Research Journal of Hospitality & Tourism

Journal of Hospitality & Tourism Research http://jht.sagepub.com/

Market Segmentation Based on Tourists' Dining Preferences Atila Yüksel and Fisun Yüksel Journal of Hospitality & Tourism Research 2002 26: 315 DOI: 10.1177/109634802237482 The online version of this article can be found at: http://jht.sagepub.com/content/26/4/315

Published by: http://www.sagepublications.com

On behalf of:

International Council on Hotel, Restaurant, and Institutional Education

Additional services and information for Journal of Hospitality & Tourism Research can be found at: Email Alerts: http://jht.sagepub.com/cgi/alerts Subscriptions: http://jht.sagepub.com/subscriptions Reprints: http://www.sagepub.com/journalsReprints.nav Permissions: http://www.sagepub.com/journalsPermissions.nav Citations: http://jht.sagepub.com/content/26/4/315.refs.html

>> Version of Record - Nov 1, 2002 What is This?

Downloaded from jht.sagepub.com at Adnan Menderes Universitesi on May 5, 2013

JOURNAL 10.1177/109634802237482 Yüksel, Yüksel OF HOSPITALITY / MARKET SEGMENTATION & TOURISM RESEARCH

MARKET SEGMENTATION BASED ON TOURISTS’ DINING PREFERENCES Atila Yüksel Adnan Menderes University, Turkey Fisun Yüksel Sheffield Hallam University, United Kingdom Managerial benefits of market segmentation have been justified; however, incorporation of this concept into restaurant management philosophy is limited, particularly in tourist resorts. A factor analytic approach undertaken in this article revealed that there were different tourist dining segments, developing their restaurant selection decisions based on different aspects of restaurant services, which requires segment-specific marketing and management strategies. The analysis identified five distinct tourist dining segments that take different sets of elements into account when making their restaurant selection decision. Marketing implications of the study findings are discussed. KEYWORDS: segmentation; restaurant selection; tourist satisfaction.

One of the most important strategic concepts contributed by marketing discipline to business firms and other types of organizations is that of market segmentation (Bowen, 1998). Segmentation is a powerful tool that serves to develop understanding of the differential influence of specific service variables across segments and to the development of more precise marketing strategies (Reid, 1983; Richard & Sundaram, 1994; Swinyard & Struman, 1986). An understanding of what various segments require and formulation of focused management strategies to fulfil these specific requirements effectively are the key to penetrate new markets and to maintaining a repeat customer base. Segmentation, when done properly, can actually enhance sales and profits because it allows the organization to target segments that are much more likely to patronize the organization’s facilities (Reid, 1983). Contrary to the benefits of becoming and remaining close to target customers, however, the market segmentation concept in foodservice operations, particularly in tourist resorts, is relatively a neglected issue. It appears that restaurateurs in tourist resorts are trying to appeal to all potential customers, and they seem to believe that by segmenting the market, they will weaken their sales volume. Generally speaking, foodservice operators seem to base their marketing practices on their own intuition and assume the price to be the most effective weapon in the batJournal of Hospitality & Tourism Research, Vol. 26, No. 4, November 2002, 315-331 DOI: 10.1177/109634802237482 © 2002 International Council on Hotel, Restaurant and Institutional Education

315


316

JOURNAL OF HOSPITALITY & TOURISM RESEARCH

tle for market share. This overemphasis on price, as the sole means of attracting and retaining customers, however, may lead to the development of the “profitless prosperity syndrome” (Crawford-Welch, 1994), whereby restaurateurs can often sell their products but the price is not high enough to ensure adequate profit levels. As a consequence, the rate of business failure of restaurants is generally high (English, Josiam, Upchurch, & Willems, 1996), particularly in tourist resorts. Given the scarcity of research on tourists visiting resort restaurants, this research sets out to develop a framework to assist restaurateurs in designing effective market-specific strategies to attract, satisfy, and retain targeted customer markets. Specifically, the first part of the research explores whether tourists can be grouped into distinct subsegments based on similarities and differences in benefits sought from restaurants. The research then ascertains whether statistically significant differences exist between the resulting segments on the basis of demographics and visit-related variables. Finally, the research investigates whether sources of satisfaction differ within each resulting segment (the results relating to this objective are not discussed here). To this end, the following section first reviews the literature on market segmentation. The research method and the findings are then presented. Next, marketing implications of the research are discussed. LITERATURE REVIEW

Although the lodging and travel literature is replete with market segmentation studies, there are only a few studies that have investigated the benefits sought by different segments within the restaurant sector (Bojanic & Shea, 1997). Lewis (1981) provided an early application of benefit segmentation with some results for management decision making. Lewis investigated the users and nonusers of family, theme, and gourmet restaurants and found that segments differed in their opinions about the importance of several service attributes. On the basis of an analysis of customers’ value benefits factors, lifestyle factors, usage patterns, and demographic descriptors, Swinyard and Struman (1986) identified three customer segments: family diners, romantics, and entertainers. By analyzing customer expectations of fast-food restaurants, Oh and Jeong (1996) found four different customer segments: neat service seekers, convenience seekers, classic diners, and indifferent diners. Bahn and Granzin (1985) have tested the merit of benefit segmentation to restaurant marketers and found how nutritional concerns could affect restaurant patronage. They discovered four distinct segments: the health segment, the gourmet segment, the value segment, and the unconcerned segment. They found that the health segment was highly concerned with issues involving nutrition, whereas the gourmet segment was less concerned with the nutrition issue. Their study revealed that these four segments were likely to patronize different restaurant types. For example, the health segment is unlikely to frequent fast-food restaurants, whereas the value group is most likely to patronize fast-food restaurants. To capture the health segment, they suggest that restaurateurs should stress the nutritiousness of their food offerings in their advertise-


Yüksel, Yüksel / MARKET SEGMENTATION

317

ments. Granzin and Olsen’s (1997) recent work revealed three groups of consumers relating to fast-food restaurants: nonusers, light users, and heavy users. Kivela (1997) undertook a study in different types of restaurants, including fine dining/gourmet, theme/atmosphere, family/popular, and convenience/fastfood restaurants, to identify main choice variables. He then analyzed whether perceived importance of choice variables differed by dining occasion, age, and income segments. The results of Kivela’s study indicated that customers’ preferences of choice variables varied significantly by restaurant type, dining occasion, age, and occupation. In a study of restaurant segmentation in the United Kingdom, Auty (1992) identified three segments: students, well-to-do middle-aged people, and older people. She found that image and atmosphere were the most critical factors in the final choice between similar restaurants. Williams, DeMicco, and Kotschevar (1997) investigated the physiological and psychological challenges that the older restaurant customer segment faces. Similarly, Becker-Suttle, Weaver, and Crawford-Welch (1994) undertook a segmentation study between groups of senior and nonsenior citizens regarding benefits sought in full-service restaurant dining. Using a conjoint analysis, they identified discrepancies between the two groups’ requirements regarding menu variety and portion size. In a study of downtown and suburban restaurants, Bojanic and Shea (1997) found a significant difference between the satisfaction drivers of customers patronizing these restaurants. In their comparative cross-national segmentation study, Kara, Kaynak, and Kucukemiroglu (1997) found differences between U.S. and Canadian consumers in terms of the relationship between frequency of purchase and attributes considered important in selecting fast-food restaurants. Shank and Nahhas (1994) applied segmentation analysis to examine dining preferences of mature and younger consumers frequenting family dining/mediumpriced restaurants and found differences in terms of dining preferences, dining habits, and loyalty. In a recent study, Shoemaker (1998) discovered five distinct segments among university students: perceptive shoppers, expedient shoppers, 24-hour social students, focused diners, and demanding diners. A review of these studies clearly demonstrates that there are distinct customer groups within the total market, and thus, managers can enhance sales volume and profits by developing market-specific strategies based on a scientific approach to segmentation, rather than on the basis of their own intuition. It is important to note that the majority of these studies were undertaken with domestic customers patronizing fast-food restaurants, whereas no segmentation research was undertaken (to the authors’best knowledge) with tourists on vacation visiting independent non-fast-food restaurants. At present, there is inadequate understanding of whether tourists’ eating-out patterns and benefits that they seek from restaurants on vacation are relatively different from when they are not on vacation. In addition, although past studies have focused on the identification of factors that may differentiate segments, the scrutiny of dining satisfaction and repeat business determinants of these segments seems to have received inadequate attention from researchers (Oh & Jeong, 1996). Identification of the segments and attracting them could be one thing, but to secure repeat business is another (Lowenstein,


318


1995). Thus, it is imperative to understand what brings or discourages satisfaction and repeat business in relation to specific markets. MARKET SEGMENTATION

Researchers concur that good market segments generally consist of consumers with homogeneous product needs, attitudes, and responses linked to marketing variables (Loker & Perdue, 1992). The segments should also be distinctive from one another, so that group membership of an individual segment is clearly based on key variables. Another characteristic of a good market segment is substantiality: A segment should be large enough to be profitable (Loker & Perdue, 1992). Researchers and marketers have long struggled with trying to identify key underlying variables that are good discriminators between groups of people having different response characteristics to a product or service. As a consequence, a number of different segmentation variables have been utilized in the market segmentation literature (Oh & Jeong, 1996). A review of the analytic literature on the hospitality sector shows that most segmentation studies are descriptive in nature, as they rely on an a priori approach and have used respondents’ sociodemographic variables as segmentation criteria. For example, Almanza , Jaffe, and Lin (1994) segmented the cafeteria restaurant market simply by basing it on such demographic variables as frequency of patronage, age, gender, income, and household size. The use of sociodemographics as segmentation variables, however, has attracted criticism (Becker-Suttle et al., 1994; Crawford-Welch, 1991; Haley, 1985; Loker & Perdue, 1992; Oh & Jeong, 1996; Swinyard, 1977). Oh and Jeong (1996), in their study of fast-food restaurant customers, reported that market segmentation by well-documented demographic variables such as gender, age, and household income was not successful in understanding market-specific customer expectations because the demographic-based markets could not isolate marketspecific expectations successfully. They state that what customers expect from fast-food restaurants may not be clearly identified by simply looking at their age, gender, and household income. Crawford-Welch (1991) criticized the use of demographics as segmentation variables and noted that “descriptive data, by their very nature, are of little analytical worth in that they are not capable of implying causality and are, in turn, poor predictors of behaviour” (p. 301). Similarly, Swinyard (1977) emphasized that sociodemographics had very low discriminatory powers in responses between segments. In search of more effective segmentation variables, some researchers have used information relating to customers’ buying behavior, customers’ buying situations, and customers’decision-making style (see Oh & Jeong, 1996). In addition, in recent years, benefit segmentation has emerged as an effective approach to market segmentation through which it is possible to identify market segments by causal, rather than descriptive, factors. The belief underlying this strategy is that benefits that people are seeking, in consuming a given product, are the basic reasons for the existence of true market segments and are better determinants of behavior than other approaches (Loker & Purdue, 1992). One of the major bene-



319

fits of this approach is to enable a service provider to implement different marketing strategies, for different segments, by offering unique benefits sought by each segment (Woo, 1998). It is reported that benefits predict behavior better than personality and lifestyle, volumetric, demographic, or geographic measures, which merely describe behavior without explaining it (Crawford-Welch, 1991; Haley, 1985; Loker & Perdue, 1992). As restaurants provide a number of services, it seems appropriate to consider the benefits in terms of attributes of the total service product provided (Bahn & Granzin, 1985). THE RESEARCH METHOD

A research instrument was developed to find out the choice criteria that tourists use when selecting restaurants on vacation and to examine tourists’ opinions of the prime components of their dining experiences in independent non-fast-food (INFF) restaurants. The research instrument comprised more than 110 items and was grouped into four major areas: general information about the respondent and dining occasion, ratings on 42 attributes of restaurant selection, ratings on 44 attributes of restaurant service performance, and ratings on overall dining satisfaction and behavioral intentions. The items were derived from interviews and previous studies, and their adequacy was checked through a pilot test. Respondents were also given ample space to make any further written comments (compliments and complaints). The questionnaire was three pages long (double-sided) and was written in English and German. Two bilingual translators from the university were employed to translate the instrument from its original in English to German and from German back to English. The respondents were first required to indicate on a 7-point scale (extremely important to not important at all) the importance of 42 items when selecting a restaurant while on a summer holiday. The respondents were then required to assess the performance of restaurant services on 7-point semantic differential scales. The study adopted the use of a single overall measure of tourist satisfaction. Although some researchers contend that satisfaction should be measured by a combination of attributes, the ease of use and empirical support for an overall measure of satisfaction led to its selection (Halstead, 1989). The study employed the Delight-Terrible Scale for measuring overall dining satisfaction, as it has been reported to be the most reliable satisfaction scale (Maddox, 1985). Similarly, respondents’ return intentions and word-of-mouth recommendations were assessed by single overall measures. The actual survey was carried out with 500 tourists who had dined in independent restaurants. The research was conducted with tourists departing from an international airport during an 18-day period in May and June 1998 in Turkey. Due to permission-related difficulties that the researchers encountered in implementing the survey in restaurants, the airport was chosen. To eliminate any possible effect of memory lapse, participants were requested to evaluate their most recent dining experience within a non-fast-food restaurant outside their accommodation. A pretest was run (N = 30) to understand whether this approach posed any memory problem. None of the respondents reported any problem recalling and evaluating


320


their last dining experiences. Assessment of customers’ last experiences, with a product or a service, can be found in other studies (e.g., Clow, Kurtz, & Ozment, 1996; Parasuraman, Zeithaml, & Berry, 1988). Given their flight times, 31 tourists refused to participate and 20 returned incomplete questionnaires. Of all the respondents, 44% were male and 54% had eaten out in the restaurant on more than one occasion. Of all the restaurants, 72% were midscale and 23% were high-scale INFF restaurants. Statistics of location are as follows: 51% were situated in Marmaris, 5% in Fethiye, 1% in Bodrum, and 43% in other resorts including TurgutReis and Hisaronu. Given the flight destinations at the time of survey implementation, together with the growing number of British tourists, the majority of respondents were British (64%), followed by German (11%), Scandinavian (11%), and others (including Italian, French, Russian, and Benelux). The accuracy of the sample representation in terms of nationality was assessed by comparing the list of departing tourists, acquired from the airport authority, with the country of origin of the actual sample. This comparison suggested that departing tourists, in this period of time, were commensurably represented in the sample. ANALYSIS

As the researcher did not know the exact market profile a priori, a factorclustering approach was employed to segment tourists into homogeneous groups that differed from each other on the basis of opinions regarding which attributes were important when selecting a restaurant. Factor-cluster segmentation is a twostep procedure (Smith, 1995). The first step involves the definition of important characteristics of the segments through a factor analysis of a large number of descriptive variables. These variables are then used to cluster individuals into statistically homogeneous segments. Consistent with the suggestions of Calantole and Johar (1984), Everitt (1993), Singh (1990), and Punj and Stewart (1983), in this study, customer restaurant selection dimensions, identified by the factor analysis, provided the data to cluster analysis instead of the original ratings on variables. This is because the raw (original) variables contain interdependence that is likely to bias the cluster analysis results (Smith, 1995). By contrast, the use of dimensions removes such interdependencies through representing data by a relatively independent and parsimonious set of factors. Correspondingly, Mo, Havitz, and Howard (1994) pointed out that factor scores are more reliable than single original variables, as they are weighted as linear combinations of variables and are more readily interpreted than a large number of variables. Applications of cluster analysis in marketing, using factor scores derived from factor analysis, have often been suggested (Singh, 1990). Figure 1 presents the statistical techniques used to meet the study objectives. Step 1: Factor Analysis

To reduce the total number of items into a more workable number of composites, the attributes of restaurant selection were first factor analyzed. The principal component factor analysis with Varimax rotation was used, because this method



321

Figure 1 Statistical Techniques Used FACTOR ANALYSIS

Factor analysis was used to extract factors from the 42 attributes of restaurant selection.

RELIABILITY ANALYSIS

Reliability analysis was used to estimate the reliability coefficients for each factor.

HIERARCHICAL CLUSTER ANALYSIS

Factor scores were then used in a hierarchical cluster analysis to obtain some idea about the number of homogeneous groups represented by the data.

K-MEANS CLUSTER

Quick cluster analysis was run on the total number of respondents.

DISCRIMINANT ANALYSIS

Discriminant analysis was then employed to profile the groups obtained in the quick cluster technique according to demographic and other related data.

was found to yield the most interpretable results (Loker & Perdue, 1992). The eigenvalues suggested that a nine-factor solution explained 61.2% of the overall variance before rotation. The overall significance of the correlation matrix was .0000, with a Barlett Test of Spherity value of 6741.7003. This indicated that the data matrix had a sufficient correlation to the factor analysis. The Keiser-MeyerOlkin overall measure of sampling adequacy was .91, which suggests that data were appropriate to factor analysis. From the orthogonal factor matrix, nine factors with 28 variables were defined. The communality of each variable was relatively high, ranging from .42 to .76. This indicates that the variance of the original values was captured fairly well by the nine factors. Each factor name was based on the characteristics of its composing variables (Table 1). The first factor was labeled Service Quality and Staff Attitude, as this


322


Table 1 Restaurant Selection Factors Factor Name Factor 1: Service Quality Service standard Service efficiency Attentive service Helpful staff Competent staff Staff appearance Prices shown clearly Factor 2: Product Quality/Hygiene Food preparation consistency Food tastiness High-quality food Fresh ingredients Hygienic food preparation Staff cleanliness Factor 3: Adventurous Menu Adventurous menu Availability of local dishes Availability of interesting food A place frequented by locals Factor 4: Price and Value Reasonable food prices Food value for money Hearty portions Factor 5: Atmosphere Restaurant atmosphere Activity and entertainment Factor 6: Healthy Food Availability of healthy food Nutritious food Factor 7: Location and Appearance Impression from the road Convenient location Factor 8 Availability of nonsmoking area Factor 9 Visibility of food preparation area

Factor Loadings Eigenvalues

Variance (%)

Reliability

12.29

29.3

.88

3.33

7.9

.86

2.45

5.9

.79

1.64

3.9

.65

1.36

3.3

.63

1.2

2.9

.81

1.17

2.8

.54

1.12

2.7

—

1.09

2.6

—

.534 .566 .576 .772 .752 .533 .739 .660 .589 .681 .718 .765 .625 .687 .773 .770 .689 .632 .695 .569 .772 .695 .759 .788 .713 .684 .723 .617

factor was formed by the variables of service standard, service efficiency, attentive service, helpful staff, competent staff, and staff appearance. This factor explained 29% of the total variance and had an eigenvalue of 12.29 (Table 1). The second factor was labeled Product Quality and Hygiene, as this factor was markedly composed of food quality and hygiene-related variables. This factor explained 8% of the total variance and had an eigenvalue of 3.33. Other factors were labeled in accordance with their composite characteristics.



323

Step 2: Reliability Analysis

A composite reliability of the construct was then calculated to measure the internal consistency of each factor. The results show that the reliability coefficients for factors exceeded the recommended level of .50 (ranging from .54 to .88) (Table 1) (Nunnaly, 1967). One might conclude that these nine dimensions were perceived as particularly important by the sample of tourists in selecting restaurants on holiday. Step 3: Cluster Analysis

The cluster analysis was used to identify and classify tourists on the basis of the similarities of their characteristics, and in this case, on the attributes of restaurant selection. As mentioned earlier, factor scores, estimated from a nine-factor rotated solution, were used to determine the number of homogeneous groups represented by the data. Because the cluster analysis is known to be sensitive to the outliers, the data were first examined for outlying observations (Singh, 1990). On examination, the outlying observations (N = 25) were deleted so as to make cluster analysis safe for the entire data (Hair, Anderson, & Black, 1995). Given the sample size, the researcher adopted a nonhierarchical clustering approach (Ahmed, Barber, & Astous, 1998; Noruris, 1994). One of the difficulties in conducting cluster analysis is that the best way to determine the appropriate number of clusters is yet to be resolved (Aldenderfer & Blashfield, 1984; Alzua, O’Leary, & Morrison, 1998; Hair et al., 1995; Lewis, 1984). Whereas some researchers suggest the analysis of distances between clusters at successive steps as a useful guideline (Hair et al., 1995; Norusis, 1994), others use the comparison of cluster means on each dimension and assessment of the distinctiveness of clusters (Ahmed et al., 1998). Plugging a different number of clusters into the data until repeated clustering iterations give the best number of clusters is also suggested as another basic approach (Lewis, 1984). Consistent with Hair et al. (1995) and Alzua et al.’s (1998) suggestions, a hierarchical technique (complete linkage with squared Euclidean distance) was used on a randomly selected subsample to obtain some idea about the number of clusters. A visual inspection was carried out of the horizontal icicle dendogram on the computer printout and the sudden jumps in the algorithm schedule (Weaver, McCleary, & Jinlin, 1994). This inspection suggested that a three-, four-, and fivecluster solution might be appropriate. Subsequently, consistent with Woo’s (1998) approach, K-means cluster analysis was performed on the three different cluster solutions (n = 3, 4, and 5). Comparing the results obtained from these solutions, the five-cluster solution was selected for further analysis because it provided the greatest difference between clusters and yielded the most interpretable results (Madrigal & Kahle, 1994). Step 4: Discriminant Analysis

To test whether significant differences in restaurant selection criteria exist across segments, a multivariate analysis of variance (MANOVA) was conducted using the segments as the independent variable and the nine factors as dependent


324


variables. Wilks’s lambda was .06 and significant at the .000 level, which indicated overall differences between clusters. It is important to note that a fundamental assumption of MANOVA was satisfied, as a test of equality of group covariance matrices using Box’s M (Box’s M = 3,449.778, F = 1.81 with 180, 117504 df, p = .000) indicated that the covariance was equal. Then, a univariate F test was used to investigate the sources of these group differences. The results reveal that clusters were significantly different on all determinant restaurant selection factors. The identified cluster structure was then subjected to discriminant analysis to double-check, in part, the classifications’ reliability. Discriminant analysis was employed to see how well the nine factors predicted membership in each cluster. It calculates the weights of different combinations of the nine determinant restaurant selection factors to maximize the distance between five clusters. The stepwise procedure was used to analyze the determinant restaurant selection factors because the research objective was to determine the best discriminating factor set between clusters. The stepwise procedure began by selecting the single best discriminating factor. This factor was then paired with each of the other determinant factors (Yoon & Shafer, 1997). The second factor was chosen that was best able to improve the discriminating power of the function in combination with the first factor and so forth. The discriminant analysis indicated that nine factors significantly predicted cluster membership at a significant level of .0000. This suggests that these factors are significant discriminators. The classification results for the use of the analysis indicated that the discriminant analysis model could correctly classify 95.7% of the individuals into groups. Discriminant analysis was further used to develop cluster profiles by using demographic information and occasion-related data that were not involved in the cluster procedure. The demographic information used in the analysis included gender, age, and nationality, and occasion-related data included the type of dining occasion, party size, and the degree of familiarity with the restaurant. The discriminant analysis indicated that only age was a significant discriminator at the .10 level (the significance was .0529). This finding, consistent with the finding of a recent study (Oh & Jeong, 1996), indicates that demographic variables may not be powerful discriminators, and thus developing a marketing strategy based on descriptive variables may be inappropriate. The analysis of variance (ANOVA) procedures were then used to determine whether statistically significant differences existed between the factor mean scores of each cluster. A Duncan range test (the alpha level was set at .01) was used to determine which means were significantly different (Table 2). Cluster 1: Value Seekers As in the factor analysis, each cluster was labeled in accordance with the characteristics of its composites. The first cluster contains 19% of the total cluster sample (58 of 304). It has a high mean score for Factor 4 (.53) (Table 2), which indicates that tourists in this group select restaurants that can provide food value for money. There is also a relatively high mean score for Factor 2, which suggests that this group attaches a great deal of importance to food quality and hygiene when selecting a restaurant. In addition, based on the mean scores on Factors 8 and



Table 2 Segment Mean Scores Factor 1 Cluster 1 (58) Cluster 2 (50) Cluster 3 (45) Cluster 4 (92) Cluster 5 (59)

F value

a

–.08785 .34055 .05908 .15151 –.28657 4.031 p < .003

a. Number of tourists.

Factor 2

Factor 3

Factor 4

Factor 5

Factor 6

Factor 7

Factor 8

Factor 9

.29073 –.93254 –.03919 .08121 .55615

–.23853 –.71444 .62119 .42359 .03498

.52866 .28207 –.59885 .46662 –.77846

–.15460 –.10756 –.14155 .60355 –.62475

–.61848 .17885 –.96049 .42266 .54155

–1.10189 .15976 .50339 .37907 .11434

.11793 –.02219 –.74047 .35181 .12983

.18646 –.48819 –.30311 .28568 –.09315

34.934 p < .000

9.082 p < .000

3.324 p < .008

58.718 p < .000

38.428 p < .000

20.057 p < .000

25.392 p < .000

10.065 p < .000

325

326


9, the availability of nonsmoking areas and the visibility of the food preparation area is also important for this group. On the other hand, this group has the lowest mean score for Factor 7 (–1.10), which indicates that they do not take location of the restaurant into consideration in their selection. In addition, mean scores on Factor 6 suggest that tourists in this group are not overly concerned about the availability of healthy and nutritious food choices either. Moreover, given their mean scores on Factor 3, it is appropriate to state that this group does select restaurants that offer local or adventurous food. In addition, restaurant atmosphere and service quality do not appear to be particularly important in these tourists’ restaurant selection. Cluster 2: Service Seekers This group accounts for 16% of the sample. Given the mean scores, it can be stated that this group attaches the highest importance to the availability of quality service when selecting a restaurant. The value of food is another factor that is taken into account when selecting a restaurant. This group also attaches moderate importance to the availability of healthy choices and the restaurant’s location. In contrast to Cluster 5 and Cluster 1, this group does not take food quality into consideration when selecting a restaurant. In addition, the low mean score on Factor 3 suggests that, in contrast to Clusters 3 and 4, the availability of local and interesting food does not appear to play a significant role in this group’s decision-making process. Cluster 3: Adventurous-Food Seekers This cluster accounts for 15% of the total cluster sample. An examination of the mean scores displayed in Table 2 reveals that this group attaches the highest importance to the availability of local, new, and interesting food when selecting a restaurant. This group has also a high mean score on Factor 7, which suggests that location of a restaurant plays a relatively significant role in the selection process. Furthermore, given the mean scores on Factor 6, it could be said that the availability of healthy and nutritious food choices is not particularly important to this group. In contrast to the value-oriented tourist group, this group does not attach any importance to prices. In addition, contrary to Cluster 4, mean scores on Factor 5 indicate that atmosphere does not play a significant role in the decision-making process. Based on the mean scores on Factors 8 and 9, it can be stated that the availability of nonsmoking areas and the visibility of the food preparation area do not have any influence on this group’s restaurant selection. Cluster 4: Atmosphere Seekers This cluster makes up 30% of the total cluster sample. In contrast to other groups, this group does not have negative mean scores on any of the selection factors, which suggests that all of the factors might be taken into account in selection. Contrary to the value-oriented, service-oriented, and healthy food–oriented groups, this group has the highest mean score on Factor 5, which indicates that this group searches for a restaurant that is capable of offering a convivial dining atmosphere and a good time. This group is also concerned about prices when



327

selecting a restaurant. The availability of local and interesting food and of nutritious and healthy food are the next two most important requirements. Moreover, based on their mean score on Factor 7, it could be argued that a restaurant’s location is an important factor in influencing the selection. The remaining factors also appear to have a moderate importance in restaurant selection. Cluster 5: Healthy-Food Seekers This cluster contains 19% of the total cluster sample. Contrary to the valueoriented group, this group has one of the highest mean scores on Factor 6, which indicates that they search for restaurants offering healthy food choices. Food quality and hygiene is another highly important requirement of this group when selecting a restaurant. Based on their mean score on Factor 7, it could be said that this group is also concerned about location. In addition, based on the mean scores on Factor 8, it seems that this group attaches a moderate importance to the availability of nonsmoking areas when selecting restaurants. On the other hand, this group has the lowest mean score on Factor 4, which suggests that price is not an important consideration for this group in selecting a restaurant. Given their mean score on Factor 5, it appears that this group does not take atmosphere into consideration when selecting restaurants. Low mean scores on Factor 1 imply that this group does not attach high importance to service quality when selecting a restaurant. In addition, adventurous food is not particularly important in their restaurant selection. DISCUSSION

Restaurateurs in tourist resorts need to know tourist requirements from restaurants and tailor-make products/services to meet these requirements to attract and retain customers. The study results indicate a number of service aspects that might be taken into consideration when tourists are making restaurant selection decisions. These include service quality, staff attitude, product quality, adventurous menu, product value and price, atmosphere, healthiness of dishes, location, availability of a nonsmoking area, and visibility of the food preparation area. The emergence of service quality and staff behavior among the significant components that tourists take into account suggests that in general, tourists look for restaurants where staff might be courteous and friendly and where service is of high standard. Tourists also seem to prefer restaurants that prepare tasty dishes of high quality, with fresh ingredients, and that are cooked and served hygienically. Another variable contributing to the selection of one restaurant over another appears to be menu diversity. Tourists seem to look for an adventurous menu that enables them to taste local and interesting food. This is not surprising, as the majority of tourists may view sampling of local dishes as a means of learning more about the local traditions and culture. Sampling of local food might extensively contribute to the discovery of host culture. However, the menu needs to be balanced to cater for different needs. This is because at one end of the need continuum, there might be people who have a desire for familiar food, and at the opposite end of the continuum might be the tourist who wants to try different cuisine.


328


It appears that availability of nutritious dishes may be an important consideration in restaurant selection. The preference of restaurants offering healthy and nutritious food is not surprising because of the growing awareness of the link between diet and disease, which initiated an increased focus on healthy eating habits (Clay, Emenheiser, & Bruce, 1995). This suggests that consumer nutrition attitudes may influence their food choices as well as their choice of restaurants. The findings of the study further indicate that availability of nonsmoking areas and visibility of the food preparation area could lead to a competitive advantage in resort restaurants. An application of the open-kitchen concept, in which customers can see sanitary conditions and how dishes are cooked inside the kitchen, may bolster customers’ confidence in the restaurant and may eliminate the health risk involved in food purchase decisions. The study findings also indicate that tourists consider price and value of dishes when selecting a restaurant. Thus, a right combination of product quality, fair price, and good service may provide a competitive edge for restaurants. As can be seen, contrary to managers’overemphasis on certain service aspects, such as price, the findings suggest that there might be a multitude of elements affecting tourists’ restaurant selection decisions. Apparently, the relative importance of each element differs between segments. For instance, the value seekers were found to be more concerned about product value when making their restaurant selection decision. The service seekers appeared to attach greater importance to quality service when selecting a restaurant. The adventurous-food seekers were found to take the opportunity of sampling new, interesting, and local dishes more into account in their selection decisions. This segment was found not to be particularly concerned with nutritiousness and healthiness of the food. The atmosphere seekers sought restaurants capable of offering a pleasant atmosphere and availability of a good time. The healthy-food seekers were found to be more concerned with availability of healthy food when selecting a restaurant. The existence of different segments, within the tourist dining market, requires restaurant managers to flag up different aspects that would appeal to the targeted segment(s) in their marketing communications. Restaurateurs targeting the Healthy-Food Seekers segment, for instance, might be better off highlighting the healthiness of the dishes and should provide information about food nutrition. Those restaurateurs who identify adventurous-food seekers as their viable customer market should emphasize relatively different elements in their marketing communication. Customers choosing their restaurants may prefer to taste authentic local dishes to learn more about the traditions and culture of the host country. Marketing messages emphasizing value for money, food quality, and hygiene may be an effective strategy for attracting value seekers. Overall, the findings of the study offer some exploratory insights into tourist dining segments and suggest that identification of segments, through selection variables, can promote better marketing efforts, as the focus would be more precise. As we noted earlier, identification of viable dining segment(s), and subsequently the development of appropriate marketing messages, is considered essential in attracting target customers. However, no matter how attractive their advertising may be, restaurateurs are likely to have difficulties in the area of mar-



329

keting and management, unless they develop a sound understanding of the processes that tourists use in service evaluations. Because the dining market is not homogeneous and the nature and relative importance of service or product attributes in the formation of postpurchase judgments might differ across segments, empirical research into segment-specific satisfaction analysis is needed. REFERENCES Ahmed, A. S., Barber, M., & Astous, A. (1998). Segmentation of the Nordic winter sun seekers market. Journal of Travel and Tourism Marketing, 7(1), 39-63. Aldenderfer, M. S., & Blashfield, K. R. (1984). Cluster analysis series: Quantitative application in the social sciences. London: Sage. Almanza, B. A., Jaffe, W., & Lin, L. (1994). Use of service attribute matrix to measure consumer satisfaction. Hospitality Research Journal, 17(2), 63-75. Alzua, A., O’Leary, T. J., & Morrison, M. A. (1998). Cultural and heritage tourism, identifying niches for international travellers. Journal of Tourism Studies, 9(2), 2-13. Auty, S (1992). Consumer choice and segmentation in the restaurant industry. The Service Industries Journal, 12(3), 324-339. Bahn, D. K., & Granzin, L. K. (1985). Benefit segmentation in the restaurant industry. Journal of the Academy of Marketing Science, 13(3), 226-247. Becker-Suttle, B. C., Weaver, P., & Crawford-Welch, S. (1994). A pilot study using conjoint analysis in the comparison of age-based segmentation strategies in full service restaurant market. Journal of Restaurant and Foodservice Marketing, 1(2), 71-91. Bojanic, C. D., & Shea, J. L. (1997). Segmentation for multiunit restaurant operation: Taking location into account when advertising. The Cornell Hotel and Restaurant Administration Quarterly, 38(4), 56-61. Bowen, T. J. (1998). Market segmentation in hospitality research, no longer a sequential process. International Journal of Contemporary Hospitality Management, 10(7), 289296. Calantole, J. R., & Johar, S. J. (1984). Seasonal segmentation of the tourism market using a benefit segmentation framework. Journal of Travel Research, 23(2), 14-24. Clay, M. J., Emenheiser, A. D., & Bruce, R. G. (1995). Healthful menu offerings in restaurants: A survey of major U.S. chains. Journal of Foodservice Systems, 8, 91-101. Clow, E. K., Kurtz, L. D., & Ozment, J. (1996). Managing customer expectations of restaurants: An empirical study. Journal of Restaurant and Foodservice Marketing, 1(3/4), 135-159. Crawford-Welch, S. (1991). Market segmentation in the hospitality industry. Hospitality Research Journal, 14(2), 295-308. Crawford-Welch, S. (1994). Restaurant and foodservice marketing into the 21st century. Journal of Restaurant and Foodservice Marketing, 1(1), 1-19. English, W., Josiam, B., Upchurch, S. R., & Willems, J. (1996). Restaurant attrition: A longitudinal analysis of restaurant failures. International Journal of Contemporary Hospitality Management, 8(2), 17-20. Everitt, B. (1993). Cluster analysis (5th ed.). New York: Edward Arnold. Granzin, L. K., & Olsen, E. J. (1997). Market segmentation for fast-food restaurants in the era of health consciousness. Journal of Restaurant and Foodservice Marketing, 2(2), 1-20.


330


Hair, J. F., Anderson, R., & Black, W. C. (1995). Multivariate data analysis with readings. Englewood Cliffs, NJ: Prentice Hall. Haley, R. (1985). Developing effective communication strategies: A benefit segmentation approach. New York: Wiley. Halstead, D. (1989). Expectations and disconfirmation beliefs as predictors of CS, repurchase intentions, and complaining behaviour: An empirical study. Journal of Consumer Satisfaction/Dissatisfaction and Complaining Behaviour, 2, 17-21. Kara, A., Kaynak, E., & Kucukemiroglu, O. (1997). Marketing strategies for fast-food restaurants: A customer view. British Food Journal, 99(9), 318-324. Kivela, J. J. (1997). Restaurant marketing: Selection and segmentation in Hong Kong. International Journal of Contemporary Hospitality Management, 9(3), 116-123. Lewis, C. R. (1981). Restaurant advertising: Appeals and consumers’ intentions. Journal of Advertising Research, 21(5), 69-75. Lewis, C. R. (1984). The basis of hotel selection. The Cornell Hotel and Restaurant Administration Quarterly, 25(1), 54-69. Loker, E. L., & Perdue, R. R. (1992). A benefit-based segmentation of a non-resident summer travel market. Journal of Travel Research, 31(1), 30-35. Lowenstein, M. V. (1995). Customer retention: An integrated process for keeping your best customer. Milwaukee, WI: ASQC Quality Press. Maddox, N. R. (1985). Measuring satisfaction with tourism. Journal of Travel Research, 23(3), 2-5. Madrigal, R., & Kahle, L. R. (1994, Winter). Predicting vacation activity preferences on the basis of value-system segmentation. Journal of Travel Research, 32, 22-28. Mo, C., Havitz, E. M., & Howard, R. D. (1994). Segmenting travel markets with the International Tourism Role (ITR) Scale. Journal of Travel Research, 33(1), 24-31. Norusis, J. M. (1994). SPSS professional statistics 6.1. SPSS Inc. Nunnaly, J. C. (1967). Psychometric theory. New York: McGraw-Hill. Oh, M., & Jeong, M. (1996). Improving marketers’ predictive power of customer satisfaction on expectation-based target market levels. Hospitality Research Journal, 19(4), 65-85. Parasuraman, A., Zeithaml, A. V., & Berry, L. L. (1988, Spring). Servqual: A multiple item scale for measuring consumer perceptions of service quality. Journal of Retailing, 64, 12-40. Punj, G., & Stewart, W. D. (1983). Cluster analysis in marketing research: Review and suggestions for application. Journal of Marketing Research, 20, 134-148. Reid, D. R. (1983). Foodservice restaurant marketing. London: CBI. Richard, D. M., & Sundaram, D. S. (1994). A model of lodging repeat choice intentions. Annals of Tourism Research, 21(4), 745-755. Shank, D. M., & Nahhas, F. (1994). Understanding the service requirements of the mature market. Journal of Restaurant and Foodservice Marketing, 1(2), 23-43. Shoemaker, S. (1998). A strategic approach to segmentation in university foodservice. Journal of Restaurant and Foodservice Marketing, 3(1), 3-35. Singh, J. (1990). A typology of consumer dissatisfaction response styles. Journal of Retailing, 66(1), 57-99. Smith, J. L. S. (1995). Tourism analysis: A handbook (2nd ed.). Essex, UK: Longman.



331

Swinyard, R. W. (1977). A research approach to restaurant marketing. The Cornell Hotel and Restaurant Administration Quarterly, 17(4), 62-83. Swinyard, R. W., & Struman, D. K. (1986). Market segmentation, finding the heart of your restaurant’s market. The Cornell Hotel and Restaurant Administration Quarterly, 27(1), 89-96. Weaver, A. P., McCleary, W. K., & Jinlin, Z. (1994). Segmenting the business traveller market. In J. V. Harsell (Ed.), Tourism an exploration (pp. 137-147). Englewood Cliffs, NJ: Prentice Hall. Williams, A. J., DeMicco, J. F., & Kotschevar, L. (1997). The challenges that face restaurants in attracting and meeting the needs of the mature customer. Journal of Restaurant and Foodservice Marketing, 2(4), 49-62. Woo, K. (1998). Using quality perceptions to segment customers in services. Marketing Intelligence and Planning, 16(7), 418-424. Yoon, J., & Shafer, L. E. (1997). An analysis of sun-spot destination resort market segments: All-inclusive package versus independent travel arrangements. Journal of Hospitality and Tourism Research, 21(1), 141-159.

Submitted November 30, 1999 First Revision submitted on November 24, 2000 Final Revision submitted on February 2, 2001 Accepted on February 12, 2001 Reviewed Anonymously Atila Yüksel (corresponding author) (e-mail: [email protected]), Ph.D., is a faculty member of the School of Tourism and Hospitality Management, Adnan Menderes University (Turkey); Fisun Yüksel (e-mail: [email protected]) is a doctoral candidate at the Centre for Tourism, Sheffield Hallam University (Sheffield, UK).
