Using visit frequency to segment ski resorts customers

Journal of Vacation Marketing Volume 12 Number 1 Using visit frequency to segment ski resorts customers Rodoula Tsiotsou Received (in revised form): April 2005 Anonymously refereed paper N. Plastira 57 Lykovrisi, TK 14123 Athens, Greece Tel. 0030 210 2849584; E-mail: rtsiotsou@yahoo.gr Rodoula Tsiotsou is currently Marketing Director for the daily sports newspaper Protathlitis and a visiting lecturer at the National & Kapodistriako University of Athens, Greece. She has worked as Marketing Manager for the Greek Professional Basketball League and she has taught courses in sport marketing and management at the Department of Physical Education of Democritus University, Greece. ABSTRACT KEYWORDS: experience, satisfaction, source of information, winter tourism, ski resorts, market segmentation The purpose of the study is to segment ski resorts customers according to their frequency of visits in order to identify homogeneous groups. In particular, the study investigates the degree to which ski experience, overall satisfaction and income discriminate customers who visit ski resorts weekly from those who visit them monthly. A self-administered anonymous questionnaire was given to 200 customers of two different ski resorts in Greece. The questionnaire was answered by 191 individuals (95.5 per cent response rate). Classification with discriminant analysis was used to identify consumers with different visitation frequency. The hypothesis that ski experience, overall satisfaction and income will classify consumers according to their visitation frequency was confirmed. The findings of the study provide theoretical and practical implications in identifying better segments and increasing destination marketing effectiveness. INTRODUCTION Tourism is one of the main industries in Greece that stimulates economic development in industries from hospitality, transport, construction and retail, to small businesses such as restaurants, bars and tourism agents. Tourism contributes to the Gross National Product by 22 per cent, more that any other sector in Greece (industry 12.4 per cent and agriculture 9 per cent). 1 Winter tourism has grown rapidly over the last two decades all over Europe. Studies have reported that a growing number of tourists visit ski resorts in European countries such as the UK, Bulgaria and Greece. 2 Particularly in Greece, more and more locals prefer to spend their weekends or winter vacations in ski resorts located in the country. Some places have become very popular and attract tourists in high number during winter especially at weekends. Currently, there are more than 19 ski resorts in Greece that constitute frequent destinations for thousands of Greek visitors every year. However, Greek ski resorts have not been a destination for foreign tourists. Because tourism is a very important sector in Greece and ski resorts are rapidly growing, it is imperative to study and understand this market and their customers. Studying ski resorts customers will provide tourism marketers with valuable information about their needs, wants and behavior in an effort to retain current and attract new Journal of Vacation Marketing Vol. 12 No. 1, 2006, pp. 15 26 & SAGE Publications London, Thousand Oaks, CA, and New Delhi. www.sagepublications.com DOI: 10.1177/1356766706059029 Page 15

Using visit frequency to segment ski resorts customers customers. Ski resorts need to segment their customers in order to offer products and offerings that will satisfy these customers and attract new ones. Consumer satisfaction, buying frequency, product experience and demographics such as income have been used in marketing to identify consumer groups. However, these variables have not been used together in an effort to find distinct tourist segments. Although activities have been used to segment tourists, activity experience has not been used before to identify homogeneous groups. Furthermore, it has been recommended that segmenting tourists based on frequency of [ski resort] use may be useful in identifying skilled consumers 3 (p. 33). Thus, studying ski resort visitors could provide us with valuable information about these customers. This study proposes the combination of demographic (income), behavioral (product experience, visit frequency), and satisfaction variables to segment ski resort customers. Segmentation using a combination of the above variables for the ski resorts market has not been proposed before in the tourism marketing literature. The purpose of the study was to segment ski resorts customers by using, as a segmentation base, visit frequency. In particular, the intention was to investigate the degree to which ski experience, overall satisfaction and income discriminate ski resorts customers with different visitation frequency in an effort to identify distinct customer groups and predict customers future visit frequency. The article is organized into five parts. First, the theoretical framework of the study is presented followed by the methodology used. Then, the results and a discussion on the results and implications are presented. The limitations of the study and future research recommendations conclude the study. The review of the literature on tourist segmentation, satisfaction and product experience provides the theoretical framework for developing the hypotheses of the study. THEORETICAL FRAMEWORK Tourist segmentation Market segmentation is one of the main strategies in marketing that assists in identifying homogeneous groups of consumers with similarities in an effort to satisfy their needs and increase marketing effectiveness. Similarities refer to needs, shopping habits, media usage, price sensitivity and other. The information gathered through market segmentation is crucial in the strategic marketing planning process of a company. Market segmentation benefits companies in four ways: (a) it provides the base for target marketing; (b) it assists in developing more effective marketing mixes in order to satisfy the needs of a specific segment; (c) it facilitates product differentiation; and (d) it provides easier identification of market opportunities and threats. The most often used bases for segmenting consumer markets are demographic (age, gender, family status, income), geographic, behavioral (benefits, frequency of use, loyalty), and psychographic segmentation (lifestyle, personality characteristics). Often a combination of different segmentation bases is used. 4 There are two types of segmentations, a priori and post hoc segmentation. A priori segmentation is when the variable used as a criterion to divide a market is known in advance whereas post hoc segmentation is when there is no knowledge about distinct groups and a set of variables is used as the base for segmentation. 5 Segmentation has often been used to identify distinct groups of tourists, because, like any other market, tourists do not respond homogeneously to marketing activities. Table 1 summarizes the results of several studies and the different variables used to segments tourists. An overview of Table 1 indicates that the most often used variables to segment tourists are demographics, socioeconomics, lifestyle and satisfaction whereas the most often statistical analysis used is cluster analysis (Kmeans). The number of segments identified a range from two to six. In terms of the role of income in tourism Page 16

Tsiotsou Table 1: Summary of segmentation studies in tourism City/ Country Segmentation variables Statistical analysis Tourism market No. of segments Authors 6 Greece Satisfaction Leisure activities Demographics Discriminant Denmark Psychographic Factor Cluster New Zealand Expenditures Austria South Africa Spain Scotland Virginia USA New Zealand Moravia Czech Repub. Guam Taiwan Balearic Islands United Kingdom Lifestyle Summer activities Demographics Socioeconomic Behavioral Psychographic Demographics Socioeconomic Geographic Emotions Satisfaction Loyalty Benefits Socioeconomic Behavioral Sentiments Demographics Trip characteristics Expenditures Lifestyle Demographics Destination choice Expenditures Demographics Trip purpose Travel behavior Adventure visitors 2 Tsiotsou & Vasioti (2006) Residents 4 Schmidt & Merser (2005) Cluster Residents 6 Lawson & Thyne (2004) Bagged International 3 5 Dolnicar & visitors Leisch (2004) Cluster Self-organizing neural networks Cluster (K-means) Cluster (K-means) X 2 Automatic Interaction Detection Cluster MANOVA Cluster One-way ANOVA Urban visitors Museum Theme park visitors Rural tourists 3-4 Bloom (2004) 2 Bigne & Andreu (2004) 4 Frochot (2005) Residents 4 Chen (2003) International tourists National International tourists National tourists Socio-demographics Chi Square International tourists Participation patterns Chi Square Ski resorts Destination customers determinants Socio-demographics Ski level 6 Becken (2003) 6 Tureckova (2002) 3 Mok & Iverson (2000) 2 Juaneda & Sastre (1999) 3 Richards, (1996) segmentation, the findings are contradictory. Lawson and Thyne 7 have found that income is significantly related to the proportion of money spent on sightseeing activities only, but not to other aspects of a trip such as accommodations, food and entertainment. Moreover, it has been found that some tourist segments differ in their income. Dolnicar and Leisch 8 have reported that health oriented holiday makers have the highest disposable income of the four segments they identified and they spend the highest amount Page 17

Using visit frequency to segment ski resorts customers of money per day and per person. However, Mok and Iverson 9 segmented Americans based on their expenditures and found that no significant difference exists between those segments in terms of their income. In the case of ski resorts, because such destinations are characterized by a high marginal cost (skiing is a relatively expensive sport), frequent visitors would have a high disposable income in order to take vacations there frequently. 10 Richards 11 studied the UK ski resorts customers and found that choice of destination was affected by the price of the product (skiing) and by income levels. Thus, income seems to be an important variable that needs to be included when segmenting ski resorts tourists. It is expected that: H1: Customers with higher income will visit ski resorts more often. Product ski experience Experience in marketing has been extensively studied in consumer knowledge, familiarity and expertise of products. In this study, experience is the ability to perform product-related tasks. 12 Product-related experience has been defined as memory for relationships between the self and the product in terms of information search, product usage, and purchase experience. 13 Experience has been utilized as a measure of consumer knowledge or familiarity. 14 Park et al. 15 have made and tested two propositions associated with product experience. The first proposition states that the amount of product-related experience is positively related to the amount of productclass information stored in memory. Moreover, personal experiences with products may increase the perceived validity and relevance of information. Thus, productrelated experiences are considered as valid cues in making judgments related to the products. The second proposition states that the amount of product-related experience is more strongly associated with the level of self-assessed knowledge than with objective knowledge. Self-assessed knowledge refers to consumers perceptions of what or how much they know about a product class, whereas objective knowledge refers to accurate information about the product class stored in long-term memory. The above two propositions were tested and confirmed by the results of their study. The findings suggested that product-related experience plays a more important role in consumers knowledge assessments than productinformation cues. This can be explained by higher accessibility of product-related experience in memory. Mano and Oliver 16 have proposed two dimensions of product experience, the utilitarian (instrumental) and the hedonic (aesthetic). The utilitarian experience refers to the usefulness function a product performs, whereas the hedonic dimension refers to intrinsically pleasing properties of a product. They found that utilitarian evaluation is more functional and cognitive than hedonic evaluation because it deals with the fulfilment of instrumental consumers expectations of the product. Hedonic evaluations are affective. This finding may have implications in areas that are considered hedonic such as leisure activities (e.g. skiing), and entertainment services. Prior experience with a product has been related to information processing, 17 product evaluation, 18 familiarity and expertise 19 and consumers goals. 20 Consumers with moderate knowledge and experience are expected to process information more than consumers with low or high knowledge and experience. In addition, consumers with the most experience use brand processing whereas consumers with less experience use attribute-based processing to a greater extent. 21 Huffman and Houston 22 related consumers goals to goal-directed experiences that assist in developing and organizing consumers knowledge. Their study indicated that the content of feedback and its consistency with consumers goals affect the goal orientation and organization of brand and feature knowledge acquired during choice experience. In tourism, prior experience has been Page 18

Tsiotsou found to be an important factor in activity participation, expenditure patterns 23 and destination choice, 24 whereas skill level (of an activity) is a good predictor of participation frequency and location. 25 In this study, product experience refers to the ski experience customers of ski resorts have, because skiing is the core product of ski resorts. Thus, based on the review of literature, it should be expected that: H2: Customers with more ski experience will visit ski resorts more often. Consumer satisfaction Satisfaction has been defined as a global evaluative judgment about product usage/ consumption. 26 A more recent attempt has conceptualized satisfaction as a summary affective response of varying intensity with a time specific point of determination and limited duration directed toward focal aspects of product acquisition and/or consumption. 27 Consumer satisfaction has been considered one of the most important constructs 28 and one of the main goals in marketing 29 because it is a key element for gaining competitive advantage. 30 Due to its importance, various theories and models have been developed in an effort to define the construct and explain satisfaction in different products/services and consumption stages. The expectancy disconfirmation paradigm, 31 the perceived performance model, 32 as well as attribution models, 33 affective models 34 and equity models, 35 are only some of the main theoretical bases developed to explain consumer satisfaction. The above approaches have raised several issues and debates among marketing scholars. Satisfaction plays a central role in marketing because it is a good predictor of purchase behavior (repurchase, purchase intentions, brand choice and switching behavior). 36 Satisfaction is an important predictor of customer loyalty 37 and the strength of the relationship between the two is strongly influenced by customer characteristics such as variety seeking, age and income. 38 Satisfied customers tend to use a service more often than not satisfied ones, 39 they present stronger repurchase intentions, and they recommend the product/services to their acquaintances. 40 It has been suggested that satisfaction has a direct effect on repurchase intentions, 41 though some scholars have found an indirect effect through adjusted expectations the post hoc expectations that are updated on the basis of consumption experiences. 42 Several studies have been conducted on satisfaction with tourism services. Satisfaction in tourism has been studied from different perspectives and for different destination types. It has been proposed that emotions such as pleasure and arousal could be antecedents of tourist satisfaction. 43 Moreover, demographic variables such as education and age, and leisure activities have been found to be good predictors of tourist satisfaction. 44 Swan et al. 45 studied compensatory satisfaction in a birding field trip. Their findings confirmed the expectancy disconfirmation paradigm. Similarly, it has been suggested that tourist (visitor) satisfaction is determined by the extent to which desired outcomes or benefits are realized. 46 Webb and Hassall 47 measured visitor satisfaction with Western Australia s conservation estate on a two-year period. They found that the type of location and the number of facilities were the strongest indicators of satisfaction. Tourism segments differ in their satisfaction and enjoyment during a trip while they do not display demographic differences in the study conducted by Andereck and Caldwell. 48 Tourist satisfaction has been proposed to be taken into account when assessing the strengths and weaknesses of a tourism organization. Moreover, it should be taken into consideration when forecasting demand for developing marketing strategies. Tourist satisfaction is central to marketing and should feed into the strategic and operational planning of tourism organizations. 49 Finally, satisfaction has been found to be directly related to destination loyalty 50 meaning that satisfied tourists will revisit and recommend a destination. Page 19

Using visit frequency to segment ski resorts customers Taking into consideration the findings of the above studies, it is expected that: H3: Satisfied customers will visit ski resorts more often. And H4: Customer ski experience, satisfaction and income will differentiate frequent visitors of ski resorts from less frequent visitors. METHODOLOGY The study utilized the survey research method to study ski resorts costumers. A self-administered anonymous questionnaire was given to 200 customers of two different ski resorts, 3 5 Pigadia and Vora- Kaimaktsalan, in Greece. The questionnaire was answered by 191 individuals (95.5 per cent response rate). The questionnaire consisted of three parts: Part I gathered demographic data; Part II asked about ski knowledge and experience and frequency of visiting ski resorts; and Part III measured customers satisfaction. Subjects had to indicate how satisfied they were with equipment, social activities and facilities using a five point likert scale (1 ¼ not at all satisfied, 5 ¼ very much satisfied). A priori segmentation was used to divide ski-resort customers into homogeneous groups by using visit frequency as segmentation base. The subjects were categorized based on their visit frequency as weekly visitors, monthly visitors and yearly visitors. However, the yearly visitors were dropped from the analysis because there were only seven cases. Satisfaction, income and ski experience were used to segment ski-resorts tourists. RESULTS Sample demographics The majority of the respondents participating in the study were men (53.4 per cent) whereas 46.5 per cent were women. The respondents of each ski resort were almost equal, 49.2 per cent came from Vora- Kaimaktsalan and 50.8 per cent came from 3 5 Pigadia. Regarding their age group, 15.5 per cent of the respondents were younger than 20 years old, 45.1 per cent were 21 30 years old, 21.2 per cent were 31 40 years old, 14 per cent were 41 50 years old, and 4.1 per cent were older than 51 years. More detailed information on sample demographics is presented in Table 2. Regarding their level of education, most of the respondents had a high school diploma (34.9 per cent), 28.1 per cent had a university degree, 7.8 per cent had a technical university degree and 10.9 per cent graduated from a college. Finally, most of the people in the study were employees in the private sector (21.1 per cent), several also worked in the public sector (15.8 per cent) whereas only 16.8 per cent were free agents, 18 per cent were students and 3.6 per cent were unemployed. Classification with discriminant analysis The basic analysis of the study was classification with discriminant analysis. Classification with discriminant analysis involves classifying subjects into one of several groups on the basis of a set of measurements. Discriminant analysis has been recommended for a priori segmentation in tourism services in order to identify and target tourists with similar needs, wants and profiles. 51 Moreover, criterion-based techniques such as discriminant analysis are supported over non-criterion methods (e.g. cluster analysis) because they have three advantages: a) the variables of interest can significantly discriminate among segments; b) they provide information on the strength of the relationship between each segment and the criterion of interest; and c) they assist in correctly classifying new observations into the identified segments. 52 Subjects that visit ski resorts once a week were assigned as y ¼ 0 and those who went once a month as y ¼ 1. Subjects who visited ski resorts yearly were also asked for, but Page 20

Tsiotsou Table 2: Sample demographics (frequencies and proportions), N 191 Gender Age Marital status Males 102 (53.4%) up to 20 30 (15.5%) Single 112 (58.9%) Females 89 (46.1%) 21 30 87 (45.1%) Married no children 73 (38.4%) 31 40 41 (21.2%) Divorced 5 (2.6%) 41 50 27 (14.0%) 51+ 8 (4.1%) Education Employment status Ski experience Middle 29 (15.1%) Unemployed 41 (21.6%), 5 years 57 (47.9%) school High 67 (34.9%) Public sector 30 (15.8%) 5 10 years 25 (21.0%) School University 54 (28.1%) Private sector 40 (21.1%) 11 20 years 25 (21.0%) College 21 (10.9%) Free agent 32 (16.8%). 21 years 12 (10.1%) Technical 15 (7.8%) Entrepreneur 22 (11.5%) college Other 5 (2.7%) Other 26 (13.2%) there were only seven cases so they were excluded from the analysis. The total sample was randomly split into a development sample of 80 subjects and a cross validation sample of 80 subjects to assess the classification accuracy of the discriminant variates (26 cases had missing data and were excluded from the analysis). The classification function was computed first on the development sample and then its hit rate was checked on the cross validation sample. The classification variables were ski experience, overall satisfaction and income. Descriptive statistics of the variables are presented in Table 3. In a preliminary analysis of the data, a case analysis was conducted to identify possible outliers and violations of the assumptions of independence, multivariate normality and the homogeneity of variance/covariance matrices. No serious violations of the assumptions were identified. The homogeneity of variance/covariance test (Box s M) indicated that the data did not violate the assumption (fail to reject at the 0.05 level; F ¼ 0.901, p ¼ 0.493). The overall multivariate relationship (MANOVA) was statistically significant at the 0.05 (chi square ¼ 13.721; Wilk s ¼ 0.836; p ¼ 0.003) indicating that the two groups of customers were statistically significantly different. Thus, the variables used were able to classify the subjects into Table 3: Descriptive statistics Variables Means Standard Deviations y ¼ 0* y ¼ 1** y ¼ 0* y ¼ 1** Ski experience 0.67 1.30 0.896 1.218 Overall satisfaction 2.23 2.73 0.945 0.952 Income 2.90 1.95 1.654 1.395 * Subjects that visit ski resorts every week (y ¼ 0). ** Subjects that visit ski resorts every month (y ¼ 1). Page 21

Using visit frequency to segment ski resorts customers the two groups based on their frequency of visiting ski resorts. Moreover, all three of the univariate F-tests were significant, as shown in Table 4. The classification was based on the Bayesian probability of group membership, assuming group priors equal to the relative group sizes (0.750 for likely to visit ski resorts every week and 0.250 for likely to visit ski resorts every month). To accomplish this classification, the Fisher s linear discriminant functions were used (Table 5). The analysis continued with the evaluation of the performance of the classification procedure. Table 6 shows the hit rate for both the development and the cross validation sample. The results for the development sample indicated 81.3 per cent correct classification rate. There were three misclassified cases. The hit rate for the cross validation sample increased to 83.8 per cent. There were four subjects misclassified. The precision of correct classification was very satisfactory and for this reason the use of the procedure for classification of future subjects is recommended. Table 4: Univariate F-tests for ski experience, overall satisfaction and income Variables Wilk s F Significance Ski experience 0.926 6.213 0.015 Overall 0.951 4.042 0.048 satisfaction Income 0.936 5.324 0.024 Table 5: Fisher s linear discriminant functions Variables Weekly visitors Monthly visitors Ski experience 1.543 2.190 Overall 2.669 3.336 satisfaction Income 1.299 1.001 Constant 5.666 8.331 DISCUSSION/IMPLICATIONS The main objective of the study was to segment ski resorts customers based on their visit frequency. This investigation confirmed previous findings on the important role of product experience, overall satisfaction and income in predicting ski resorts customers behavior. The results of the study provide important theoretical and managerial implications. The classification with discriminant analysis was used to identify consumers who would visit ski resorts weekly from those visiting them monthly. Ski experience, customer satisfaction and income were the variables used to predict frequency of ski resort visits. The main hypothesis of the study was confirmed. Ski experience, customer satisfaction and income classified weekly from monthly visitors of ski resorts. The two segments were significantly different in terms of their ski experience, overall satisfaction and income. It should be noted that the satisfaction results (univariate F-tests) were close to rejecting such a difference ( p ¼ 0.048). However, the classification procedure performed very well in correctly classifying visit frequency indicating that ski experience, overall satisfaction and income could be used for future predictions as well. Two out of the four hypotheses were not confirmed by the study. Ski experience and customer satisfaction affected visit frequency inversely. Monthly visitors of ski resorts scored higher in ski experience and satisfaction. Customers with more experience and satisfaction do not visit ski resorts as frequently as it was expected whereas less experienced and satisfied customers visit ski resorts weekly. This finding might be explained by the demographics of the study. More than 66 per cent of the sample were younger than 30 years old and thus, with an expected lower income than older customers. Because skiing is a relatively new activity in Greece, it is mostly young people that know how to ski. The dominant role of income in predicting visit frequency becomes apparent from the study and confirms previous findings. 53 Weekly visitors had a higher income mean Page 22

Tsiotsou Table 6: Classification accuracy Results for the development sample*: Actual group No. of cases Classification Weekly visitors Monthly visitors Weekly visitors 60 56 (93.3%) 4 (6.7%) Monthly visitors 20 11 (55.0%) 9 (45.0%) Total correct classification ¼ 81.3% * There were three ungrouped cases Results for the cross validation sample**: Actual group No. of cases Classification Weekly visitors Monthly visitors Weekly visitors 73 66 (90.4%) 7 (9.6%) Monthly visitors 7 6 (85.7%) 1 (14.3%) Total correct classification ¼ 83.8% ** There were four ungrouped cases score than monthly visitors. Thus, high income tourists visit ski resorts more often, though they do not have much ski experience and they are not very satisfied. Those customers are probably older and, either learnt to ski over the last few years, or, are not experienced skiers but visit ski resorts for other reasons (e.g. social). The high cost associated with visiting ski resorts and skiing prevents customers, who probably have more experience but less income, from visiting them more often. Segmentation has become increasingly important for successful marketing practices in the tourism industry. To get closer to their customers, marketers need to concentrate on the needs of specific homogeneous groups. The segments identified from the study provide useful information to managers. Having information about the satisfaction level, the ski experience, the income and the frequency of visits is possible to identify distinct segments. Moreover, the first three variables could be used as predictors of the frequency of visits in order to forecast product and service needs and better satisfy their customers. By gaining understanding on how the two segments are different, it could facilitate managers efforts in developing the appropriate marketing plan for each segment. First, ski resort managers need to retain their weekly high income customers (the most important segment) by increasing their satisfaction level and ski experience. High income with less ski experience customers might be more demanding and could be satisfied by extrinsic attributes such as offering quality services and products (e.g. quality food, more entertainment). Second, monthly lower income customers could become more frequent visitors by making them special offers (e.g. special prices on the lift use or in renting ski equipment). Profit could be made by selling other products and offering additional services (e.g. restaurants, bars, excursions) that could satisfy their needs. Another aspect that needs to be looked at by ski resorts managers in Greece is the international tourist. The rapid growth in popularity of ski resorts as tourism destinations in Greece was due to the large market increase of Greek tourists. Foreign visitors do not come to Greece to take their vacation in ski resorts. This is a market that Greek ski resorts should try to attract. Greece has snow for 4 5 months a year, whereas in several Page 23

Using visit frequency to segment ski resorts customers other European and Mediterranean countries it does not snow much. The lack of snow in some other European countries along with lower prices could be two main factors that would attract foreign skiers and visitors to Greece. Several ski resorts are close to archeological places and museums, so other type of activities could be offered in order to attract more tourists. Developing travel packages in cooperation with travel agents, advertising in international specialized magazines and developing internet sites in several foreign languages are some of the actions that could be taken in order to attract foreign tourists to Greek ski resorts. LIMITATIONS/FUTURE RESEARCH The study was intended to produce meaningful data that would provide a tool and data source on which quality marketing efforts can be based. However, this research is limited to the customers of the two ski resorts in northern Greece. Generalizations of the findings should be made with caution. Ski experience, overall satisfaction, income and visit frequency should be used to segment customers of other ski resorts in Greece and abroad. A replication of the study in other ski resorts could also provide valuable information on the role of satisfaction in predicting visit frequency. The statistical results on satisfaction were close to the 0.05 level (p ¼ 0.05), indicating that monthly visitors might not differ significantly from weekly visitors. Larger sample size could assist in gaining more confidence from the results. Moreover, future studies should investigate the kind of products and services that could satisfy ski resorts customers with different ski experience and income. REFERENCES (1) Institute for Tourism Research and Predictions (2004), URL (consulted 16 December 2004): http:www.itep.gr (2) Tuppen, J. (2000) The restructuring of winter sports resorts in the French Alps: Problems, processes and policies, International Journal of Tourism Research 2: 327 44; Richards, G. (1996) Skilled consumption and UK ski holidays, Tourism Management 17(1): 25 34; Tsiotsou, R. (2003) General characteristics of ski resorts customers in Greece, Physical Education and Sport 5: 20 4. (3) Richards, ref. 2 above. (4) Kotler, P. (2000) Marketing Management. New Jersey: Prentice-Hall Inc. (5) Chen, J. S. (2003) Market segmentation by tourists sentiments, Annals of Tourism Research 30(1): 178 93. (6) Tsiotsou, R. and Vasioti, E. (2006) Satisfaction: A segmentation criterion for short term visitors of mountainous destinations, Journal of Travel & Tourism Marketing 19(4); Schmidt, M. and Merser, H. (2005) Segmenting the tourist market using factor and cluster analysis, (Working Paper) (Available: http://www.mic.cbs.dk/marcus/gbpapers/ Nashvil/NASHVIL.htm); Lawson, R. and Thyne, M. (2005) Expenditure patterns in six travel lifestyle segments, (Working Paper); Dolnicar, S. and Leisch, F. (2004) Segmenting markets by bagged clustering, Australasian Marketing Journal 12(1): 51 65; Bloom, J. Z. (2004) Tourist market segmentation with linear and non-linear techniques, Tourism Management 25: 723 33; Bigne, E. J. and Andreu L. (2004) Emotions in segmentation: An empirical study, Annals of Tourism Research 31(3): 682 96; Frochot, I. (2005) A benefit segmentation of tourists in rural areas: A Scottish perspective, Tourism Management 26: 335 46; Chen, J. S. (2003) Market segmentation by tourists sentiments, Annals of Tourism Research 30(1): 178 93; Becken, S. (2003) An integrated approach to travel behaviour with the aim of developing more sustainable forms of tourism, Landcare Research Internal Report, URL (consulted 15 November 2004): http://www.landcareresearch.co.nz/research/ sustain_business/tourism/integrated_approaches. asp; Tureckova, O. U. R. (2002) Segmenting the tourism market using perceptual and attitudinal mapping, Agriculture Economy 48(1): 36 48; Mok, C. and Iverson, T. J. (2000) Expenditure-based segmentation: Taiwanese tourists to Guam, Tourism Management 21: 299 305; Juaneda, C. and Sastre, F. (1999) Balearic islands tourism: A case study in demographic segmentation, Tourism Management 20: 549 52; Richards, ref. 2 above. Page 24

Tsiotsou (7) Lawson and Thyne, ref. 6 above. (8) Dolnicar and Leisch, ref. 6 above. (9) Mok and Iverson, ref. 6 above. (10) Richards, ref. 2 above. (11) Richards, ref. 2 above. (12) Alba, J. W. and Hutchinson, J. W. (1987) Dimensions of consumer expertise, Journal of Consumer Research 13(March): 411 54. (13) Park, W. C., Mothersbaugh, D. and Feick, L. (1994) Consumer knowledge assessment, Journal of Consumer Research 21(June): 71 83. (14) Bettman, J. R. and Park, W. C. (1980) Effects of prior knowledge and experience and phase of the choice process on consumer decision processes: A protocol analysis, Journal of Consumer Marketing Research 7(December): 234 48. (15) Park et al., ref. 13 above. (16) Mano, H. and Oliver, R. L. (1993) Assessing the dimensionality and structure of the consumption experience: Evaluation, feeling, and satisfaction, Journal of Consumer Research 20(3): 451 66. (17) Bettman and Park, see ref. 14 above; Park et al., see ref. 13 above. (18) Mano and Oliver, ref. 16 above. (19) Alba and Hutchinson, ref. 12 above. (20) Huffman, C. and Houston, M. J. (1993) Goal-oriented experiences and the development of knowledge, Journal of Consumer Research 20(2): 190 210. (21) Bettman and Park, ref. 14 above. (22) Huffman and Houston, ref. 20 above. (23) Ryan, C. (1994) Leisure and tourism: The application of leisure concepts to tourist behaviour a proposed model, in A. V. Seaton (ed.) Tourism: The State of the Art, pp. 294 307. London: Wiley. (24) Lehto, X. Y., O Leary, J. T. and Morrison, A. M. (2004) The effect of prior experience on vacation behavior, Annals of Tourism Research 31(4): 801 18. (25) Ewart, A. and Hollenhorst, S. (1994) Individual and setting attributes of the adventure recreation experience, Leisure Science 16: 177 91; Ditton, R. B., Loomis, D. K. and Choi, S. (1992) Recreation specialization: Re-conceptualiyation from a social worlds perspective, Journal of Leisure Research 24: 33 51. (26) Oliver, R. L. and DeSarbo, W. S. (1988) Response determinants in satisfaction judgments, Journal of Consumer Research 14(March): 495 507. (27) Giese, J. L. and Cote, J. A. (2000) Defining consumer satisfaction, Academy of Marketing Science Review, URL (consulted 10 October 2004): http://www.amsreview.org/articles/ giese01 2000.pdf (28) Morgan, M. J., Attaway, J. S. and Griffin, M. (1996) The role of product/service experience in the satisfaction formation process: A test of moderation, Journal of Consumer Satisfaction, Dissatisfaction and Complaining Behavior 9: 104 14; McQuitty, S., Finn A. and Wiley, J. B. (2000) Systematically varying consumer satisfaction and its implications for product choice, Academy of Marketing Science Review, URL (consulted 10 October 2004): http://www.amsreview. org/articles/mcquity10 2000.pdf (29) Erevelles, S. and Leavitt, C. (1992) A comparison of current models of consumer satisfation/dissatisfaction, Journal of Consumer Satisfaction, Dissatisfaction and Complaining Behavior 5: 104 14. (30) Woodruff, R. B., and Gardial, S. F. (1996) Know Your Customer: New Approaches to Understanding Customer Value and Satisfaction. Cambridge: Blackwell Publishers. (31) Oliver, R. L. (1980) A cognitive model of the antecedents and consequences of satisfaction decisions, Journal of Marketing Research 17(Nov.): 460 9. (32) Churchill, G. A. Jr. and Suprenant, C. (1982) An investigation into the determinants of consumer satisfaction, Journal of Marketing Research 19(Nov.): 491 504. (33) Folkes, V. S. (1984) Consumer reactions to product failure: An attributional approach, Journal of Consumer Research 10: 398 409. (34) Westbrook, R. A. (1987) Product/ consumption-based affective responses and postpurchase processes, Journal of Marketing Research 24(August): 258 70. (35) Oliver and DeSarbo, ref. 26 above. (36) McQuitty, S., Finn A. and Wiley, J. B. (2000) Systematically varying consumer satisfaction and its implications for product choice, Academy of Marketing Science Review, URL (consulted 10 October 2004): http:// www.amsreview.org/articles/mcquity10 2000.pdf (37) Yang, Z. and Peterson, R. (2004) Customer satisfaction, perceived value, and loyalty: The role of switching costs, Psychology & Marketing 21(10): 799 822. (38) Homburg, C. and Giering, A. (2001) Personal characteristics as moderators of the Page 25

Using visit frequency to segment ski resorts customers relationship between customer satisfaction and loyalty An empirical analysis, Psychology & Marketing 18(1): 43 66. (39) Bolton, R. N. and Lemon, K. N. (1999) A dynamic model of customers9 usage of services: Usage as an antecedent and consequence of satisfaction, Journal of Marketing Research 36: 171 86. (40) Zeithaml, V. A., Berry, L. L. and Parasuraman, A. (1996) The behavioral consequences of service quality, Journal of Marketing 60: 31 46. (41) Reichheld, F. F. and Teal, T. (1996) The Loyalty Effect. Boston, MA: Harvard Business School Press; Tsiotsou, R. (2005) The role of perceived product quality and overall satisfaction on purchase intentions, Proceedings, 9 th International Conference on Marketing and Development (forthcoming). (42) Yoon, Y. and Uysal, M. (2005) An examination of the effects of motivation and satisfaction on destination loyalty: A structural model, Tourism Management 26: 45 56. (43) Bigne and Andreu, ref. 6 above. (44) Tsiotsou and Vasioti, ref. 6 above. (45) Swan, J. E., Martin, S. M. and Trawick, I. F. Jr. (2003) Compensatory satisfaction: An ethnography of avoiding disappointment and producing satisfaction in birding, Journal of Consumer Satisfaction/Dissatisfaction and Complaining Behavior 16: 157 65. (46) Tian-Cole, S. and Cromption, J. L. (2003) A conceptualization of the relationships between service quality and visitor satisfaction, and their links to destination selection, Leisure Studies 22: 65 80. (47) Webb, D. and Hassall, K. (2002) Measuring visitor satisfaction with western Australia s conservation estate, Journal of Consumer Satisfaction, Dissatisfaction and Complaining Behavior 15: 81 98. (48) Andereck, K. L. and Caldwell, L. L. (1994) Variable selection in tourism market segmentation models, Journal of Travel Research 33(2): 40 6. (49) Satish, C. and Menezes, D. (2001) Applications of multivariate analysis in international tourism research: The marketing strategy perspective of NTOs, Journal of Economic and Social Research 3(1): 77 98. (50) Yoon and Uysal, ref. 42 above. (51) Satish and Menezes, ref. 49 above. (52) Chen, ref. 5 above. (53) Richards, ref. 2 above; Dolnicar and Leisch, ref. 6 above; Lawson and Thyne, ref. 6 above. Page 26