The Emerald Research Register for this journal is available at www.emeraldinsight.

com/researchregister

The current issue and full text archive of this journal is available at www.emeraldinsight.com/0309-0566.htm

EJM 39,7/8

910
Received March 2004 Revised July 2004

The analysis of antecedents of customer loyalty in the Turkish mobile telecommunication market
zer khan O Serkan Aydin and Go
Gebze Institute of Technology, Cayirova Kampusu, Gebze Kocaeli, Turkey
Abstract
Purpose Corporate image, perceived service quality, trust and customer switching costs are the major antecedents of customer loyalty, and loyal customers may buy more, accept higher prices and have a positive word-of-mouth effect. Also we know that the cost of selling to new customers is much higher than the cost of selling to existing customers. Although this fact is apparent to everyone, many companies are still losing customers at a formidable rate. In this context the main aim of this paper is to examine the relationships between these factors and customer loyalty, and the relationships among these factors in the Turkish GSM sector. Design/methodology/approach Data was obtained from 1,662 mobile phone users in Turkey via questionnaire. The data was analyzed by structural equation modeling (SEM) in order to test all the relationships between variables in the model. Findings The ndings supported the proposed hypotheses, which are consistent with the theoretical framework. Analysis results showed that perceived service quality is a necessary but not sufcient condition for customer loyalty. Research limitations/implications In order to generalize the ndings from the paper, the model should be studied in different sectors. The contribution of this paper is to model all the relationships between customer loyalty and its antecedents, and to test these relationships simultaneously. Practical implications In order to better understand customer loyalty, as well as perceived service quality, corporate image, perceived switching costs and trust should be taken into consideration. Lately, technological change has shifted competition in the GSM sector from price and core service to value-added services. Therefore, operators should differentiate their services and guarantee their services quality because of this shift in competition. Originality/value In this paper, the effects of all the factors on customer loyalty are analyzed simultaneously via SEM. Keywords Customer loyalty, Telecommunications, Customer services quality, Corporate image, Trust, Turkey Paper type Research paper

European Journal of Marketing Vol. 39 No. 7/8, 2005 pp. 910-925 q Emerald Group Publishing Limited 0309-0566 DOI 10.1108/03090560510601833

Introduction A critical issue for the continued success of a rm is its capability to retain its current customers and make them loyal to its brands (Dekimpe et al., 1997, p. 405). Loyal customers build businesses by buying more, paying premium prices, and providing new referrals through positive word of mouth over time (Ganesh et al., 2000, p. 65). In fact, companies in telecommunications are losing 2-4 percent of their customers monthly; disloyal customers can amount to millions of lost revenue and prot. For example:

20% of customers of the mobile phone operator Orange defect each year and, on average it cost Orange 256 in 1996 to recruit each new customer, reecting the cost of introductory offers, subsidized phones and advertising. With almost a million customers, therefore, reducing the churn rate from 20% to 10% would bring about annual savings of over 25 million (Palmer, 1998, p. 117).

Antecedents of customer loyalty

In the same way, studies conducted in the nancial services industry show that increasing customer retention (or customer loyalty) by 5 percent could lead to 25-75 percent prot growth (Chan et al., 2001, p. 5):
Especially in telecommunications services, it is frequently pointed out that once customers have been acquired and connected to the telecommunications network of a particular operator; their long-term relations with the focal operator are of greater importance to the success of the company in competitive markets than they are in other industry sectors (Gerpott et al., 2001, p. 249).

911

Oliver (1997, p. 392) denes customer loyalty as:

. . . a deeply held commitment to rebuy or repatronize a preferred product/service consistently in the future, thereby causing repetitive same-brand or same brand-set purchasing, despite situational inuences and marketing efforts having the potential to cause switching behavior.

Although there are so many different denitions about customer loyalty, there seems to be two basic approaches stochastic and deterministic loyalty (Odin et al., 2001). The stochastic approach assumes customer loyalty as a behavior (Ehrenberg, 1988). In this approach it is considered that the preference structure of the customer is reected in the customers behavior. Some of the operational measures of this approach are shares of purchase, purchasing frequency, etc. On the other hand, the deterministic approach assumes customer loyalty as an attitude (Fournier and Yao, 1997). It is considered that merely describing the actual behavior of the customer does not sufce, but a proper analysis and description are clearly required for underlying attitudes structure of the customer. Some of the operational measures in deterministic approach are preference, buying intention, supplier prioritization and recommendation willingness. Jacoby and Kyner (1973, p. 2) express customer loyalty by a set of six necessary and collectively sufcient conditions by integrating two approaches:
[T]hese conditions express that brand loyalty is (1) the biased (i.e., random), (2) behavioral response (i.e., purchase), (3) expressed over time, (4) by some decision-making unit, (5) with respect to one or more alternative brands out of a set of such brands, and (6) is a function of psychological (decision making, evaluative) process.

Although there are many antecedent factors affecting brand loyalty, it is impossible to nd a study that has examined the effects of all factors simultaneously and jointly. Therefore, the main objective of this study is to determine the potential antecedents of customer loyalty and to test the relationships among these factors. It is known that corporate image (Nguyen and Leblanc, 2001), customer switching costs (Jones et al., 2002), trust (Lau and Lee, 1999) and service quality (Bolton and Drew, 1991) have signicant effects on customer loyalty. In this context, the relations among customer loyalty, trust, customer switching costs and service quality are analyzed by using structural equation modeling.

EJM 39,7/8

This study designed is as follows. First we present an overview of all of the antecedent factors of customer loyalty and the relationships. Second, the hypotheses and the structural model are proposed. Then the research methodology is described, and nally the ndings are discussed from the point of view of theoretical and managerial implications and future research. Theoretical background Service quality Even though there is no consensus about conceptualizing and measuring service quality (Carman, 1990), in this study we assume service quality to be the consumers judgment about the overall excellence or superiority of a service (Zeithaml, 1988). In order to have a better understanding about service quality, we should understand general attributes of services rst. The attributes reveal that: . services are intangible; . services are heterogeneous, meaning that their performance often varies with respect to the provider and the customer; . services cannot be placed in a time capsule and thus be tested and re-tested over time; and . the production of services is likely to be inseparable from their consumption (Gronroos, 1990). Because of the attributes of services, the evaluation of service quality is more difcult than the evaluation of product quality. Also, the evaluation may be connected with the service delivery process, along with output (Cody and Hope, 1999). In general, service quality is seen as a critical factor for protability, and thereby a rms success. Two underlying processes generally explain the contribution of service quality to protability. First, service quality is regarded as one of the few means for service differentiation and competitive advantage that attracts new customers and contributes to the market share (Venetis and Ghauri, 2000, p. 215). Second, service quality enhances customers inclination to buy again, to buy more, to buy other services, to become less price-sensitive and to tell others about their favorable experiences (Venetis and Ghauri, 2000, p. 215). For example, Bloemer et al. (1998) and Jones et al. (2002), among others, have pointed out that there is a positive relationship between service quality and repurchase intention, recommendation, and resistance to better alternatives. All these repurchase intention, recommendation and resistance to better alternatives are behavioral intentions and constitute customer loyalty. Therefore, it is proposed that service quality has a positive effect on customer loyalty. Trust Anderson and Narus (1990) emphasize that trust occurs when one party believes that the other partys actions would result in positive outcomes for itself. Consequently, in order to trust a brand, customers should perceive quality as being positive. Trust has been recognized as an important role in affecting relationship commitment (Morgan and Hunt, 1994) and so customer loyalty (Gundlach and Murphy, 1993). It appears that if one party trusts another, it is likely to develop some form of positive behavioral intention towards the other party. Accordingly, when a

912

customer trusts a brand, this means that he is also likely to form a positive buying intention towards the brand (Lau and Lee, 1999). In this context, trust works at preserving relationship investments by cooperating with exchange partners, resists attractive short-term alternatives in favor of the expected long-term benets of staying with existing partners, and views potentially high-risk actions as being prudent because of the belief that their partners will not act opportunistically (Morgan and Hunt, 1994). Hence, it is claimed that there is a positive relationship between trust in a rm and customer loyalty, consistent with the past research (e.g. Chaudhuri and Holbrook, 2001; Lau and Lee, 1999). Doney and Cannon (1997) suggest that the construction of trust involves a calculative process based on the ability of a party to continue to meet its obligations and on an estimation of the costs versus rewards of staying in the relationship. Therefore, to trust a brand, customers should not only perceive positive outcomes but also believe that these positive outcomes will continue in the future. Consequently, service quality should positively affect trust in the operator. Trust also reects credibility (Ganesan, 1994, p. 3), and credibility affects the long-term orientation of a customer by reducing the perception of risk associated with opportunistic behaviors by the rm (Erdem et al., 2002; Ganesan, 1994). Specically, trust reduces uncertainty in an environment in which consumers feel vulnerable, since they know that they can rely on the trusted brand (Chaudhuri and Holbrook, 2001, p. 82). That the customer trusts the operator in the GSM sector reduces uncertainty associated with services which were delivered previously or which are yet to be delivered. Reducing the uncertainty associated with the operator increases relatively the uncertainty of alternatives and thereby the perceived switching cost. Corporate image Corporate image is described as the overall impression made on the minds of the public about a rm (Barich and Kotler, 1991). Nguyen and Leblanc (2001, p. 228) claim that corporate image is related to the physical and behavioral attributes of the rm, such as business name, architecture, variety of products/services, and to the impression of quality communicated by each person interacting with the rms clients. Corporate image is the result of a process (MacInnis and Price, 1987). The process stems from ideas, feelings and consumption experiences with a rm that are retrieved from memory and transformed into mental images (Yuille and Catchpole, 1977). Therefore, corporate image is the result of an evaluation process. Although a customer may not have enough information about a rm, information obtained from different sources such as advertisements and word of mouth will inuence the process of forming the corporate image. Fishbein and Ajzen (1975) argue that attitudes are functionally related to behavioral intentions, which predict behavior. Consequently, corporate image as an attitude must affect behavioral intentions such as customer loyalty (Johnson et al., 2001, p. 224). Nguyen and Leblanc (2001) demonstrate that corporate image relates positively with customer loyalty in three sectors (telecommunication, retailing and education). The same relationship is demonstrated by Kristensen et al. (2000) for Danish postal services, and by Juhl et al. (2002) for the Danish food retailing sector. As pointed out above, corporate image stems from all of a consumers consumption experiences, and service quality is a function of these consumption experiences. Hence,

Antecedents of customer loyalty

913

EJM 39,7/8

customer perception about service quality directly affects the perception of corporate image. Switching costs Porter (1998, p. 10) denes switching costs as one-time costs facing the buyer when switching from one suppliers product to anothers. In addition to objectively measurable monetary costs, switching costs may also pertain to the time and psychological effort involved in facing the uncertainty of dealing with a new service provider (Bloemer et al., 1998, Klemperer, 1987). Hence, switching costs are partly consumer-specic (Shy, 2002). For this reason, a switching cost can be seen as a cost that deters customers from demanding a rival rms brand. Jackson (1985) describes the switching cost as the sum of economic, psychological and physical costs. The economic or nancial switching cost is a sunk cost which appears when the customer changes his/her brand, for example the costs of closing an account with one bank and opening another with a competitor, the cost of changing ones long-distance telephone service (Klemperer, 1987) or the costs of changing ones GSM operator. Procedural switching costs stem from the process of customers purchase decision making and their implementation of the decision. The buying process (Etzel et al., 1997) contains the following phases: (1) need recognition; (2) information search; (3) evaluation of alternatives; (4) purchase decision; and (5) post-purchase behavior. For example, if a consumer wishes to change their operator, they should evaluate alternative operators with regard to different criteria, such as coverage area, billing, customer service, value-added service, etc., complete the procedure for purchasing a new GSM line, and nally inform people of the new GSM number. Psychological cost is perceived as the cost stemming from social bonds (e.g. staff-customer relations, etc.) that appear over the course of time and the uncertainty/risk of the unused brand. The customer perceives high risk regarding a brand he/she has never used (Sharma and Patterson, 2000). Especially in services, where customers prefer a rival service provider, risk exists because service quality cannot be evaluated before purchasing (Sharma et al., 1997). A customer who has collected information in order to decrease their anxiety about a wrong purchasing decision will use all previous purchase experience. This is called post-purchase cognitive dissonance (Etzel et al., 1997). In this process, if the customer were to switch brand, he/she would compare the switched brand and the previous brand. Therefore, the better the switched brands performance, the higher the alternatives uncertainty. Hence, customers who want to decrease cognitive dissonance prefer brands that they have used before (Klemperer, 1995). Markets with switching costs are generally characterized by consumer lock-in, where it is observed that consumers repeatedly purchase the same brand even after competing brands become cheaper. One important consequence of having consumer lock-in is the ability of rms to charge prices above marginal costs (Shy, 2002, pp. 71-2).

914

In the case of a market having switching costs, when customers select from a number of functionally identical brands, they display brand loyalty and go on buying the same brand (Klemperer, 1987). In short, ex ante homogeneous products may, after purchase, be differentiated ex post by switching costs (Klemperer, 1987). Moreover, if customers are sensitive to a products attributes, such as quality, uncertainty will decrease price sensitivity (Erdem et al., 2002): in other words, the customer behaves loyally. For these reasons, switching costs are a factor that directly inuences customers sensitivity to price level, and so inuences customer loyalty (Jones et al., 2002; Bloemer et al., 1998; Burnham et al., 2003; Lee et al., 2001). Hypotheses and structural model Based on the results of earlier studies discussed in the previous section, we formulated the hypotheses detailed below. Figure 1 presents a proposed model, representing all of the hypotheses. Arrows in Figure 1 indicate causal directions. H1. There will be a positive relationship between perceived service quality and customer loyalty. H2. There will be a positive relationship between perceived service quality and trust in the operator. H3. There will be a positive relationship between trust in the operator and perceived switching cost. H4. There will be a positive relationship between trust in the operator and customer loyalty. H5. There will be a positive relationship between corporate image and customer loyalty. H6. There will be a positive relationship between perceived service quality and corporate image. H7. There will be a positive relationship between perceived switching cost and customer loyalty.

Antecedents of customer loyalty

915

Figure 1. The proposed model

EJM 39,7/8

H8. There will be a positive relationship between perceived service quality and perceived switching cost. Methodology Measures All of the constructs in the model were measured using a multiple-item measurement scale. All measures used a ve-point Likert-type response format, with strongly disagree and strongly agree as the anchors. A list of measurement items was developed using input from the review of the literature related to our study. All of the measures were translated from English into Turkish. To ensure content validity, the measures translated into Turkish were assessed by three academics so that respondents would understand the questions correctly. Then the measures were translated into English by a different academic, and the original measures were compared with these measures. The items can be seen in Table I. To measure customer loyalty (CL), the ve-item scale developed by Narayandas (1996) was adapted to the Turkish GSM sector. Accordingly, the operational measures in measuring customer loyalty are: . repurchase intention (next-use); . resistance to switching to competitors product that is superior to the preferred vendors product; and . willingness to recommend preferred vendors product to friends and associates. Perceived switching cost (SC) was assessed by seven items adapted from Burnham et al. (2003), Guiltinan (1989) and Jones et al. (2002). Operational measures used in the scale are: . perceived monetary costs; . perceived uncertainty costs; . perceived evaluation costs; . perceived learning costs; and . perceived set-up costs. Regarding the measure of trust (TR), a ve-item scale was developed by using different but complementary denitions. The operational measures used in measuring trust were: . reliability; . ethics; . service quality; and . cumulative process. In measuring perceived service quality, instead of the 22-item SERVQUAL instrument, a unidimensional measure (ve-item scale) of perceived service quality relating to an evaluation of all the base services was used for reasons of data collection efciency (Bloemer et al., 1998, p. 441). In the GSM sector, the base services are coverage of calling area, value-added services, customer support services, the suppliers services of the operator, and services in campaigns.

916

Construct Item Switching costs: GFI 0:94, Cronbachs a 0:674, Bentler-Bonett NFI 0:83 1. Switching to a new operator causes monetary cost 2. If I switched to a new operator, the service offered by the new operator might not work as well as expected 3. I am not sure that the billing of a new operator would be better for me 4. To switch to a new operator; I should compare all operators (on account of services, coverage area, billing, etc.) 5. Even if I have enough information, comparing the operators with each other takes a lot of energy, time and effort 6. If I switched to a new operator, I could not use some services (MMS, GPRS, WAP, etc.), until I learned to use them 7. I would be concerned about the people who would dial my previous number and could not reach me Customer 1. 2. 3. 4. 5. Corporate 1. 2. 3. 4. 5. loyalty: GFI 0:97, Cronbachs a 0:824, Bentler-Bonett NFI 0:97 I will go on using this GSM line If I bought a new GSM line, I would prefer this GSM operator I recommend this operator to people I encourage friends who plan to buy a GSM line Even if the other operators billing was cheaper, I would go on using this GSM line image: GFI 0:96, Cronbachs a 0:871, Bentler-Bonett NFI 0:97 This company is stable and rmly established This company is innovative and forward-looking This company has a social contribution for society This company is a leading rm in the Turkish GSM sector This company has a positive image

Antecedents of customer loyalty

917

Trust: GFI 0:94, Cronbachs a 0:856, Bentler-Bonett NFI 0:95 1. I trust this company 2. I feel that I can rely on this company to serve well 3. I trust the billing system 4. I believe that I can trust this company will not try to cheat me 5. This company is reliable because it is mainly concerned with the customers interests Service quality: GFI 0:90, Cronbachs a 0:827, Bentler-Bonett NFI 0:89 1. Given the quality of your operators coverage area, how would you rate the coverage area for your operator? 2. Given the quality of your operators customer services, how would you rate the customer services for your operator? 3. Given the quality of your operators adding service (GPRS, WAP etc.), how would you rate the adding service for your operator? 4. Given the quality of your operators vendor, how would you rate the vendor for your operator? 5. Given the quality of your operators campaign, how would you rate the campaign for your operator? 6. When you compare advertisements with services, how would you rate the services for your operator?

Table I. The items and unidimensionality, reliability, and convergent validity indices

EJM 39,7/8

Corporate image was measured by ve items developed by Bayol et al. (2001), which used the scale in analyzing the customer satisfaction index in GSM sector. Sample and data Data was collected from mobile phone users in the biggest cities in Turkey (Istanbul, Ankara, Izmit and Bursa). A total of 1,950 mobile phone users were contacted over a six-week period via a questionnaire. A number of questionnaires were eliminated by means of examining control questions in the questionnaire form. For this reason, the nal data set contained 1,662 GSM subscribers. The samples distribution on GSM operators was consistent with their real market share: . Aria, 10.1 percent; . Aycell, 7.8 percent; . Telsim, 32.2 percent; and . Turkcell, 50.4 percent. In the same way, consistent with market share, 43.6 percent of subscribers in the sample used a post-paid line, and 56.4 percent of subscribers used a pre-paid line. Sample characteristics appear to be representative of mobile phone users in Turkey. A total of 37 percent of the sample is female, the mean age of the sample is 29 years, the mean monthly mobile phone bill is 34.8 million Turkish lira, and the respondents monthly revenue is 884 million Turkish lira. Non-response bias To assess the possibility of non-response bias in the collected data set, we performed a chi-square difference test. To this end, we used the approximate percentages of each GSM operator in the market and the percentage of each GSM operator in our sample. Whether two distributions differed from each other was tested by chi-square test. From the result of the test (x2 3 0:1652; p , 0:01), we can state that there is no difference between the two distributions. Accordingly, non-response bias may not be a signicant problem. Measure validation Conrmatory factor analysis (CFA) was used for establishing the validity of the constructs (Ahire et al., 1996). Prior to the reliability analysis, the series mean was replaced instead of missing values in the data set. Unidimensionality is a necessary condition for reliability and construct validation (Mak and Sockel, 2001, p. 271). The unidimensionality of the constructs was analyzed reskog and by specifying a measurement model for each construct. According to Jo rbom (1993), a goodness of t index (GFI) of 0.90 or above suggests that each of the So constructs is unidimensional. As seen in Table I, the GFI value of the construct is 0.90 or above. Convergent validity is examined by using the Bentler-Bonett normed t index (NFI) (Bentler and Bonett, 1990). All of the constructs, except perceived switching cost and perceived service quality, have NFI values above 0.90. Perceived service quality has an NFI value of 0.83 and perceived service quality has an NFI value of 0.89. These values are close to 0.90. Therefore, convergent validity was achieved for all the constructs in the study.

918

In assessing discriminant validity, CFA was performed on a selected pair of scales, allowing correlation between two constructs. The analysis was rerun with the correlation between the two constructs xed at 1. If the correlation is a free parameter and not this xed constant, the x 2 of the initial model (where correlation is free) should be much smaller than the latter model (where it is xed at one). In addition, the difference between the chi-squares of these two models should be signicant when checked against the chi-square test statistic at p , 0:01 with degrees of freedom equal rsoy et al., 2004). All to the difference in degrees of freedom between the two models (Gu 2 ; 069:55; p , 0:01] of the chi-square difference tests [minimumx2 dif 1 demonstrated that discriminant validity has been achieved. For reliability, the items were submitted to reliability analysis via Cronbachs alpha. Reliability analysis of ve factors is shown in Table I. The reliability values of all the factors were either close to or greater than 0.70, the threshold Nunnally (1978) recommended for research. The reliability and validity analysis results indicate that the scales for the constructs appear to have satisfactory measurement qualities. Tests of hypotheses Structural equation modeling (SEM) was used to test the hypothesized relationships in the proposed model shown in Figure 1. The structural equation modeling technique enables the simultaneous estimation of multiple regression equations in a single framework. Notably, all direct and indirect relationships in the model are estimated simultaneously, and thus the method allows all the interrelationships among the variables to be assessed in the same decision context (Oh, 1999, p. 74). The proposed model was analyzed via the maximum likelihood estimator of LISREL 8.3 by using the covariance matrix of the measured variables as input. Table II reports goodness of t indices, standardized parameter estimates and their t-values for the structural model. The overall chi-square statistic is signicant (x2 336 2; 767:69; p , 0:01), which is expected given the large sample size (Bagozzi and Yi, 1988). All other goodness of t indices are within the acceptable ranges (GFI 0:89, AGFI 0:87, SRMR 0:049, CFI 0:92, RMSEA 0:066). All of the t indices indicate that the proposed model exhibits a reasonably good t to the data. In accordance with the parameter estimates shown in Table II, perceived service quality is positively and signicantly related to customer loyalty (H1: g1 0:14, p , 0:01). In the same way, as proposed in H4, trust has a positive and signicant effect on customer loyalty (H4: b2 0:59, p , 0:01), and as proposed in H7, perceived switching costs positively and signicantly affect customer loyalty (H7: b4 0:14, p , 0:01). However, although corporate image affects positively customer loyalty (H5: b3 0:07), this effect is not statistically signicant. Hence, H5 is rejected. The analysis results also yield that perceived service quality relates positively and signicantly with trust (H2: g2 0:60, p , 0:01). This nding supports H2. Moreover, the path from perceived service quality to perceived switching cost is signicant (H8: g4 0:09, p , 0:05). Referring to Table II, the ndings indicate that trust in the operator relates positively and signicantly with perceived switching cost (H3: b1 0:30, p , 0:01). This result supports H3. In the same way, since the positive relationship between perceived service quality and corporate image (H6: g6 0:62, p , 0:01) is signicant at p , 0:01, H6 is supported.

Antecedents of customer loyalty

919

EJM 39,7/8

Path Perceived service quality ! customer loyalty Perceived service quality ! trust Trust ! perceived switching cost Trust ! customer loyalty Corporate image ! customer loyalty Perceived service quality ! corporate image Perceived switching cost ! customer loyalty Perceived service quality ! perceived switching cost

Hypothesis H1 H2 H3 H4 H5 H6 H7 H8

Estimate 0.14 0.60 0.30 0.59 0.07 0.62 0.14 0.09

t-Value 5.02* 19.96* 7.44* 13.87* 1.83 20.64* 5.88* 2.39**

920

x2 336 2; 767:69 GFI 0:89 AGFI 0:87 SRMR 0:049 CFI 0:92 RMSEA 0:066 SMCcustomer loyalty 0:63 SMCtrust 0:36 SMCcorporate image 0:38 SMCperceived switching cost 0:13
Table II. Summary of results Notes: CFI comparative t index; GFI goodness of t index; AGFI adjusted goodness of t index; SRMR standardized root mean square residual; RMSEA root mean square error of approximation; SMC squared multiple correlation; *p , 0:01 (one-tailed test); **p , 0:05

Table II indicates that the model shows a fairly high level of explanatory power for four endogeneous constructs, with customer loyalty, trust, corporate image and perceived switching cost having SMC values of 0.63, 0.36, 0.38 and 0.13, respectively. Discussion In this study, the relationship between customer loyalty and perceived service quality, perceived switching cost, trust and corporate image is investigated. To this end, the data was analyzed by path analysis. The results of the path analysis show that all of the factors have positive effects on customer loyalty. The analysis results reveal that trust is the most important determinant of customer loyalty. Even though perceived service quality and perceived switching cost seemed to have the same level of effect on consumer loyalty, the switching cost should be considered as a more important factor due to its indirect effect. For example, the indirect effect of perceived service quality on customer loyalty stems from the positive relations between perceived service quality and such factors as trust, corporate image and perceived switching cost. Since these factors directly affect customer loyalty, we found the indirect effect of perceived service quality to be 0.43. In the same way, the indirect effect of trust on customer loyalty is based on the direct effect of perceived switching cost, and its indirect effect was measured as 0.04. As a result, the total effect of trust on customer loyalty is 0.63. This indirect effect is based on the path from trust to perceived switching cost and the path from perceived switching cost to customer loyalty. As can be seen in Figure 1, perceived switching costs and corporate image have no indirect effect on customer loyalty.

Although it is not signicant, the analysis results indicate that corporate image affects positively customer loyalty. The general experience of marketing professionals also veries that such an effect is valid. The situation can be explained by the general characteristics of Turkish mobile phone market, since the purchasing power of subscribers is less than in developed countries, and the indirect tax rate in the GSM sector is relatively higher in Turkey. Consequently, the most important factor for the decision making process of subscribers is services provided by operator compared to cost. The operators image has no signicant effect on consumer loyalty in the Turkish GSM sector. Perceived switching cost is inuenced by trust and perceived service quality, whereas perceived service quality affects both perceived switching cost and trust. For this reason, perceived service quality relates both directly (0.09) and indirectly (0.18) to perceived switching cost. As regards trust, it has a direct effect (0.30) on perceived switching cost. These ndings indicate that trust is a more crucial factor for perceived switching costs than perceived service quality. Generally, as discussed above, the ndings show that perceived service quality is a necessary but not sufcient condition for customer loyalty to emerge and to exist. In addition, consistent with prior research, corporate image, perceived switching costs and especially trust should be taken into consideration in developing strategies for developing customer loyalty. Essentially, the SMC value of customer loyalty in this study is higher than that in der, 2003; Bloemer et al., 1998; prior research (e.g. DeWulf and Odeberken-Schro Chaudhuri and Holbrook, 2001; Nguyen and Leblanc, 2001; Sharma and Patterson, 2000; Lee et al., 2001, etc.), since our model includes all these factors effects simultaneously and jointly. Therefore, our proposed model has a great explanatory power in comparison to prior research. Managerial implications In the Turkish GSM market, there are four operator rms. Two of the operators (Aria and Aycell) are new entrants. On the other hand, it may be concluded from the fall in the markets growth rate that the GSM market has reached maturity in Turkey. Acquiring new customers is both costly and difcult in terms of marketing for GSM operators when the number of subscribers has reached its peak level. Hence, it is becoming an industry-wide belief that the best core marketing strategy for the future is to try to retain existing customers by heightening customer loyalty and customer value (Kim et al., 2004, p. 146). To this end, operators should decrease their subscribers sensitivity to price. It can be said that factors such as trust, perceived service quality, perceived switching cost and positive corporate image are very important for GSM operators to establish a loyal customer base and decrease their sensitivity to price. On the other hand:
. . . bolstered by the rapid development of information and communication technologies (ICT) and high demand from customers, the paradigm of mobile telecommunication services is now shifting from voice-centered communication to a combination of high-speed data communication and multimedia (Kim et al., 2004, p. 145).

Antecedents of customer loyalty

921

This change shifts competition in GSM sector from price and core services (coverage of calling area and clarity of sound) to value-added services. For this, the operators should

EJM 39,7/8

922

differentiate their services and guarantee quality of their services in order to maintain their market share. Due to technological changes and differentiation strategies of rms, services become more and more complex than ever. As the services provided become more complex, the learning period for service use becomes longer and the evaluation process becomes more difcult for customers. Differentiation will increase perceived switching nikasyon Kurumu), cost. Accordingly, despite the attempts of the regulator (Telekomu which focus on minimizing the deterrent to switch operator (switching cost), perceived switching cost may be seen as a crucial factor for customer loyalty. In addition, differentiating services and the superior quality standards of these services will not only increase trust in the operator, but also enable the formation of a positive corporate image in subscribers minds. In many European countries new users are tied to contracts extending over a certain period of time, during which the subscriber has to maintain the service and pay a pre-agreed xed fee. This contractual bond between the operator and subscribers may cause switching costs. In Turkey, however, new users are tied to contracts extending over uncertain periods of time. Therefore, operators in Turkey may offer subscriptions with denite periods of time in order to increase switching costs.

Limitations and future research directions This study has some limitations. First, the perceived switching cost was assessed by a seven-item scale. The switching cost consists of different sub dimensions, such as psychological, nancial and procedural. Therefore, these sub dimensions should be measured and their effects should be examined. This approach may provide both empirical and theoretical information. Second, rival operators attempts to attract new subscribers (such as lower price level, phone lines without fee, etc.) affect the preference of subscribers. Therefore, subscribers perceptions about these attempts should be assessed, and the direct and indirect effect of these perceptions on customer loyalty and on its antecedents should be studied. Consequently, in order to generalize the ndings in the study, mobile telecommunication services should be compared with other industries. Finally, Dick and Basu (1994) conceptualize customer loyalty as the strength of the relationship between customers relative attitude towards an entity (brand, rm, store, etc.) and repeat patronage (behavior). According to this approach, customers are divided into four segments using two levels of behavioral loyalty and two levels of attitude toward the brand. These groups are true loyal (high behavioral loyalty and high relative attitude), hidden loyal (low behavioral loyalty and high relative attitude), spurious loyal (high behavioral loyalty and low relative attitude) and no loyal (low behavioral loyalty and low relative attitude). With respect to this typology, even if customers may tend to be attitude loyal because of high perceived service quality, perceived switching cost and trust, they may not behave loyally, or although they have low relative attitude loyalty because of low perceived service quality, perceived switching cost and trust, they may behave loyally. Therefore, the effect of perceived service quality, switching costs, trust, corporate image and other possible situational factors (e.g. attractiveness of alternatives, lack of rival brands, etc.) on both behavioral loyalty and attitude loyalty should be studied in future research.

References Ahire, S.L., Golhar, D.Y. and Waller, M.A. (1996), Development and validation of TQM implementation constructs, Decision Sciences, Vol. 27 No. 1, pp. 23-56. Anderson, J.C. and Narus, J.A. (1990), A model of distributor rm and manufacturer rm working partnerships, Journal of Marketing, Vol. 54, January, pp. 42-58. Bagozzi, R.P. and Yi, Y. (1988), On the evaluation of structural equation models, Journal of the Academy of Marketing Science, Vol. 16, Spring, pp. 77-94. Barich, H. and Kotler, P. (1991), A framework for marketing image management, Sloan Management Review, Vol. 32 No. 2, pp. 94-104. Bayol, M.P., LaFoye, A., Tellier, C. and Tenenhaus, M. (2001), Use of PLS path modelling to estimate the European Consumer Satisfaction Index (ECSI) model, available at: www.stat. ucl.ac.be/ISarchives/ Bentler, P.M. and Bonett, D.G. (1990), Comparative t indices in structural models, Psychological Bulletin, Vol. 107 No. 2, pp. 238-46. Bloemer, J., Ruyter, K. and Wetzels, M. (1998), On the relationship between perceived service quality, service loyalty and switching costs, International Journal of Industry Management, Vol. 9 No. 5, pp. 436-53. Bolton, R.N. and Drew, J.H. (1991), A multistage model of customers assesment of service quality and value, Journal of Consumer Research, Vol. 17 No. 4, pp. 365-84. Burnham, T.A., Frels, J.K. and Mahajan, V. (2003), Consumer switching costs: a typology, antecedents and consequences, Journal of The Academy of Marketing Science, Vol. 31 No. 2, pp. 109-26. Carman, J.M. (1990), Consumer perceptions of service quality: an assesment of the SERVQUAL dimensions, Journal of Retailing, Vol. 66, Spring, pp. 33-55. Chan, M., Lau, L., Lui, T., Ng, S., Tam, E. and Tong, E. (2001), Final report: customer relationship management, Customer Relationship Management Consortium Study, Asian Benchmarking Clearing House, Hong Kong. Chaudhuri, A. and Holbrook, M.B. (2001), The chain effects from brand trust and brand affect to brand performance: the role of brand loyalty, Journal of Marketing, Vol. 65, April, pp. 31-93. Cody, K. and Hope, B. (1999), EX-SERVQUAL: an instrument to measure service quality of extranets, Proceedings of the 10th Australasian Conference on Information Systems, Wellington, 1-3 December, p. 207. Dekimpe, M.G., Steenkamp, J.-B.E.M., Mellens, M. and Abeele, P.V. (1997), Decline and variability in brand loyalty, International Journal of Research in Marketing, Vol. 14, pp. 405-20. der, G. (2003), Assesing the impact of a retailers relationship DeWulf, K. and Odeberken-Schro efforts on consumers attitudes and behavior, Journal of Retailing and Consumer Services, Vol. 10, pp. 95-108. Dick, A.S. and Basu, K. (1994), Customer loyalty: toward an integrated conceptual framework, Journal of the Acedemy of Marketing Science, Vol. 22, pp. 99-113. Doney, P.M. and Cannon, J.P. (1997), An examination of the nature of trust in buyer-seller relationships, Journal of Marketing, Vol. 61, April, pp. 35-51. Ehrenberg, A.S.C. (1988), Repeat Buying: Facts, Theory and Applications, Oxford University Press, Oxford. Erdem, T., Louviere, J. and Swait, J. (2002), The impact of brand credibility on consumer price sensitivity, International Journal of Research in Marketing, Vol. 19, pp. 1-19.

Antecedents of customer loyalty

923

EJM 39,7/8

924

Etzel, M.J., Stanton, W.J. and Walker, B.J. (1997), Marketing, McGraw-Hill, New York, NY. Fishbein, M. and Ajzen, I. (1975), Belief, Attiude, Intention, and Behavior: An Introducing to Theory and Research, Addison-Wesley, Reading, MA. Fournier, S. and Yao, J.L. (1997), Reviving brand loyalty: a conceptualization within the framework of consumer-brand relationships, International Journal of Research in Marketing, Vol. 14 No. 5, pp. 451-72. Ganesan, S. (1994), Determinants of long-term orientation in buyer-seller relationships, Journal of Marketing, Vol. 58, April, pp. 1-19. Ganesh, J., Arnold, M.J. and Reynolds, K.E. (2000), Understanding the customer base of service providers: an examination of the differences between switchers and stayers, Journal of Marketing, Vol. 64, July, pp. 65-87. Gerpott, T.J., Rams, W. and Schindler, A. (2001), Customer retention, loyalty and satisfaction in the German mobile cellular telecommunications market, Telecommunications Policy, Vol. 25, pp. 249-69. Gronroos, C. (1990), Service Management and Marketing: Managing the Moments of Truth in Service Competition, Lexington Books, Lexington, MA. Guiltinan, J.P. (1989), A classication of switching costs with implications for relationship marketing, in Childers, T.L., Bagozzi, R.P. et al. (Eds), AMA Winter Educators Conference: Marketing Theory and Practice, American Marketing Association, Chicago, IL, pp. 216-20. Gundlach, G.T. and Murphy, P.E. (1993), Ethical and legal foundations of relational marketing exchanges, Journal of Marketing, Vol. 57, October, pp. 35-46. rsoy, D., Kim, K. and Uysal, M. (2004), Perceived impacts of festivals and special events by Gu organizers: an extension and validation, Tourism Management, Vol. 25, pp. 171-81. Jackson, B.B. (1985), Winning and Keeping Industrial Customers, Lexington Books, Lexington, MA. Jacoby, J. and Kyner, D.B. (1973), Brand loyalty vs. repeat purchasing behavior, Journal of Marketing Research, Vol. 10, February, pp. 1-9. Johnson, M.D., Gustafsson, A., Andreassen, T.W., Lervik, L. and Cha, J. (2001), The evolution and future of national customer satisfaction index models, Journal of Economic Pcychology, Vol. 22, pp. 217-45. Jones, M.A., Beatty, S.E. and Mothersbaugh, D.V. (2002), Why customers stay: measuring the underlying dimensions of services switching costs and managing their differential strategic outcomes, Journal of Business Research, Vol. 55, pp. 441-50. reskog, K.G. and So rbom, D. (1993), LISREL 8: A Guide to the Program and Applications, Jo Scientic Software International, Chicago, IL. Juhl, H.J., Kristensen, K. and Ostergaard, P. (2002), Consumer satisfaction in European food retailing, Journal of Retailing and Consumer Services, Vol. 9 No. 6, pp. 327-34. Kim, M.K., Park, M.C. and Jeong, D.H. (2004), The effects of customer satisfaction and switching barrier on customer loyalty in Korean mobile telecommunication services, Telecommunications Policy, Vol. 28, pp. 145-59. Klemperer, P. (1987), Markets with consumer switching costs, The Quarterly Journal of Economics, Vol. 102, pp. 376-94. Klemperer, P. (1995), Competition when consumers have switching costs: an overview with applications to industrial organization, macroeconomics and international trade, Review of Economic Studies, Vol. 62, pp. 515-39.

Kristensen, K., Gronholdt, L. and Martensen, A. (2000), Customer satisfaction measurement at Post Denmark: results of application of the European Customer Satisfaction Index methodology, Total Quality Management, Vol. 11 No. 7, pp. 1007-15. Lau, G. and Lee, S. (1999), Consumers trust in a brand and link to brand loyalty, Journal of Market Focused Management, Vol. 4, pp. 341-70. Lee, J., Lee, J. and Feick, L. (2001), The impact of switching costs on the customer satisfaction-loyalty link: mobile phone service in France, Journal of Services Marketing, Vol. 15 No. 1, pp. 35-48. MacInnis, D.J. and Price, L.L. (1987), The role of imagery in information processing: review and extensions, Journal of Consumer Research, Vol. 13, pp. 473-91. Mak, B.L. and Sockel, H. (2001), A conrmatory factor analysis of IS employee motivation and retention, Information and Management, Vol. 38, pp. 265-76. Morgan, R.M. and Hunt, S.D. (1994), The commitment-trust theory of relationship marketing, Journal of Marketing, Vol. 58, July, pp. 20-38. Narayandas, N. (1996), The link between customer satisfaction and customer loyalty: an empirical investigation, working paper, 97-017, Harvard Business School, Boston, MA. Nguyen, N. and Leblanc, G. (2001), Corporate image and corporate reputation in customers retention decisions in services, Journal of Retailing and Consumer Services, Vol. 8, pp. 227-36. Nunnally, J.C. (1978), Psychometric Theory, McGraw-Hill, New York, NY. Odin, Y., Odin, N. and Florence, P.V. (2001), Conceptual and operational aspects of brand loyalty an empirical investigation, Journal of Business Research, Vol. 53, pp. 75-84. Oh, H. (1999), Service quality, customer satisfaction, and customer value: a holistic perspective, International Journal of Hospitality Management, Vol. 18, pp. 67-82. Oliver, R. (1997), Satisfaction: A Behavioral Perspective on the Consumer, McGraw-Hill, New York, NY. Palmer, A. (1998), Principles of Services Marketing, 2nd ed., McGraw-Hill, New York, NY. Porter, M. (1998), Competitive Strategy: Techniques for Analyzing Industries and Competitors, Free Press, New York, NY. Sharma, N. and Patterson, P.G. (2000), Switching costs, alternative attractiveness and experience as moderators of relationship commitment in professional consumer services, International Journal of Service Industry Management, Vol. 11 No. 5, pp. 470-90. Sharma, N., Patterson, P.G., Cicic, M. and Dawes, P. (1997), A contingency model of relationship commitment for professional consumer services, Proceedings of the 26th EMAC Conference, Warwick, May 20-23. Shy, O. (2002), A quick and easy method for estimating switching costs, International Journal of Industrial Organization, Vol. 20, pp. 71-87. Venetis, K.A. and Ghauri, P.N. (2000), The importance of service quality on customer retention: an empirical study of business service relationships, Proceedings of the Marketing in a Global Economy Conference, Buenos Aires, June 28-July 1, pp. 215-224. Yuille, J.C. and Catchpole, M.J. (1977), The role of imagery in models of cognition, Journal of Mental Imagery, Vol. 1, pp. 171-80. Zeithaml, V.A. (1988), Consumers perceptions of price, quality, and value: a means-end model and synthesis of evidence, Journal of Marketing, Vol. 52, July, pp. 2-22.

Antecedents of customer loyalty

925