Fox, J. (1987). Corporate attitudes towards frequent flier programs. Transportation Journal, 1, 10-22. Stern, L. W. & El-Ansary, A. E. (1996). Marketing channels, 5th, ed. Upper Saddle River, NJ, Prentice Hall. Stevens, S. (1997). Brand planning: Travel . Brandweek, 38, 34. Strickland, L. H. (1958). Surveillance and trust. Journal of Personality, 26, 200-215. Strutton, D., Tomar, A., & Pelton, L. E. (1993). Relationships between psychological climate and trust between sales associates and their managers in sales organizations. Psychological Reports, 72, 931-939. Stump,R. L. & Srinan, V. (1997). Employing information technology in purchasing: buyer seller relationships and the size of the supplier base. Industrial Marketing Management, 26, 2, 127-136. Stone, E. (1978). Research methods in organizational behavior, Glenview, IL: ScottForeman. Tabachnick, B. G. & Fidell, L. S. (1996). Using multivariate statistics. HarperCollins College Publishing. Taibut, J. W. & Kelly, H. H. (1959). The social psychology of groups. New York, John Wiley Taylor, S. International Management, Portland State University, 1999. Personal communication. Taylor, T. (1998). Better Loyalty measures leads to business solutions. Institute of Marketing Science Newsletter. Thurstone, L. L. (1947). Multiple factor analysis. Chicago: University of Chicago Press. Toffler, A. & Toffler, H. (1994). Creating a New Civilization. The politics of the third wave. Turner Publishing. Treacy, M. & Wiersema, F. (1993). Customer intimacy and other discipline values. Harvard Business Review, 84-93. Turoff, M. & Mitroff, I. I. (1974). A case study of assessment applied to the ‘Cashless Society’ concept. Technological Forecasting and Social Change, 7. New York, American Elsevier. Turnbull, P. W. & Wilson, D. T. (1989(. Developing and protecting profitable customer relationships. Industrial Marketing Management, 18, 233-238. Tyler, T. R. & Degoey, P. (1996). Trust in organizational authorities: The influence of motive attributions on the willingness to accept decisions. In Kramer, R. M. & T. R. Tyler (Eds.), Trust In organizations: Frontiers of theory and research, 331-356. Thousand Oaks, CA: Sage. Uttal, B. (1978). Selling is no longer Mickey Mouse at At&T. Fortune, 17, 98-104. Vloski, D. T., Smith, R. P., & Wilson, P. M. (1994). Electronic data interchange implementation strategies: a case study. Journal of Business and Industrial Marketing, 9, 5-18. Wall, J. A. (1975). The effects of constituent trust and representative bargaining visibility on inter-group bargaining organization. Organizational Behavior and Human Performance. Wayne, S. J. & Green, S. A. (1993). The effects of leader member exchange on employee citizenship and impression management behavior. Human Relations, 46, 1431-1440. Web-Pressler, M. (1995). Looking to rack up a recovery: the limited struggles with a sagging image and sagging sales. The Washington Post, 24. Webb, E. G. (1966). Trust and crisis. In R. M. Kramer, & T. R. Tyler (Eds.), Trust in organizations: Frontiers of theory and research, 288-301, Thousand Oaks, Ca: Sage. Webster, Jr. F. E. (1992). The changing role of marketing in the corporation. Journal of Marketing, 56, 1-17. Weinstien, P. (1998). Preparing for the millennium. Progressive Grocer, 4, 20-24. Wichman, H. (1970). Effects of isolation and communication on cooperation in a two person game. Journal of Personality and Social Psychology, 16, 114-120. Williams, M. R. Attaway, J. S. (1997). Exploring salespersons’ customer orientation as a mediator of organizational culture’s influence on buyer-seller relationship. Journal of Personal Selling and Sales Management, 16, 4, 33-52. Williams, A. R. & Caughlin, C. L. G. (1993). In God we trust, all others pay cash. A prologue to trust vulnerability, deceit in business organizations. Business and Professional Ethics Journal, 12, 67-90. Williamson, O. R. (1993). Calculativeness, trust and economic Organization. Journal of Law and Economics, 36, 433-486. Wilson, D. T. (1990). Creating and managing buyer-seller relationships. ISBN Report, Pennsylvania State University. Whitely, R., Henderson, J., Czaban, L., Lengley, G. (1996). Trust and contractual relations in an emerging capital economy. The changing trading relationship of ten large Hungarian enterprise. Organizational Studies, 397-420. Whitner, E. M. Brodt, S. E. Korsgaard, M. A. & Werner, J. M. (1998). Managers as initiators of trust. An exchange relationship framework for understanding managerial trust worthy behaviors. Academy of Management Review, 23, 513- 531. Woolf, B. P. (1996). Customer specific marketing. The New Power in Retail. Teal Books, Greenville. Wrightsman, L. S. (1991). Interpersonal trust and attitudes towards human nature, In J. P. Robinson, P. R. Shaver, & L. S. Wrightsman (Eds.), Measures of personality and social psychological attitudes. San Diego, Academic Press. Wurdinger, V. (1997). Hair care report. Drug and Cosmetic Industry, 160, 24-32. Xin K. & Pearce, J. L. (1996). Guanxi: connections as substitutes for structural support. Academy of Management Journal, 39, 1641-1658. Young, C. L., Louise, & Wilkison, I. F. (1989). The role of trust and cooperation in marketing channels: A preliminary study. European Journal of Marketing. Yamagishi T. & Yamagishi, M. (1994). Trust and commitment in the United States and Japan, Motivation and Emotion, 18, 129-166. Zand, D. E. (1972). Trust and managerial problem solving. Administrative Science Quarterly, 17, 229-239. Zaheer, A. & Venkatraaman, N. (1995). Relationship governance as an international strategy: an empirical test of the role of trust in economic exchange . Strategic Management Journal, 16, 373-392. Zemeke, R. & Connelian, T. K. (1993). Sustaining knock your socks off service. New York, Amacom. Ziethaml, V., Berry, L. L., & Parasuraman, A. (1996). The behavioral consequences of service quality. Journal of Marketing, 60, 31-46. Zucker, L. G. (1986). The production of trust: institutional sources of economic structure, 1840-1920. In B. M. Staw & L. L. Cummings (Eds.), Research in Organizational Behavoir, 8, 55-111. Greenwich, CT: JAI Press. Zucker, L. G., Darby, M. R., Brewer, M. B., & Peng, Y. (1996). Collaboration structure and information dilemmas in biotechnology. Organization boundaries and trust production. In R. M. Kramer & T. R. Tyler (Eds.), Trust in organizations: Frontiers of theory and research, Thousand Oaks, CA:: Sage. APPENDIX A1 Levels of Observation in Declaring a System A Vertical Level of Observation Every system has subsystems. Every system is also an element of a suprasystem. To apply a vertical view the perceiver should move across levels of observation. If the observer moves one level up, she or he is adopting a perception of the suprasystem level, a perceptual level of which the unit at issue is a subunit. Moving one level down, the observer is adopting a perceptual level of the subsystem of the unit at issue. The subsystem is a partial collection of subunits of the system. Each perceiving role is to be adopted separately from other roles. A Horizontal Level of Observation In this level of observation the observer adopts perceiving roles of different suprasystems and applies the vertical perceived roles in regard to each suprasystem. The main difficulty is to accurately define the separate roles and perceptual filters to be simulated. The multiple perspectives accomplish two purposes: first, they provide an approach to consider and keep track of multitude data and relationships. Second, they bring the observer to perceive the problem from perspectives that are often not considered. Whether it is one problem solver or a team, the same approach should be applied. Thus, this approach assists in dealing with complex situations by chunking the information and abstracting the process. The perceiver is encouraged to role-play a variety of perspectives in the process of understanding a complex system. For example the researcher can choose the topic of service orientation of retail companies as her or his focus, first defining the system as the retail industry and the subunit as a specific retail store. In the second stage, the store becomes the unit, and departments in it become subunits, turning the industry into a suprasystem. Lastly, a department becomes the unit and dyadic relationships of sales associates with customers become subunits. Thus, using at least these five stances, a system can be declared and analyzed. The focus of the researcher can change from service, to relationships, etc., adding a horizontal perspective to the vertical one. APPENDIX A2 Inquiry Models What Are Inquiry Models Inquiry models are philosophical stances that lie at the base of every ill or well structured problem. Each inquiry system emerges from a different philosophical stance. A primary constraint in problem solving is having a rigid preconception of the problem’s nature leading to a solution that is unsuitable for the problem. A major consideration is to determine what the problem really is and what is the question for which solutions are generated. Hegelian and Sinergian Inquiry Models The Hegelian model contends that current plans and opposing alternatives should be examined comparably leading to a third alternative that incorporates the best of two plans. Since, the data can always be manipulated to support contrary views, the least a conflict would bring is more confidence in the current approach. Hegelian inquiry systems hold that the truth is conflictual and is a result of a complicated process that depends on the existence of a plan and an opposed counterplan. These two plans strongly diverge, engage in each others debate over the true nature of a system. The debate results in new plan that reconciles the plan and the counterplan. Data input becomes meaningful and complete only when it is consistent with both the plan and the counterplan. A dialectical debate can be formulated with respect to any issue. On any important issue, intense division of opinion is expected. This inquiry system starts with prior existence or creation of two opposing theories. Information results from data out of the dialectical confrontation. The conflict will expose the underlying assumptions which are usually obscured because of the agreement among experts. Debate is over the underlying assumptions and not always is an agreement achieved. Dialectical inquiry systems are best suited for studying ill structured problems that will produce intense debate over the true nature of the problem. Sinergian models contend that ethical concerns of the philosopher are inseparable from the approach that is being put forth. A holistic approach is required. Considering the problems of ethics, values, and psychology, while considering and analyzing the problem, broadens the range of issues analyzed. Sinergian inquiry systems hold that the truth is pragmatic and relative to the overall goals of the inquiry systems. The model is measured in terms of the ability to define certain systems objectives, to create several means for securing objectives, and to specify new goals that remain to be accomplished by some future inquiry. No single aspect of the system has any fundamental priority over any other. Holistic thinking is a main concept leading to additional components and new variables that broaden the concern base. Sinergian inquiry systems are suitable for interdisciplinary issues. This system includes all previous inquiry systems as submodels in their design. Commands legitimize assumptions and ethics are inherent to any systems’s design. Inquiry systems expand on the potential set of systems designers and users. Every Sinergian inquiry system is dependent upon the future to complete elucidation. Every system is designed to satisfy both present objectives and future objectives. Every description of the present is based on the normative conception of the future. This model provides the broadest possible modeling of any problem, and view all problems using a vertical, horizontal levels, and multiple issues. APPENDIX B The Travel Industry Travel and tourism expenditures were projected to reach $473 billion in 1998, making it the largest services export industry (Galper, 1998). Growing from $26 billion in 1986 to $90 billion in 1996, travel and tourism contributions to the US economy have grown nearly 250% (Doggett, 1998). The travel and tourism congressional districts impact study in 1998, was the first report to rank congressional districts by visitor spending. This study removed historic impediment to raising awareness of the industry’s vital role in the nation’s economy (Seal, 1998). All segments of the US travel industry sales are at an all time record high in profitability, in part because of the strong economy (Boyers, 1998). Travel agencies are an example of a mature industry, offering commodity like products. The travel industry is currently undergoing a major transition, due to the shift in distribution channels which reflects the impact of information technology. Business and leisure travelers are increasingly aware of the travel options available to them and favor using user-friendly software that gives them direct access to relevant information. The Travel Industry Association issued a report stating that on- line sales jumped to 28% in 1997. However, travel providers such as airlines have a vested interest in steering the traveler toward their own services, leading to considerable justification for the continued presence of agencies. Travel agents provide value-added services by integrating the needs of carriers with those of travelers, and continue to be in a unique position to develop value-added services to leisure and corporate travelers (Lewis, Semeijn, & Talalayevski, 1998). Feldman (1998) claims that a study of the Travel Industry Association in 1997 shows that while travelers have boosted the internet for finding schedules and fares but they had not increased their booking rate. For airlines the time is right and the economic growth continues to spur new equity offerings. For the travel industry the outlook is bright. Not only do baby boomers have more cash at their disposal, but they are also spending a greater chunk of that money on leisure activities. This trend is expected to continue. Choices include hotels, casinos, cruises, rental cars, and airlines with beauty services at hotels, outfits providing movies and other entertainment. Thus, the travel industry is seeing a period of high prices and heavy profits. Companies are capitalizing on healthy economy while improving services so they can battle on value rather than on price (Stevens, 1997). Record prices do not mean, however, that there is heavy network spending. The industry is huge but extremely fragmented. Even giants control less than 5% of the market each (Chipkin, 1997). The highly competitive nature of the industry and declining commissions paid by airlines, are projected to lead to widespread travel industry consolidation. Travel organizations increasingly need to target niche groups by appealing to heritage and lifestyles. Travel organizations are differentiating themselves from one another by appealing to niche groups or specific cultures (Fellman, 1998). Hal Rosenbluth from Rosenbluth International believes that by creating an environment of collaboration, innovation, and joy at work, companies of any size can attract and retain the best employees, foster long term relationships with customers, increase profits and compete successfully in a fierce and impatient marketplace. The Professional Hair Care Industry The professional hair care industry is the largest market in the cosmetics and toiletries (17%). At one end the idea of natural and organic products continues to remain strong. In 1995 the professional hair care industry represented $38.7 billion in services and in-salon sales. In 1996 the estimated total dollar value is $40 billion. Most of the overall compounded annual growth of 3.4% is coming from hair color (McArdle, 1997). The aging baby boomers want hair color, hair health- enhancing and volumizing products. African and Hispanic American populations are growing rapidly and so is the demand for special services. The biggest challenges will be in developing products and services that consumers want or in attracting more clients. Forty seven percent of the population patronizes salons and more will (Wurdinger, 997). The hair care industry is one of the three contexts that produces recent streams of research on both service quality and on the relationship between service provider’s gender and the perception of quality. The moderating variable was the consumer gender. APPENDIX C RS Original Items List for Validation Dear Profeesor/Fellow Ph.D. student, Relationship selling is a new- old concept. RS is defined as a one-on-one close relationship between a sales associate and a consumer. A company provides the customer with personal, customized service, thus earning the customer’s satisfaction, loyalty, and a better image of a quality provider. Consequently, sales performance is higher. The goal of RS is to enhance sales, business share and market share through customer loyalty. Relationship selling can be practiced in multiple levels, depending on the type of bonds used to foster customer loyalty. Each group of rewards provided to customers is a relationship selling level. The scale this writer wishes to validate is constructed to measure which psychological and physical rewards customers expect from the relationship. Thus, reflecting the type of bond: emotional, social, physical that exists between a sales associate and a customer. Please read the definition for each one of the three levels of RS. Following the definitions are items representing each level. Please use Table 1 to match each item with the appropriate RS level. Level one: Pricing Incentives RS in this level relies primarily on pricing incentives to secure customer’s loyalty i.e., a free rental video after ten paid rentals, or higher interest rates for longer duration bank accounts. Level two: Social Incentives RS in this level relies primarily on social bonds. Social bonding involves personalization and customization of the relationship i.e., a local chapter newspaper, sponsored events. Level three: Structural Solutions RS relies primarily on structural solutions to important customer problems. Marketers offer customers value-adding benefits that are difficult or expensive for the customer to provide i.e., solidifying services, installing computer terminal in offices of high volume customers to connect them with the company. Items Augments the service with educational activities Offers quarterly price incentives Charges lower fees as the amount purchased units grows Communicates regularly with me Gives me a fair deal Uses my name during transactions Invites me to entertaining events Is attentive to me as an individual Is personal and warm Maintains the same price for customers who have been using the SA’s services for a long time Offers me value-added services Provides a continuity of service Provides services that are not readily elsewhere Responds creatively to my needs Tailors the service and the products to my specific situation Were there items that could fit into more than one category? Please explain: Thank You for Your Cooperation APPENDIX D First Pilot of RS Scale Offers quarterly price incentives Maintains the same price for customers who have been using the SA’s services for a long time Charges lower fees as I the amount of assets managed grows Guides me for better value purchases Is attentive to me as an individual Uses my name during transactions Is attentive to me as an individual Invites me to entertaining events Communicates regularly with me Offers me value-added services Provides service that is above and beyond the norm Provides me with services that are not readily available elsewhere Augments the service with educational activities APPENDIX E Scales Incentives 1. RS Level 1: Pricing incentives a.This SA sends me coupons that are directed at my preferences b.This SA reminds me of relevant members benefits c.This SA helps me save money on products of similar value 2. RS Level 2: Social Incentives a. This SA is attentive to me as an individual b. This SA is always friendly c. This SA uses my name during transactions 3. RS Level 3: Structural Solutions a. This SA does not offer me value-added services b. This SA does not provide me with services that are not readily elsewhere c. This SA does not provide me with options and advise about what to do 4. Relative Perceived Price a. Compared with charges made by alternative stores for similar products this store is X b. Compared with charges made by alternative stores for similar service this store is (X) 5. Relative Perceived Value a. Comparing what I pay for what I get the value here is (X). b. Comparing what I pay to what I get here to that at the competitor’s store the value here is (X) 6. Trust Inventory Scale a. My SA is around when I need her/him b. I can find my SA when I want to talk with her/him c. My SA does things competently d. My SA performs her/his tasks with skill e. My SA does things in a consistent manner f. My SA does the same thing every time the situation is the same g. My SA keeps secrets that I tell her/him h. If I give confidential information my SA keeps it confidential i. My SA treats me fairly j. My SA always gives me a fair deal k. My SA treats me on an equal basis with others l. My SA always tells me the truth m. My SA deals honestly with me n. My SA would not do anything to make me look bad o. I can discuss problems with my SA without her/him using the information against me p. My SA tells me what’s on her/his mind q. My SA tells me what she/he thinks r. I can count on my SA to be trustworthy s. I trust my SA t. My SA follows thorough on promises made to me u. My SA does things that she/he promises to do for me v. My SA readily takes in my ideas w. My SA really listens to me. x. My SA makes an effort to understand what I have to say 7. Perceived Service Quality a. Overall the personnel I interacts with provides high quality service b.This store has a high customer orientation c. This store has responsive service 8. Perceived Merchandise Quality a. Overall sells high quality of merchandise b. Items displayed appropriately c. Wide selection of items 9.Loyalty to the SA a. It is probable that will contact this SA again b. I am willing to shop with this SA again c. I plan to continue doing business with this SA d. I plan to purchase from this SA again 10. Intentional Loyalty to the Store a. I plan to continue to shop at the store b. I plan to use the store for more of my travel/hair needs in the next 12 months c. I plan to recommend this store to my friends 11.Behavioral Loyalty to the Store a. How many word of mouth referrals did you make in the past? b. Have you ever waited to purchase a product or a service even if a comparable one was available from a competitors? c. How many of your family members use travel/hair care service? How many of those use this store? 12. SOCO a. I try to help customers achieve their goals b. I try to achieve my goals by satisfying the customers c. A good agent has to have the best interest of the customer in mind d. I try to get the customers to discuss their needs with me e. I try to influence a customer with information rather than by pressure f. I offer the product that is best suited to the customer’s problem g. I try to find out what kind of product would be most helpful to the customer h. I answer a customer’s question as accurately as I can i. I am willing to disagree with a customer in order to help her/him make a better decision j. I try to bring a customer together with a product/service that helps her/him solve that problem k. I try to give a customer an accurate expectation of what the product/service will do for her/him l. I try to figure out what are the customer needs 13. ADAPTS a. Each customer requires a unique approach. b. When I feel that my sales approach is not working, I can easily change my approach. c. I like to experiment with different sales approaches. d. I am very flexible in the selling approach that I use. e. I can easily use a wide variety of selling approaches. f. It is easy for me to modify my sales presentation if the situation calls for it. g. I am very sensitive to the needs of my customers. h. I vary my sales style from situation to situation. i. I try to understand how one customer differs from another. j. I feel confident that I can effectively change my planned presentation when necessary. APPENDIX F Customer Survey Please return your complete questionnaire in the enclosed envelope. Please feel free to contact me at (503) 524-4857 with any questions about this survey. All survey responses will be confidential and will not be individually reported in any way. THANK YOU! Gilli Ben-Rechav, Ph.D. student, Portland State University. First, please provide some information about your visits to GreatClips: 1. What is the name of the stylist you most frequently interact with?____________________________ 2. Which salon location? ____________________________________________________________ __ 3. How often do you contact this stylist? ___________________________________________________ 4. How much time, on average, does your stylist spend when serving you per visit? ________________ 5. How long have you been a GreatClips customer? __________________________________________ 6. How long have you known this stylist? __________________________________________________ 7. How frequently do you visit GreatClips? ________________________________________________ Please rate the quality of services and merchandise provided by GreatClips and its personnel. Note that ‘service’ refers to haircuts, or appointments and ‘product’ refers to shampoos, hair conditioners, etc. Strongly Strongly Disagree Agree The personnel I interact with provide high quality service…………………... 1 2 3 4 5 GreatClips has a strong customer orientation………………………………… 1 2 3 4 5 GreatClips has a friendly and responsive service……………………………... 1 2 3 4 5 GreatClips sells high quality merchandise……………………………………. 1 2 3 4 5 Items at GreatClips are displayed appropriately……………………….. …….. 1 2 3 4 5 GreatClips had a wide brand selection of items………………………………. 1 2 3 4 5 In the following questions please compare GreatClips to another hair salon you know Compared to a competitor, GreatClips charges more for similar services……... 1 2 3 4 5 Compared to a competitor, GreatClips charges more for similar product….…… 1 2 3 4 5 Compared to a competitor, GreatClips provides high value for the money I pay 1 2 3 4 5 Compared to a competitor, GreatClips provides higher value. …………………. 1 2 3 4 5 Now please answer the following questions IN REGARD TO WHAT THE ABOVE STYLIST ACTUALLY DOES FOR YOU This stylist informs me about hair health……………………………..………… 1 2 3 4 5 This stylist mails me coupons that are directed at my preferences..……………. 1 2 3 4 5 This stylist is attentive to me as an individual…………………. ………………. 1 2 3 4 5 This stylist uses my name during transactions………………………………… 1 2 3 4 5 This stylist is always friendly…………………………………….…………….. 1 2 3 4 5 This stylist reminds me of relevant benefits for regular customers…………….. 1 2 3 4 5 This stylist offers me value added services……………………………………… 1 2 3 4 5 This stylist presents the options and advises me what to do……………………. 1 2 3 4 5 This stylist helps me save money on products or service of similar value……… 1 2 3 4 5 Please respond to the following items by referring to the person identified above. Please keep in mind that some stylist-customer relationships are stronger than others. Just respond as you honestly feel about questions by circling the number that best shows your agreement to that statement. Strongly Strongly Disagree Agree My stylist is around when I need her/him………………………………………. 1 2 3 4 5 I can find my stylist when I want to talk with her/him…………………………. 1 2 3 4 5 My stylist does things competently……………………………………………… 1 2 3 4 5 My stylist performs her/his tasks with skill…………………………………….. 1 2 3 4 5 My stylist does things in a consistent manner………………………………….. 1 2 3 4 5 My stylist does the same things every time the situation is the same…………… 1 2 3 4 5 My stylist keeps secrets that I tell her/him……………………………………… 1 2 3 4 5 If I give my stylist confidential information my stylist keeps it confidential…… 1 2 3 4 5 My stylist treats me fairly……………………………………………………….. 1 2 3 4 5 My stylist gives me a fair deal…………………………………………………… 1 2 3 4 5 My stylist deals honestly with me……………………………………………….. 1 2 3 4 5 My stylist would not do anything to make me look bad………………………… 1 2 3 4 5 I can discuss problems with my stylist without her/him using the information against me……………………………………………………………………….. 1 2 3 4 5 My stylist tells me what’s on her/his mind……………………………………… 1 2 3 4 5 My stylist tells me what she/he thinks………………………………………….. 1 2 3 4 5 I can count on my stylist to be trustworthy……………………………………… 1 2 3 4 5 I trust my stylist…………………………………………………………………. 1 2 3 4 5 My stylist follows through on promises made to me……………………………. 1 2 3 4 5 My stylist does things that she/he promises to do for me…….. ………………… 1 2 3 4 5 My stylist readily takes in my ideas…………………………………………….. 1 2 3 4 5 My stylist really listens to me……………………………………………..…….. 1 2 3 4 5 My stylist makes an effort to understand what I have to say……………….…… 1 2 3 4 5 I plan to continue doing business with this stylist……………………………… 1 2 3 4 5 I plan to purchase from this stylist again……………………………………….. 1 2 3 4 5 I will contact this stylist again………………………………………………….. 1 2 3 4 5 I am willing to style my hair with this stylist again…………………………….. 1 2 3 4 5 Now please answer the following questions IN REGARD TO WHAT YOU LIKE FROM STYLISTS IN GENERAL: I would like stylists to inform me about hair health………………………… 1 2 3 4 5 I would like stylists to send me coupons that are directed at my preferences. . .. 1 2 3 4 5 I would like stylists to be attentive to me as an individual…………………… 1 2 3 4 5 I would like stylists to always be friendly……………………………………… 1 2 3 4 5 I would like stylists to use my name during transactions……………………. 1 2 3 4 5 I would like stylists to remind me of relevant benefits for regular customers…. 1 2 3 4 5 I would like stylists to offer me value added services…………………………. 1 2 3 4 5 I would like stylists to provide service that is not readily available elsewhere… 1 2 3 4 5 I would like stylists to help me save money on products or service …………… 1 2 3 4 5 Please refer to the LAST SIX MONTHS in regard to the frequency of the following behaviors: 1) How many referrals have you made to GreatClips in the past 6 months? 2) Have you ever waited to receive a service or buy a product from GreatClips even if a comparable one was available from a competitor? (please circle one): YES NO 3) What percentage of your personal hair needs do you get in GreatClips? _______________________ 4) How many members in your family visit a hair salon? ____________________________________ 5) How many of them visit GreatClips? ___________________________________________________ 6) What is your average dollar purchase amount for service per visit? $ ________________________ 7) What is your dollar purchase amount for products per visit? $________________________________ Optional Demographic Questions 1) What is your gender (please circle one) Male Female 2) What is your occupation? 3) What is your age? _______________________________________________________ 4) What is the highest level of education you have reached so far? (please circle one): None b) GED c) High School d) Degree/Some College e) Associate f) Bachelor Degree g) Masters Degree h) Ph.D. Degree Travel Agency Customer Survey Please return your complete questionnaire in the enclosed envelop. Please feel free to contact me at 503.524.4857 with any questions about this survey. All survey responses will be confidential and will not be individually reported in any way. THANK YOU! Gilli Ben – Rechav, Ph.D. student, Portland State University. First, please provide some information about your visits to Emmett Travel: 1. What is the name of the travel agent you most frequently interact with?_________________________ 2. Which location? ____________________________________________________________ ________ 3. How often do you contact this agent? ____________________________________________________ 4. How much time, on average does your travel agent spend when serving you per visit? _____________ 5. How long have you been customer of this agency? __________________________________________ 6. How long have you known this agent? ___________________________________________________ 7. How frequent do you contact this agency? ________________________________________________ Please rate the quality of services and merchandise provided by Emmett Travel and its personnel. Note that ‘service’ refers to what the agent does (e.g., informs, accommodates) and ‘product’ refers t o your package components (e.g., hotel room, flight, cruise). Strongly Strongly Disagree Agree The personnel I interact with provides high quality service……………………. 1 2 3 4 5 This agency has a strong customer orientation………………………………… 1 2 3 4 5 This agency has friendly and responsive service………………..……………… 1 2 3 4 5 Overall I purchase high quality services at this agency ………………………. 1 2 3 4 5 This agency has a wide selection of services…………………………………… 1 2 3 4 5 In the following questions please compare Emmett Travel to a competitor: Compared to a competitor, this Travel agency charges more for similar service.. 1 2 3 4 5 Compared to a competitor, this Travel agency charges more for similar products 1 2 3 4 5 Compared to a competitor, this Travel agency provides high value for my money 1 2 3 4 5 Compared to a competitor, this Travel agency provides higher value……………1 2 3 4 5 Now please answer the following questions IN REGARD TO WHAT THE ABOVE AGENT ACTUALLY DOES FOR YOU This travel agent does not inform me about the travel industry………………… 1 2 3 4 5 This travel agent sends me coupons that are directed at my preferences……… 1 2 3 4 5 This travel agent is attentive to me as an individual….………………………… 1 2 3 4 5 This travel agent is always friendly…………………….. ……………………… 1 2 3 4 5 This travel uses my name during transactions…………. ……………………… 1 2 3 4 5 This travel agent does not offer me value added services……………………………. 1 2 3 4 5 This travel agent helps me save money on products or service of similar value……. 1 2 3 4 5 This travel agent does not present options and advise about what to do …………… 1 2 3 4 5 This travel agent reminds me of relevant benefits for regular customers…………… 1 2 3 4 5 Please respond to the following items by referring to the person identified above. Please keep in mind that some agent-customer relationships are stronger than others. Just respond as you honestly feel about questions by circling the number that best shows your agreement to that statement. Strongly Strongly Disagree Agree My travel agent is around when I need her/him…………………………………. 1 2 3 4 5 I can find my travel agent when I want to talk with her/him…………………….. 1 2 3 4 5 My travel agent does things competently……………………………………… 1 2 3 4 5 My travel agent performs her/his tasks with skill……………………………… 1 2 3 4 5 My travel agent does things in a consistent manner…………………………… 1 2 3 4 5 My travel agent does the same things every time the situation is the same…… 1 2 3 4 5 My travel agent keeps secrets that I tell her/him………………………………… 1 2 3 4 5 If I give my travel agent confidential information my travel agent keeps it Confidential………………………………………………………………………. 1 2 3 4 5 My travel agent treats me fairly…………………………………………………. 1 2 3 4 5 My travel agent gives me a fair deal……………………………………………. 1 2 3 4 5 My travel agent deals honestly with me…………………………………………. 1 2 3 4 5 My travel agent would not do anything to make me look bad………………… 1 2 3 4 5 I can discuss problems with my travel agent without her/him using the information against me……………………………………………………………………….. 1 2 3 4 5 My travel agent tells me what’s on her/his mind……………………………… 1 2 3 4 5 My travel agent tells me what she/he thinks……………………………………. 1 2 3 4 5 I can count on my travel agent to be trustworthy……………………………….. 1 2 3 4 5 I trust my agent…………………………………………………………………. 1 2 3 4 5 My travel agent follows through on promises made to me……………………… 1 2 3 4 5 My travel agent does things she/he promises to do for me……..………………. 1 2 3 4 5 My travel agent readily takes in my ideas……………………………………… 1 2 3 4 5 My travel agent really listens to me……………………………………………. 1 2 3 4 5 My travel agent makes an effort to understand what I have to say……………… 1 2 3 4 5 I am willing to travel with this travel agent again………………………. ……….. 1 2 3 4 5 I plan to continue doing business with this agent ……………………………………1 2 3 4 5 I plan to purchase from this travel agent again……………………………………….1 2 3 4 5 It is possible that I will contact this travel agent again…………………………. 1 2 3 4 5 Now please answer the following questions IN REGARD TO WHAT YOU LIKE FROM TRAVEL AGENTS IN GENERAL: I would like agents to present options and advise about what to do…………………. 1 2 3 4 5 I would like agents to send me offerings that are directed at my preferences………… 1 2 3 4 5 I would like agents to communicate regularly with me……………………… ……… 1 2 3 4 5 I would like agents to use my name during transactions………………………………...1 2 3 4 5 I would like agents to be attentive to me as an individual…………………… ………. 1 2 3 4 5 I would like agents to always be friendly……………………………………………… 1 2 3 4 5 I would like agents to remind me of relevant benefits for regular customers…………. 1 2 3 4 5 I would like agents to invite me to entertainment events……………………………… 1 2 3 4 5 I would like agents to offer me value added services………………………………….. 1 2 3 4 5 I would like agents to help me save money on products or service …………………… 1 2 3 4 5 Please refer to the last year in regard to the frequency of the following behaviors: 1) How many word of mouth referrals have you made to Emmett Travel in the past 6 months?_________________ 2) Have you ever waited to receive a service or buy a product even if a comparable one was available from a competitor? (Please circle one): YES NO 3) What percentage of your travel needs do you get at this agency?_________________________________ 4) How many of your family members use a travel agency? ________________________________________ 5) How many of them use Emmett Travel? ? ____________________________________________________ 6) What is your average dollar purchase amount for service per encounter? $ _________________________ 7) What is your average dollar purchase amount for products per encounter? $ _______________________ Optional Demographic Questions 1) What is your gender (please circle one) Male Female 2) What is your occupation? ____________________________________________________ 3) What is your age? __________________________________________________________ 4) What is the highest level of education you have reached so far? (please circle one): a) None b) High School c) Degree/Some College d) Associate e) Bachelor Degree f) Masters Degree g) Ph.D. Degree SA survey Please return your complete questionnaire in the enclosed envelope. Please feel free to contact me at (503) 524-4857 with any questions about this survey. All survey responses are confidential and will not be individually discussed in any way. Thank You! Gilli Ben-Rechav, Ph.D. student, Portland State University. The statements below describe various ways a travel agent might act with a customer. For each statement please indicate the proportion of your customers with whom you act as described in the statement. Do this by circling one of the numbers from 1 to 7. The meaning of these numbers is: 1 True for none of my customers – Never 2 True for a few 3 True for somewhat half of my customers 4 True for half 5 True for somewhat more than a half 6 True for a large majority 7 True for all of your customers – Always Never Always I try to help customers achieve their goals 1 2 3 4 5 6 7 I try to achieve my goals by satisfying the customers…………. 1 2 3 4 5 6 7 I try to get customer to discuss their needs with me 1 2 3 4 5 6 7 I try to influence a customer with information rather than by Pressure…………………………………………….…..………. 1 2 3 4 5 6 7 I offer the product that is best suited to the customer’s problem 1 2 3 4 5 6 7 I try to find out what kind of product would be most helpful to a Customer 1 2 3 4 5 6 7 I answer a customer’s question as to a product/service as correctly as I can 1 2 3 4 5 6 7 I try to bring a customer together with a product/service that best Solves her/his problem 1 2 3 4 5 6 7 I am willing to disagree with a customer in order to help her/him make a better decision 1 2 3 4 5 6 7 I try to give a customer an accurate expectation of what the product Will do for her/him 1 2 3 4 5 6 7 I try to figure what are the customer needs 1 2 3 4 5 6 7 Now please answer the following questions in regard to which of these things you actually do for customers by circling the number of that represents how strongly you agree or disagree with each statement below. Please keep in mind that some of these may not be applicable to your position. In that case, note that you strongly disagree as you do not engage in these behaviors. Strongly Strongly Disagree Agree I use the customer name during transactions…………………………………… 1 2 3 4 5 I inform and provide tips this customer about the travel industry ………..…….. 1 2 3 4 5 I send this customer offerings that are directed at her/his preferences………… 1 2 3 4 5 I am attentive to this customer as an individual……………………………….. 1 2 3 4 5 I am always friendly…………………………………………………………… 1 2 3 4 5 I remind this customer of relevant benefits for regular customers……………. 1 2 3 4 5 I invite this customer to entertainment events at Fairview Travel …………… 1 2 3 4 5 I offer this customer value added services……………………………………. 1 2 3 4 5 I help this customer save money on products or service of similar value……… 1 2 3 4 5 APPENDIX G Factor Analysis on the Incentives Scale Berry (1995) theorized that a relationship between consumers and service providers is based on three bases. These bases are levels of the relationship in which distinct types of bonds that can exist between sellers and buyers. Based on Berry’s theory, this author created a scale to measure these Levels. At the end of a scale development process, this author came up with twelve conceptually consistent items that measure three bases of a relationship. The three bases of a relationship are pricing incentives, social incentives, and structural solutions. Pricing Incentives: These incentives are provided in Level 1 of Relationship Selling. In this level the primary bond between sellers and buyers is pricing incentives: savings, free service/product with a certain number of purchases at regular price, etc., Social Incentives: These incentives are provided in Level 2 of relationship Selling. This level refers to the personalization aspect of a relationship: friendliness, openness, warmth, and communication. Structural Solutions: These incentives are customized service solutions that are facilitated by the store and are not readily available somewhere else. These incentives are highly valued by the customer but are simply and inexpensively provided by a company. The purpose of this factor analysis was to confirm the existence of the three theorized relationship levels. The research questions are: a) Are there any underlying structures to incentives provided in relational exchanges? b) Do these incentive groups differ on the constructs underlying them? The first null hypothesis states that there are no constructs underlying customer incentives. The second null hypothesis states that there are no significant differences among RS levels or provided incentives on underlying constructs. Method A questionnaire was handed out to 127 marketing students. The questionnaire referred to incentives that are provided to students by their hair stylists. This author presented the goals of this survey after students filled out the survey. The presented goal was to validate a scale that measures incentives that bond customers to their sales associates. Next a presentation of the questionnaire items: For all variables students responses ranged from “1” (strongly disagree) to “5” (strongly agree). · The stylist sends the customer coupons that are directed at the customer’s preferences. · The stylist is attentive to the customer as an individual (Attentive) · The stylist informs the customer about hair health (Informs) · The stylist is always friendly (Friendly) · The stylist uses the customer name during transactions (Names) · The stylist presents options and advises me what to do (Options) · The stylist guides the customer to get more for her/his money (GetMore) · The stylist offers the customer value-added services (AddedValue) · The stylist tailors services and product to specific needs of the customer (Tailors) · The stylist reminds the customer of relevant regular customer benefits (Benefits) · The stylist helps the customer save money on products and services of similar value (Save) · The stylist provides services at a level that is “above and beyond the norm” (Beyond Norm) Results Assuming that measurement and unique errors exist in most measures and under many circumstances of surveys, a principal factors extraction with Oblimin rotation was performed to derive underlying constructs in the variable set. Communalities were estimated using squared multiple correlations (SMC). The extraction procedure was performed through SPSS on 12 items for a sample of 127 students. This author assumed normality and linearity of the variables. Exploration Stage Prior to principal factors extraction, Principal Components technique was used as a first step to confirm the probable number of factors. Since the principal components uses variance of standardized scores for all variables, only variables that carry Eigenvalues greater than 1 were extracted. The Principal Components technique extracted 3 factors. Table 1 presents initial statistics. Table 1 Initial Statistics Variable Eigenvalues Variance Pct. Cum. Pct. Coupons 4.35 33.5 33.465 Attentive 2.117 16.3 49.749 Informs 1.121 8.6 58.369 Friendly .819 6.3 64.669 Names .802 6.2 70.842 GetMore .715 5.5 76.341 Options .610 4.7 81.034 AddedValue .566 4.4 85.387 Tailors .460 3.5 90.516 Benefits .432 3.3 94.328 Save .388 2.9 97.752 AboveNorm .265 2.0 100.00 Factors Retention This author used five criteria in the decision of retaining factors: Eigenvalues greater than 1, Scree Plot test, significance of percentage of variance explained, iteration convergence, and a simple structure. Principal Components extracted three factors in the initial statistics. This author believes that three factors should be extracted. A few reasons support this author’s decision: · Three factors accounted for approximately 60% of the variance among the variables. · The contribution of each of the other factors to the variance among the variables was flattening out. · A Scree Plot test clearly showed that three factors account for most of the variance, while the remaining factors all range form .82 to .27 Eigenvalues. · These initial statistics were derived using only twelve iterations. Examining the structure matrix, this author noticed clear clusters of variables around each factor. Table 2 presents the rotated solution. Table 2: Rotated Solution Variable Factor 1 Factor 2 Factor 3 Coupons 9.849E-02 .710 -.283 Attentive .375 1.322E-02 .693 Informs .801 .273 .185 Friendly .105 -.8623E-02 .675 Names .224 5.961E-02 .761 GetMore .714 .238 .149 Options .834 .202 .433 AddedValue .723 .432 .482 Tailors .611 .356 .584 Benefits .209 .748 3.168E-02 Save .348 .785 .124 AboveNorm .475 .422 .615 Interpretation Table 2 illuminates a simple structure with items clustering around three factors. The first factor explained 33% of the variance, the second factor explained 16% of the variance and the third factor explained 8% of the variance. The simple structure confirms the existence of the following three factors. Factor 1 is the most important factor with five variables loading on it. Significant correlationa are greater than .40, ranging from .611 to .834. Variables that cluster around factor 1 involve structural solutions incentives that are applied in Level 3 of relationship selling. These are incentives such as providing the customer with information, presenting options and advising the customer, offering value-added services, and tailoring services to the customer needs. In this level the variable with the highest loading was found to be informing the customer about hair health. The second most important variable in this category was found to be presenting options and providing the customer with advise. Variables that clustered around factor 2 involved pricing incentives. All loadings were above .40 and ranged from .710 to .790. These incentives were sending the customer coupons, reminding the customer of relevant benefits to regulars, and helping the customer save money. All these incentives were theorized to be provided in Level 1 of relationship selling. All incentives were found to be almost equally important, led by sending customers coupons that are directed at their preference, followed by reminding the customer of relevant benefits and helping the customer save money. Variables that loaded on factor 3 involved social incentives. All loadings were greater than .40 and ranged from .615 to .761. These incentives included being attentive to the customer as an individual, being friendly, using the customer name during transactions, and providing service that is above and beyond the norm. The most important incentive in this level was found to be using the customer name during transactions followed by being friendly. AddedValue, Options, Informs, and Tailors were theorized to load on the same factor and loaded on the first factor. These items measure structural solutions (RS Level 3). AddedValue and Options had low weights on the third factor (.43) and high weights on the first factor (.83). To reduce cross loadings these items were negatively worded. Tailors loaded on two factors with small differences in weights (.611, .584) and was taken out of the scale. Thus, Level 3 of RS was measured by three items: INFORMS, OPTIONS, ADDEDVALUE. GetMore clustered around the first cluster but was theorized to cluster with other items that measure pricing incentives. This item was thrown out of the scale. Three items that were theorized to cluster around the same factor loaded on Factor 2: SAVE, BENEFITS, COUPONS. These items measure pricing incentives (RS Level 1). Attentive, Friendly, Names, and AboveNorm were theorized to load on the same factor and loaded on factor 3. AboveNorm loaded on three factors with small differences among loadings. AboveNorm was therefore taken way from the scale leaving the three other items as measuring social incentives (RS Level 2). Table 3 presents items by factors. Table 3: Items by Factors Structural Solutions Pricing Incentives Social Incentives The SA does not informs me about hair health The SA sends me coupons that are directed at my preferences The SA is always attentive to me as an individual The SA does not present options nor advises me what to do The SA reminds me of relevant benefits for regular customers The SA is always friendly The SA does not provide me with added value The SA helps me save money on products or services of similar value The SA uses my name during transactions Factor Intercorrelatitons Intercorrelations between factor 1 and 2 (r= .34) show that as structural solutions increase pricing incentives increase as well. This reflects customer expectations that price would remain reasonable despite additional value that is provided through the relational exchange. It is also possible that customers in this sample are aware of the value of structural solutions but are not willing to pay more for it and thus, stores do not charge more although they provide more value. Intercorrelations between factor 1 and 3 (r= .35) show that as structural solutions increase social incentives increase as well. This is explained by the fact that in order to clearly understand customer needs and provide customized service, the sales associate needs to develop more communication that may strengthen a relational exchange. Also, more tailoring of products and service, requires more communication that is included in the social incentives, Level 2 of relationship selling. Table 3 shows the correlations among the factors. Discussion The first null hypothesis was rejected. Overall the analysis confirmed the existence of three underlying constructs that reflect levels of relationship selling with distinct incentives that are provided to customers in each level. The simple structure was adequate. Factor 1 explained most of the variance, factor 2 explained more of the variance that was not explained by factor 1, and factor 3 explained more of the variance that was not explaines by factors 1 and 2. All loadings of the nine items were high and each variable loaded high on one factor and low on the others. Items that loaded differently than hypothesized or cross loaded on more than one factor were taken out of the scale. The second null hypothesis was also rejected. Factor 1, structural solutions was different than other factors in the number of variables that loaded on it and in its importance in terms of explained variance (33.5%). Implications of this factor analysis are to focus on meeting expectation of customers. Application of structural solutions when customers desire other incentives could result in greater costs and customer dissatisfaction leading to customer defection. Since factors are intercorrelated, a change in one factor is likely to cause a change in other factors. Lower application of some incentives will decrease the application of other incentives. Decision makers are to be aware of the intercorrelations and consequences of specific changes. Changes in specific incentives that have high estimates of communalities and high weights, i.e., informing the customer on hair health, will affect all variables in the structural solutions category. APPENDIX H Inter-Correlations Between Perceived Quality of Service and Perceived Quality of Merchandise Travel Agency Sample Model Zero Order Partial Semi-Partial Service Quality .481 .301 .275 Merchandise Quality .405 .109 .095 Hair Salons Sample Model Zero Order Partial Semi-Partial Service Quality .655 .536 .468 Merchandise Quality .488 .223 .168 APPENDIX I Covariate Structural Modeling for the Travel Agency Sample