The effectiveness of loyalty programs: an application in the hospitality industry.

Hikkerova, Lubica

I. INTRODUCION

Increased competition has forced firms to constantly look for new ways to be more attractive and thus more competitive. As the cost of keeping an existing customer is lower than the cost of acquiring new customers, many firms have launched loyalty programs. These programs are marketing strategies aimed at rewarding and thus encouraging loyal behaviour. However, their cost when compared to the benefits which they bring the firms has made several authors question their real effectiveness (Meyer-Waarden, 2002). The cost of the programs is often underestimated (and has led certain firms to re-examine them i.e. Air France with its program Flying Blue). Moreover, as competitors imitate the schemes the initial competitive edge is gradually eroded. The tourist sector was the pioneer in this field. Firstly, it was airline companies followed by hotel chains. This article proposes to question the effectiveness of loyalty programs in the hotel sector. This involves ranking the determinants of loyalty: is a customer loyal to a hotel or a chain of hotels for practical reasons and comfort; because he is satisfied; because the loyalty program arouses his interest, or finally for all these factors together? The answer to these questions should make it possible to determine if a loyalty program and its associated advantages can render the other choice factors secondary and thus increase the probability of the hotel being chosen. More precisely, we want to both gain a better understanding of the antecedents of commitment and trust and how these factors influence customer loyalty, and to determine the impact of the loyalty programs. After reviewing the existing literature on loyalty theories, we propose a conceptual model which aims to explain the loyalty of customers to a hotel or a chain of hotels and thus evaluate the efficiency of loyalty programs.

To validate our model empirically, we questioned 390 users of hotel services in Bratislava. The specification of the structural model was made in two stages in order to identify the antecedents of loyalty. We first tested the general structural model, then the reduced model after removing non-significant variables. Lastly, we tested the impact of the loyalty programs by analyzing the possible differences in response between members and non-member of these schemes.

II. ANTECEDENTS OF LOYALTY

In this section, we offer a concise review of the main contributions to marketing research literature on the concepts of loyalty and customer loyalty schemes in order to better understand their intrinsic links, and to define our conceptual framework.

A. From Loyalty to Loyalty Program

The exchange marketing approach has evolved over the last few decades as companies have changed their preoccupations. The concept of traditional marketing was interested mainly in satisfying the needs and desires of the customer and in creating an orientation towards a product or a brand without taking into account the possible social relationship between the actors. From this perspective, communication between the two parties (companies and customers) was almost non-existent. The latter were regarded as complete strangers with no interest in the personality of their partner because the role which was allotted to them consisted in respecting their obligations in accordance with the terms of the exchange. This exchange, named by researchers "transactional exchange" (Dwyer et al, 1987) or "restricted exchange", is not seen as enduring as each transaction is regarded as being unique and independent of other transactions which had already taken place or which could take place in the future. The customer has no preference, or consideration, or loyalty to the seller. Reciprocally, the seller does not express any desire for knowing the preferences of the customer, nor his perception and personal opinion about the service or product sold. This highly restrictive vision of an exchange was criticized for its limited ability to explain behaviour and real practice. It has been gradually replaced by a new line of research which has developed over the past decades, both on the theoretical and practical level. It focuses on the concept of relational exchange and makes the relationship between the partners the main object of the analysis (Oliver, 1999).

This concept of relational exchange is characterized by its realistic aspect based on two basic hypotheses; the continuous temporal character of the exchange process, and the existence of a social link between the two parties in the exchange (company and customer). We notice the introduction of the concept of co-operation between the two partners as contributing to the success of the exchange within a relational context. That implies a certain level of knowledge and of reciprocal understanding between the two parties, which supposes that the development of social links between the actors encourages the partnership to endure.

The existence of a partnership relationship between the two actors involves the reduction of uncertainty between them, the increase in the frequency of exchanges, and a better information flow, thus allowing each party to gain better knowledge of the other's expectations, and create an environment of shared trust and mutual commitment. It is within this framework that customer loyalty is created.

Researchers in marketing then became increasingly interested in the concept of loyalty (especially brand loyalty) and tried to define and measure it more precisely. To define this concept, we chose the definition offered by Jacoby and Kryner (1973) which asserts that loyalty is "a biased behavioural response (i.e. non random) expressed over time by some decision making unit, with respect to one or more alternative brands out of a set of such brands, and is a function of a decision-making process". Therefore, repeat purchase (over time), and oriented (non random) behaviour, as well as a favourable attitude must be observed to introduce the concept of loyalty. This definition raises a subjacent question: when speaking about loyalty, should researchers limited themselves to observation and description of behaviour, or should they also take into account the feelings, beliefs, intentions and attitudes of the actors? Initial research based on the observation and description of repeat purchase behaviour to characterize loyalty (behaviourist approach) was widely criticized because it does not take into account the cognitive process influencing the purchase decision. New approaches have emerged which consider that loyalty is formed, in addition to repeat purchase behaviour, from a whole range of psychological (decision-making, evaluation) factors of a cognitive, emotional or conative nature. The literature has tried to determine certain explanatory variables of loyalty which we will outline in the following article, whilst accepting that all the factors which could be associated with loyalty have not as yet been identified:

1. Loyalty depends on attitudes towards a brand or service provider. These attitudes were defined as being all the intentions and predispositions created from previous purchases which create loyal behaviour. This concept has been the focus of many research projects; however, the major reproach relates primarily to problems of the reliability and validity of attitudinal measurements.

2. For a customer, the intention to repurchase does not lead automatically to an actual decision. In fact, a phase of revaluation is introduced to identify the conformity of expected real quality (following the consumption experience), which introduces the notion of satisfaction (1) as a precursor to loyalty.

Thus, perceived satisfaction towards a brand will stimulate the preference for this brand, lead to building a positive attitude, reinforcing the intention to repurchase, and will lead to loyalty.

3. Loyalty is built and developed within the framework of an enduring relationship that the customer builds with the company following his different consumption experiences. Trust and mutual commitment are therefore determining factors for an enduring relationship and building customer loyalty.

Research literature has already dealt extensively with these two concepts which determine relationship quality. We therefore limit ourselves, in this article, to giving short definitions.

Commitment has been defined in the literature as being an implicit or explicit promise to continue the relationship between the partners in the exchange; it is also the desire to maintain a mutually appreciated relationship.

More recently, commitment has been identified as a concept with two dimensions (Ganesan and Hess, 1997; Gilliland and Bello, 2002). The first is cognitive or calculated linking the continuity of the exchange to the anticipation of gains or losses resulting from keeping up or terminating the relationship. The other dimension is emotional, reflecting the degree of attachment towards the brand or the company, appreciation of the partner, having shared values and interests.

Research literature has also covered trust by dealing with the different aspects which make it up such as honesty, reliability, the reduction of uncertainty, credibility and goodwill (Gurviez and Korchia, 2002).

4. Loyalty is also influenced by the switching costs in that efforts made by a consumer to change service provider or brand are high compared to the gains expected from the switch which can sometimes prove to be insignificant.

As a result, creating an enduring and solid relationship with the customer becomes a major concern for companies. It has become of paramount importance to find a means to satisfy the customer's interests (both cognitive and emotional), and companies are looking for ways to build an enduring, long-term relationship with their customers through relationship marketing. Many firms have launched expensive loyalty programs whose effectiveness may not be guaranteed. When questioning the effectiveness of these programs, it appears necessary to look at antecedents to loyalty and to understand how customer loyalty programs have influenced them.

B. Loyalty of Hotel Customers: A Proposed Conceptual Model

On the basis of the literature review, we propose a conceptual model (2) which helps explain customer loyalty to a hotel or a hotel chain. This model is based on two concepts; commitment and trust. We try to identify the antecedents: economic and social factors as well as the influence of the customer loyalty programs.

[FIGURE 1 OMITTED]

A. The Triptych of Trust-Commitment-Loyalty

To define the concept of trust, we refer to Gefen et al. (2003) for whom "trust is a set of beliefs relating to the honesty, the goodwill (the partner in the exchange should act in the interests of the other party), the competence (the ability of the partner to respond to the other party's needs) and to the predictability (the partner's behaviour should be constant and predictable) of the service provider".

Commitment has many definitions in marketing literature and the concept has evolved significantly in recent years. That is why we have chosen to adhere to the definition offered by Rylander et al. (1997) for whom commitment corresponds to the desire to develop an enduring relationship between a consumer and a firm with an aim to generating long-term benefits for both partners. With reference to McKnight et al. (2002), we link trust and commitment to the theory of reasoned action (Fishbein and Ajzen, 1975). Thus, the reactions of the customer can be modelled by the following sequence: Belief, attitude, intention, and behaviour. By making an analogy with this

proposal, we suppose that trust (expressed as belief) leads to consumer commitment (attitude). This commitment then leads to an intention and repurchase behaviour expressed through behavioural loyalty. Trust seems to be an implicit element in the intention of loyalty and constitutes a fundamental factor in the long-term orientation of the consumer. Although the argument for the influence of trust on the attitudinal dimension of loyalty (commitment) finds more support, the argument suggesting the positive impact of trust on the behavioural dimension of loyalty also finds a certain amount of theoretical and empirical support (Jarvenpaa and Tractinsky, 1999).

With reference to the results of the research mentioned above, we formulate the following hypotheses:

H1: For a consumer, trust in a company is positively linked to his commitment to the company,

H1b: Commitment is associated with greater loyalty,

H1c: Trust is associated with greater loyalty.

B. The Mediating Impact of Loyalty Programs

Consumers adhere to a loyalty program to be rewarded by additional benefits (monetary or in kind) or better recognition (additional services linked to the program). Loyalty programs appear in our conceptual model to be a mediator variable. It is important then to identify how their advantages are seen by the customers, if they influence their commitment and trust with respect to the hotel and consequently their decision to book a room.

We therefore made the following hypotheses:

H2: Loyalty programs are associated with a high degree of commitment.

H3: Loyalty programs are associated with a high degree of trust.

H1a: Loyalty programs are associated with greater loyalty.

We will now look at the antecedents of commitment and trust, which we put into three categories: economic factors, related factors and social factors. According to Meyer-Waarden (2002), trust, commitment and loyalty programs reinforce the functional dependence of the customer compared to the firm both in a positive way because of a higher perceived economic value of the supplier-customer relationship, and negative due to an increase in the perception of switching costs. In the services sector, the switching costs include costs in terms of time, money and effort required to change the service provider. The hotel trade cannot charge for explicit switching costs, contrary to some other sectors (such as banking through fees for closing and transferring an account). The only recourse for hotels is to create a value offer which would make it difficult for the customer to find a substitute among competitors. Thus, one of the aims of loyalty programs is to increase the value offer for the consumer and consequently the perceived cost of switching so as to create customer lock-in. Moreover, the perception of this value offer is related to its degree of commitment with respect to the firm. A customer who has significantly invested in the loyalty program finds that his switching costs are far higher than a customer who is less committed in his relationship with the firm (Lee et al., 2005).

We came to the following hypotheses:

H4a: For a consumer, a firm's perceived economic value is positively related to his commitment to the firm.

H4b: Loyalty programs are associated with a high level of perceived economic value.

H5b: For a consumer, the perceived costs of switching firms are positively linked to his commitment to the firm.

H5a: Loyalty programs are associated with a high level of switching costs.

Beyond the perceived value, the second condition to develop consumer commitment is, according to resource theory, the presence of certain factors linked to the firm's resources which determine its competitive position. The research literature generally identifies two: reputation in terms of quality and personalization of service (Hunt and Morgan, 1995). How a consumer attributes a firm's reputation for quality depends on his perception of the quality of past service but also anticipated service, therefore he makes many value judgments about the firm's services or products (Nguyen, 1991). It generally relates to the overall reputation of the firm and not to a particular service or product. The literature showed the direct influence of this factor on loyalty but also via commitment and trust (Frissou, 2004). Indeed, the better the reputation, the greater the trust in the service provider. This is also due to the fact that the customer feels less uncertainty about the service and is thus reassured.

This leads us to make the following hypotheses:

H6a: For a consumer, the firm's reputation for quality is positively linked to commitment to the firm.

H6b: For a consumer, the firm's reputation for quality is positively linked to trust in the firm.

H6c: Loyalty programs are associated with the firm having an excellent reputation for quality.

Another factor linked to commitment and trust is the personalization of the service. It creates an emotional attachment to the brand or service (Rust et al., 2000) because the customer feels recognized and that he is being given special treatment. Thus, the personalization of the service becomes a crucial element in the process of developing customer loyalty. In the hotel trade one of the aims of loyalty programs is to provide specific services for members, such as choice of room (smoking or nonsmoking, type of bed, etc.). Moreover, these programs make it possible to save the customer's preferences and consequently to adapt the service to their needs without having to ask them again. The reservation process is also made easier as the need to exchange information is reduced.

We make the following hypotheses:

H7a: For a consumer, a personalized service is positively linked to commitment to the firm.

H7b: For a consumer, a personalized service is positively linked to trust in the firm.

H7c: Loyalty programs are associated with a better personalized service from the firm.

The third element among the relational factors which encourage the intensification of a marketing relationship is due to social factors. This includes communication and sharing certain ethical or moral values (Czepiel, 1990). The communication between the hotel and its customers is defined as process of transmitting a message to its customers in a formal or informal way which is aimed at informing them as well as promoting products, services and activities. If the communication of the hotel is frequent and that information relevant and is transmitted at the right moment, then the consumer's commitment and trust will be strengthened as his perception of the service meets his expectations. The communication relates to all the information given out during the different phases of the purchase: pre-sale (advertising), sale (information given out during the reservation and in the hotel especially at the reception), and after-sale (satisfaction questionnaire and feedback). Loyalty programs are generally used by the firms as a preferential vehicle of their communications policy. This communication has been made easier and less expensive with the spread of electronic tools such as email and newsletters.

These results lead us to make the following hypotheses:

H8a: For a consumer, the firm's communication is positively linked to commitment to the firm.

H8b: For a consumer, the firm's communication is positively linked to trust in the firm.

H8c: Loyalty programs are associated with high-quality communication.

Lastly, shared values correspond within our conceptual framework to the consumer's perception that the aims, policies and beliefs of the hotel are compatible with his. The perception of shared values with the hotel increases the marketing relationship and supports relational commitment as well as trust. Even if it was not in the initial strategy of the loyalty programs to include ecological or social concerns (via for example systems of donating to charities or environmental organisations instead of directly rewarding the members of the program), companies use them to communicate about their sustainable development policy which is increasingly put at the heart of their strategy. One can quote as example the case of Accor which has made sustainable development one of the major axes of its strategy and which strongly encourages all hotels in the brand, even those which are franchises, to apply its 'Environment' charter (Guillon, 2006).

We make the three following hypotheses:

H9a: For a consumer, shared values are positively linked to commitment to the firm.

H9b: For a consumer, shared values are positively linked to trust in the firm.

H9c: Loyalty programs are associated with strong shared values.

III. EMPIRICAL RESULTS

Firstly, we will describe the sample and then study the variable pairs in order to detect possibly connections using a Chi-2 test and to describe it better. Then we will carry out the test of the conceptual model, and the analysis of the impact of loyalty programs.

A. Descriptive Analysis of the Sample

The questionnaire was completed, using face-to-face interviews, from September 7th to 18th, 2009 in Bratislava (as guests left 3, 4 and 5-star hotels). We completed 390 questionnaires but after elimination of those containing errors, we finally retained 385. Since in the hotel trade the practice of using loyalty programs concerns mainly hotels with at least 3 stars, we favoured respondents who had used this class of hotel at least once. However, we were careful to make sure that there was some variance between respondents in their usual practice of hotel frequentation: the type and frequency of hotel stays, the category of the hotels used and whether the respondent was a loyalty card holder.

The sample is unbalanced in terms of male-female ratio (the difference between 58.2% and 41.8% is significant with a risk of 1%). The higher proportion of men is understandable as they are often business customers: 55% travel for business and 36% for business and leisure. We noted that in terms of age and profession our target concerned a specific segment of customers: 65% of people questioned were between 26 and 45 years old and 52% were in higher social brackets (entrepreneur, freelancer, manager) whereas we excluded from this category civil servants and pensioners who could belong to this category. This is due to the specific sector chosen--customers of 3, 4 and 5-star hotels in Bratislava. It is therefore logical that 52% answer that they generally stay in 4-star hotels.

Moreover, 69% stay in hotels more than 7 times a year and 26% have at least one loyalty card for hotel chains. It is thus a population which regularly uses hotel services and of which 26% are members of a loyalty program.

We also note that these results are due to the fact that 27% of the respondents are of foreign nationality; reinforcing the business customer profile. Indeed, the crossing of variables "reason for stay" and "nationality" with a Chi-2 test shows a significant relation (with a risk of 2%) between foreign nationality and the business motive for the stay. Of the respondents of Slovak nationality only 21% are members of a loyalty program, compared to 43% for those of foreign nationality.

B. Analysis of the Structural Model

The specification of the structural model was carried out in two stages. First of all, we tested the general structural model for the sample from the entirity of the conceptual model in order to test the relationships observed. These relationships do not include the part relating to loyalty programs since we will analyze in the second step the impact of these programs while testing if the members of these programs respond in the same way as the others. Then, we worked out so called 'reduced' structural models by removing the non-significant variables (t-value < 1.96) to check the significance of the identified variables and to increase the explanatory ability of the model. If the structural models obtained are significant as well as the different relationships retained, we will be able to validate or disprove the various hypotheses of our conceptual model. We also carried out in parallel 20 guided interviews (3) with some respondents taken randomly from our sample in order to go beyond the analysis of our quantitative results and to better understand certain consumer behaviour concerning hotel products.

The structural model reduced to significant relationships gives both the t-value of remaining relationships and a higher overall explanatory capacity of the model. In particular, Goodness of the Fit Index goes from 0.78 to 0.84. Our structural model makes it possible to explain 84% of the covariations observed.

From this reduced model, we validate the hypotheses for which relationships are significant. First of all, the two samples confirm the three hypotheses relating to the triptych of trust-commitment-loyalty, that is to say:

H1: For a consumer, trust in a company is positively linked to his commitment to the company.

H1b: Commitment is associated with greater loyalty.

H1c: Trust is associated with greater loyalty.

Concerning the antecedents of commitment, the five following hypotheses are validated:

H4a: For a consumer, the perceived economic value of a firm is positively linked to his commitment to the firm.

H8a: For a consumer, the firm's communication is positively linked to commitment to the firm.

H9a: For a consumer, shared values are positively linked to commitment to the firm.

Then, the three following hypotheses, relating to the antecedents of trust, are validated:

H6b: For a consumer, the firm's reputation for quality is positively linked to trust in the firm.

H8b: For a consumer, the firm's communication is positively linked to trust in the firm.

H9b: For a consumer, shared values are positively linked to trust in the firm.

The absence of relationships, shown in the general structural model, enables us to reject the four following hypotheses:

H5b: For a consumer, the perceived costs of switching firms are positively linked to his commitment to the firm.

H6a: For a consumer, the firm's reputation for quality is positively linked to commitment to the firm.

H7a: For a consumer, a personalized service is positively linked to commitment to the firm.

H7b: For a consumer, a personalized service is positively linked to trust in the firm.

Our structural model resulted in removing the variable "switching costs" because the Student's t-distribution was not significant. This means that the impact of loyalty programs on this variable cannot be tested. We thus remove our analysis hypothesis H5a (loyalty programs are associated with a high level of switching costs).

With regard to the impact of the loyalty programs, the Chi-2 test shows differences in significance for four relationships when one separates the members of loyalty programs from the non-members. It concerns the following relationships:

Economic value [right arrow] Commitment

Communication [right arrow] Trust

Shared values [right arrow] Trust

Trust [right arrow] Loyalty

In fact, only one relationship sees its coefficient decreasing when one looks at members of a loyalty program. This is the relationship between "Economic Value [right arrow] Commitment" This means that loyal customers are less sensitive to economic value (in particular to the price of quality--price). According to the guided interviews, this is because the population is unaware of the advantages which they benefit from as a member of the loyalty program.

The difference for the relation "Communication [right arrow] Trust" shows that the hotel's communication policy has more influence on loyal customers' trust.

The difference for the "Shared values [right arrow] Trust" relationship shows that trust is based more on shared values for loyal customers than for non-members. In particular, the fact that the Slovak business customer (who is highly dominant in the sample) puts so much more trust in a hotel of which he is a member of the loyalty programme reflects the type of person he is.

The difference for the relationship "Trust [right arrow] Loyalty" shows that trust is a dominating antecedent of loyalty for the members of a loyalty program in the sample. It will be therefore more effective to act on the antecedents of trust to develop loyalty in this type of customer. Thus, communication and shared values are the main levers to develop this trust and consequently the loyalty of this population.

From these results, we can validate the following hypotheses:

H3: Loyalty programs are associated with a high degree of trust.

H4b: Loyalty programs are associated with a high level of perceived economic value.

It is partially validated. The relationship is significant but only through the "commitment" variable.

H8c: Loyalty programs are associated with high-quality communication. It is partially validated. The relationship is significant but only through the "trust" variable.

H9c: Loyalty programs are associated with strong shared values. It is partially validated. The relation is significant but only through the "trust" variable.

Two hypotheses are rejected completely. They are:

H2: Loyalty programs are associated with a high degree of commitment.

H6c: Loyalty programs are associated with the firm having an excellent reputation for quality.

H7c: Loyalty programs are associated with a better personalized service from the firm,

Lastly, the hypothesis of the impact of loyalty programs (H1a) was tested separately due to its two behavioural and attitudinal components. Indeed in this research, loyalty is measured by five variables. The first two relate to behavioural loyalty (intention of repurchasing a room in a hotel), while the following three relate to attitudinal loyalty (positive attitude to the hotel which I transmit to others):

* Behav1: In the next 12 months, are you likely to return more often to hotels in this chain?

* Behav2: In the next 12 months, are you likely to book more rooms from hotels in this

chain compared to other hotels?

* Att1: I am ready to recommend this chain of hotels to my parents and friends.

* Att2: When the subject comes up in conversation, I am ready to recommend this chain of hotels.

* Att3: I am delighted to tell others that I stayed in this chain of hotels.

We will test the impact of loyalty programs on customer loyalty by measuring the percentage of answers, from members and non-member, whose score is equal to or higher than 5 for these five items (on our Likert scale with 7 levels, 4 corresponds to a situation where the consumer is indifferent), then by checking our results with a Chi2 test if the differences obtained are significant.

We observe that the loyalty programs positively influence customer loyalty. The Chi2 test of average difference between members and non-members of loyalty programs are significant for two variables with a risk of 5%. In fact, only behaviour is affected. This shows that the loyalty of these consumers is connected with a "false loyalty" as they are not ready to recommend a hotel to their close relations (friends and parents), or when the subject comes up in conversation and they do not feel especially proud to have stayed in this type of hotel.

The hypothesis H1a (loyalty programs are associated with stronger loyalty) is thus partially validated for this sample (SK).

IV. CONCLUSION

This article makes it possible to better understand the determinants of loyalty and consequently the effectiveness of loyalty programs in the hotel sector. In particular, we analyzed the impact of the relational strategies of the company on trust and commitment which influence customer loyalty. This research led to the three following types of results in hotel trade:

* The antecedents of loyalty are economic value, reputation, communication and shared values, but only the two last variables act through both commitment and trust. They are thus the two most important antecedents.

* Members of loyalty programs are more loyal than non-members.

* The antecedents of loyalty for loyalty program members and non-members are significantly different. Loyalty programs therefore do have an influence on customer loyalty.

Although these general results can appear intuitive, they confirm the results of certain studies which affirm that these programs enlist consumers who are not really committed or who the firm might wish to avoid (Meyer-Wardeen, 2000).

The loyalty of the Slovak consumers is connected with a "false loyalty" as they are not ready to recommend a hotel to their close relations (friends and parents), or when the subject comes into the conversation and they do not necessarily feel proud to have stayed in this type of hotel. In targeting these customers, the loyalty programs only partially attain their objective. This study shows that it appears more efficient to improve certain attributes of the service (such as communication and shared values, the two most important antecedents according to this study) than to add monetary or other advantages to loyalty programs which are already extremely expensive.

REFERENCES

Czepiel, J.A., 1990, "Service Encounters and Services Relationships: Implications for Research," Journal of Business Research, 20, January, 13-21.

De Wulf., K., G. Oderkerken-Schroder, and D. Iacobucci, 2001, "Investments in Consumer Relationships: A Cross-Country and Cross-Industry Exploration," Journal of Marketing, 65, October, 33-50.

Dwyer, F.R, P.H. Schurr, and S. Oh, 1987, "Developing Buyer-Seller Relationships," Journal of Marketing, 51, April, 11-27.

Fishbein, M.A., and I. Ajzen, 1975, Belief, Attitude, Intention and Behaviour: An Introduction to Theory and Research, Reading, MA, Addison Wesley.

Frisou, J., 2004, Le controle du comportement de fidelite : premiers jalons pour une theorie dynamique et eclectique, These de doctorat en sciences de gestion, Universite de Pau et des Pays de l'Adour.

Ganesan, S., and R. Hess, 1997, "Dimensions and Levels of Trust: Implications for Commitment to a Relationship," Marketing Letters, vol. 8, no. 4, 439-448.

Gefen, D., and D. Straub, 2003, "Managing User Trust in B to C e-Services", e-Service Journals, vol. 1, no. 1, 7-24.

Gravitz, M., 1988, Lexique des sciences sociales, Paris: Dalloz.

Guillon, B., 2006, "La Generalisation De La Responsabilite Environnementale Au Niveau Internationale : Le Cas D'un Groupe Specialise Dans Le Tourisme", dans Responsabilite Sociale De L'entreprise Et Changement De Contrat Social, dirige par JJ Rose, Coll. Methode et Recherche en Management, De Boeck, octobre.

Gurviez, P., and M. Korchia, 2002, "Proposition D'une Echelle De Mesure Multidimensionnelle De La Confiance Dans La Marque", Recherche et Applications en Marketing, vol. 17, no. 3, 41-61.

Hunt, S.D., and R.M. Morgan, 1995, "The Comparative Advantage Theory of Competition", Journal of Marketing, 59, April, 1-15.

Jacoby, J., and D.B. Keyner, 1973, "Brand Loyalty Versus Repeat Purchasing Behaviour", Journal of Marketing Research, 10, 1-9.

Jarvenpaa, S.L., and N. Tractinsky, 1999, "Consumer Trust in an Internet Store: A Cross-Cultural Validation", Journal of Computer Mediated Communication, vol. 5, no. 2, 34-49.

Lee, J., J. Lee, and L. Feick, 2001, "The Impact of Switching Costs on the Customer Satisfaction-Loyalty Link: Mobile Phone Service in France", The Journal of Services Marketing, vol. 15, no. 1, 35-48.

Mc Knight, H.D., V. Choudhury, and C. Kacmar, 2002, "The Impact of Initial Consumer Trust on Intentions to Transact with a Web Site: A Trust Building Model", Strategic Information Systems, vol. 11, 297-323.

Meyer-Waarden, L., 2002, "Les sources d'efficacite des programmes de fidelisation Une etude empirique sur la base d'un panel single source", These de doctorat soutenue le 09 decembre, Universite de Pau et des Pays de l'Adour--IAE.

Meyer-Waarden, L., and C. Benavent, 2006, "La theorie de la motivation intrinseque et extrinseque appliquee a la problematique des gratifications des programmes de fidelisation", Actes du 22eme Congres de I'Association Frangaise du Marketing, 11 et 12 Mai 2006, Nantes.

Nguyen, N, 1991, "Un modele explicatif de revaluation de la qualite d'un service : une etude empirique", Recherche et Applications en Marketing, vol. 6, no. 2, 83-98.

Oliver, R.L., 1999, "Whence Consumer Loyalty?", Journal of Marketing, 63, Special Issue, 33-44.

Price, L.L., and E.J. Arnould, 1999, "Commercial Friendships: Service Provider-Client Relationships In Context", Journal of Marketing, 63, October, 38-56.

Rylander, D., D. Strutton, and L.E. Pelton, 1997, "Toward a Synthesized Framework of Relational Commitment: Implications for Marketing Channel Theory and Practice", Journal of Marketing Theory and Practice, vol. 5, no. 2, 58-71.

Sirdeshmukh, D., J. Singh, and B. Sabol, 2002, "Consumer Trust, Value and Loyalty in Relational Exchanges", Journal of Marketing, 66, January, 15-37.

ENDNOTES

(1.) De Wulf., Oderkerken-Schroder & Iacobucci (2001), Garbarino & Johnson (1999).

(2.) Selon Gravitz (1988), a conceptual model is "an ideal representation which seeks to explain a phenomenon".

(3.) http://fr.wikipedia.org/wiki/Entretien_semi_directif

Lubica Hikkerova (a)

(a) University of Banska Bystrica, Slovakia [email protected]

Table 1

Reduced structural model

Variables Weighting t-value

Economic Value [right arrow] Commitment 0.09 3.77 **
Communication [right arrow] Commitment 0.17 2.86 **
Shared Values [right arrow] Commitment t 0.10 7.44 **
Reputation[right arrow] Trust 0.10 7.99 **
Communication [right arrow] Trust 0.12 7.11 **
Shared Values [right arrow] Trust 0.1 4.53 **
Trust [right arrow] Commitment 0.22 6.30 **
Trust [right arrow] Loyalty 0.25 7.82 **
Commitment [right arrow] Loyalty 0.17 12.38 **
Comparative Fit Index 0.91
Normative Fit Index 0.88
Goodness of Fit Index 0.84
Adjusted Goodness of Fit Index 0.81
Root Mean Squared Error of Approximation 0.059

** Significant at a risk of 5%

Table 2

Typology of loyalty

 Behavioural Loyalty

 Strong Weak

Attitudinal Strong True Loyalty Latent Loyalty
Loyalty Weak False Loyalty (SK) No Loyalty

Source: Frissou (2004)