VACATION LIFESTYLE AND TRAVEL BEHAVIORS

Abstract

In marketing research, tourism scholars have often developed a variety of scales and integrated them into segmentation studies. Since vacation lifestyle variables, which could be utilized as a base for marketing research, have not received much attention from market researchers, this article attempts to further purify the extant scale and use Taiwanese travelers as a case study. The study survey was conducted of 443 tourists staying at least one night at the Ken‐Ting National Park in Taiwan from April to May, 2002. This study collected 420 useful samples. Three groups of users merged from the cluster analysis: (a) Family Oriented Travelers, (b) Social Oriented Travelers, and (c) Stylish Travelers. The study shows that the differences in trip characteristics are the frequencies of domestic travel, the frequencies of international travel, the hotel preferences, and the average expenditure on accommodations. The variations in demographic characteristics relate to age, family life cycle, education, and household income. This study on vacation lifestyle subsequently renders a new assessment tool to tourism professionals in Taiwan to grasp the lifestyle profile of their customers and develop strategies to entice the travel market.

Keywords:

• Vacation lifestyle
• market segmentation
• travel behavior

INTRODUCTION

Tourism marketing research focuses a good deal on exploring travel behavior, structure demand, and destination choice (Zins, 1998). With rapidly growing consumption patterns and increasingly competitive consumer markets, scholars have constantly conducted a battery of studies to supply marketing information needs (Yaman & Shaw, 1998). Marketing managers and service providers recognize that there are highly differentiated markets of customers and that consumers are basically heterogeneous in their demands and wants. If consumers could be classified on the basis of their behavior, some predictive ability for the categories will have been shown (Perreault, Darden, & Darden, 1977). Thus, from service management points of view, it is pivotal to examine the degree of the similarities and variations in travel demands between/among different groups of customers. In order to successfully tailor the needs of distinct groups of customers, segmentation has been employed to identify particular characteristics of tourists (Hudson, 2000). In marketing research, market segmentation as a tool of marketing management has been widely discussed in the scientific literature. Market segmentation is the aggregation of relatively heterogeneous individual consumers into small groups, where the members of each group or segment are relatively similar, yet different from other groups (Schewe & Calantone, 1978). In tourism, market segmentation as a method of marketing management has caught up with other industries in terms of degree of complexity and application of complicated methodologies (Baker, Hozier, & Rogers, 1994).

Segmentation and typology construction have been developed and used by tourism scholars to grasp and predict traveler behaviors. In the past few decades, tourism marketing researchers have carried out various segmentation studies by using observable variables or unobservable variables such as demographic variables (Woodside & Pitts, 1976), travel activities (Tsang, Choi, & Tsang, 1999), benefits sought (Calantone & Johar, 1984; Yannopoulos & Rotenberg, 1999), and motivation (Crompton & McKay, 1997; Nicholson & Pearce, 2000). However, in order to effectively segment tourists in groups to understand and predict travel behaviors, a new construct—lifestyle—has been developed and utilized in marketing research. Lifestyles are employed as an approach of market segmentation that offers marketers particular advantages (Lawson et al., 2000). Plummer (1974) indicated that lifestyle concept combines the merits of demographics with the richness and dimensionality of psychological characteristics and depth research. In addition, compared to most market segmentation methods, lifestyles could give added insight for the development and management of the marketing mix (Lawson et al.). Lifestyle is used to segment the market because it provides a wide, everyday view of customers (Plummer). The use of the lifestyle variable provides an understanding of tourists by looking at their activities, interests, opinions. Knowing how an individual behaves in one area of his life may allow us to predict how he will act in others (Feldman & Thielbar, 1975). Through applying the lifestyle approach to understand the tourist as a customer, the marketers could develop more efficient and effective marketing programs (Schewe & Calantone, 1978). Further, tourism scholars have begun to apply lifestyle variables as segmentation criteria to explore travel behaviors. An overview of the application in tourism research is given by several authors (e.g., Darden & Perrreault, 1975; Gladwell, 1990; Silverberg, Backman, & Backman, 1996; Zins, 1998). Zins pointed out that the consumption of tourism products and services follow different conditions than other consumer goods and products. The analysis of consumption of tourism products should develop proper specific lifestyle measurements, such as the vacation lifestyle variable, to reveal travel behaviors. In addition, day‐to‐day lifestyle is contrasted with the pattern of lifestyle while on vacation (Zins). The use and application of the vacation lifestyle construct as an effective segmentation criterion could provide adequate delineation of customer behaviors. Inspection of groups resulting from vacation lifestyle classification should provide valuable insights for marketing researchers into how leisure time is spent and what type of vacationers exist. For tourism studies, marketing researchers should use vacation‐specific lifestyle measurements, which are vacation lifestyle variables, to distinguish the differences of tourist consumption behaviors.

In conclusion, tourism studies (Cheng & Lio, 2001; Darden & Perrreault, 1975; Plummer, 1974; Zins, 1998) suggested that psychographic variables were superior to demographic variables. Vacation lifestyle attribute is a very useful assessment of tourist psychology for market segmentation research and could render practical information of tourist behaviors in developing marketing strategies. While a majority of the vacation lifestyle studies have been conducted in the Western countries, launching an investigation on vacation lifestyle in an Asian country—Taiwan—may add new cross‐cultural prospects into current tourism marketing literature.

LITERATURE REVIEW

This section discusses the development of the construct of lifestyle and the relationship of lifestyle to market segmentation. In addition, this section also explores the reason for choosing lifestyle as a variable to detect the differences of travel behaviors. Further, a review of marketing research demonstrates relationships among lifestyle and vacation behaviors. The review is designed to determine the relations found in tourism research and how the findings from these prior studies are applied to travel vacation behavior. Finally, this section examines the methods of lifestyle measurement and reviews how previous researchers have applied lifestyle measurements in tourism studies.

Lifestyle Segmentation

The study of lifestyle patterns and their relationships to marketing was introduced by William Lazer in 1963. Feldman and Thielbar (1975) pointed out three key aspects of the lifestyle concept: (a) lifestyle is a group phenomenon; (b) lifestyle pervades many aspects of life; (c) lifestyle implies central life interests. During the 1970s, the methods of measuring lifestyle patterns and their relationship to consumer behaviors were developed and purified. The concept of lifestyle has been useful for marketing and advertising planning (Plummer, 1974). Kucukemiroglu, Harcar, and Spillan (2006) indicated that lifestyle patterns could be used to identify the consumer market segments, and that lifestyle dimensions influence consumer behaviors. Previous research indicated that lifestyle variables can provide more effective and efficient marketing information about a tourist as a customer than demographic or socio‐economic variables (Plummer; Abbey, 1979; Zins, 1998). Researchers conducting studies regarding the provision of more marketing information have used the lifestyle variable to better understand tourists as a consumer (Gladwell, 1990). Further, Abbey stated that lifestyle measurements are superior to demographic variables in explaining travelers' preferences. Mayo and Jarvis (1981) found that the lifestyle characteristics reflect personality and stated that the study of lifestyle could explain more about consumers (tourists). According to early studies (Woodside & Pitts, 1976) regarding the effects of consumer lifestyle, lifestyle information may be more important than demographic variables in the prediction of travel behavior. As the result of their study, Woodside and Pitts advocated that tourism marketing research should use lifestyle data as a marketing segmentation tool. Over the past several decades, there has thus been growing recognition that lifestyle could be utilized as an effective segmentation base to provide adequate delineation of tourist market segments.

The Measurement of Lifestyle and Its Relationship to Tourism Marketing

The application of lifestyle segmentation could provide a unique and important view of the tourism market. The concept of lifestyle is defined as “patterns in which people live and spend their time and money reflecting people's activities, interests, and opinions (AIO)” (Blackwell, Miniard, & Engel, 2001, p. 219). The lifestyle variable measures people's activities, in terms of (a) how they spend their time; (b) their interests, or how they place importance on their immediate surroundings; and (c) their opinions in terms of their view of themselves and the world around them (Plummer, 1974). Lifestyle segmentation begins with people instead of products and classifies customers into different lifestyle types, each characterized by a unique style of living based on a wide range of activities, interests, and opinions (Plummer). The lifestyle measurements attempt to provide an in‐depth understanding of consumers based on their way of living (Schewe & Calantone, 1978). In tourism research, Darden and Perreault (1975) initiatively used AIO measurement to investigate the relationship between vacation behaviors and lifestyles. They suggested that lifestyles are the genesis for many vacation behaviors and additionally pointed out that lifestyle differences are more influential in determining vacation behaviors. Later, Perreault, Darden, and Darden (1977) used leisure literature and focused group interviews to generate a pool of vacation lifestyle statements that reflect vacation AIOs, developing 28 important dimensions of vacation lifestyles to classify tourists. In addition, Hawes (1988) used 33 AIO statements to clarify the travel lifestyle profiles of women over 50 years of age and pointed out that the travel lifestyle variable could be useful to distinguish media preference patterns among different subgroups of older women. Gladwell (1990) applied activities, interests, and opinions (AIO) statements to identify the types of state park inn user and found three travel segments including knowledgeable travelers, budget‐conscious travelers, and travel planners; he also concluded that differences exist between user groups in vacation behaviors and sources of vacation information. Likewise, Silverberg et al. (1996) converted 46 lifestyle variables into six benefit‐sought factors which are utilized as the criteria to find subgroups of natured‐based travelers and pointed out that lifestyle could be viable for tourism marketers to predict which benefits travelers are seeking. Further, Backman et al. (1999) used 46 AIO measurements to categorize senior nature‐based travelers and to determine whether differences and similarities between the groups exist. The result of the study stated that senior nature‐based travelers are looking for different packages of benefit; in sum, lifestyles are very useful to marketers of nature‐based travel products.

Not only can lifestyle segmentation be important in travel marketing; but, as Zins (1998) proclaims, psychographic concepts (e.g., lifestyle) also supposedly offer the highest predictive power. He studied consumers' hotel choice patterns by integrating lifestyle with values, vacation style, and benefits, finding that all four psychographic constructs create both direct and indirect effects on hotel choice. Furthermore, Chandler (2004), based on the concept of lifestyle and personal values, compared visitor profiles at heritage tourism destinations in North Carolina and identified both homogeneous and heterogeneous traits of heritage tourists concerning their lifestyle and activity preferences. Similarly, Wilson and Thilmany (2006) segmented Colorado's golfers by a lifestyle variable and also developed a set of profiles to assist the development of promotional strategies for the golf business. Moreover, Lee and Sparks (2007) employed AIO instruments to investigate the differences between Korean Australians and Koreans in Korea and found that the travel experiences of the two groups were affected by their respective cultural experiences. Cheng and Lio (2001) conducted a preliminary study on the construction of Taiwanese vacation Lifestyle Scale and resulted into a seven‐factor dimension of vacation lifestyle. Thus, lifestyle could be a useful variable for tourism marketers, not only to classify customers and examine the differences of travelers, but also to predict tourist consumption behavior by looking at their activities, interests, and opinions which are based on their ways of living.

The applications of the lifestyle concept could classify tourists into different groups and identify the demands and needs between groups. In the early stage of lifestyle research, AIO (Activities, Interests, and Opinions) constitutes the measurement of lifestyle. By examining consumers' activities, interests, and opinions, lifestyle segmentation could provide a unique and important view of markets, characterizing customers based on their ways of living.

PURPOSE OF THE STUDY

Due to the new labor law in Taiwan that gives full‐time workers a 2‐day break (Saturday and Sunday) per week—compared to a 1½‐day break beforehand—domestic tourism is booming and looks very rosy. The government has noticed a vital role of tourism expansion in economic development and launched a national development plan—the 2002 Doubling Tourist Arrivals Plan (DTAP)—to promote tourism. In addition, the promotion of tourism is also a major concern of many private corporations and service providers. In order to provide appropriate tourism products and services, marketing managers should identify target markets which are core groups of customers whom the service providers desire to attract. For the tourism researcher and market manager, the first step to effectively understand the customers is to divide the total market into relatively homogenous groups with similar product needs or service interests. Once these market segments are identified, the researchers and practitioners could design appropriate service blueprints and marketing strategies to provide better services for target markets. As per the literature, the lifestyle variable has been proven as a vital tool in assessing consumer needs in Western countries; this study attempts to further explore it by introducing a case study of an Asian country—Taiwan. This study about vacation lifestyle subsequently could render a new assessment tool for industrial professionals to grasp the lifestyle profile of their customers and develop strategies to attract the target markets. The results of this study also provide a perspective on the Taiwanese for understanding vacation lifestyle and travel behavior. Therefore, three research questions are proposed: (a) How may mutually exclusive clusters rooted in a lifestyle variable could be identified?; (b) What are the socio‐demographic and trip characteristics of respondents from each cluster?; (c) Do trip characteristics and socio‐demographic traits differ between/among the resultant vacation lifestyle clusters?

METHODOLOGY

The study population consisted of individuals, 18‐years‐old or above, taking their domestic trip within the island of Taiwan. According to the Travel Survey of the Taiwanese (Taiwan Tourism Bureau Ministry, 2001), the Ken‐Ting National Park was one of the top‐three most visited domestic tourist spots in Taiwan (Taiwan Tourism Bureau Ministry). Therefore, the study conducted surveys on visitors to the Ken‐Ting National Park. Consequently, this study surveyed 443 tourists staying at least one night at the Ken‐Ting National Park in Taiwan from April to May, 2002. However, this study entailed 420 useful samples and the response rate of this study was 36.91%.

A questionnaire was designed to elicit information from respondents about vacation activities, interests, opinions, and travel behaviors, as well as socio‐economic characteristics. In developing the research variable of vacation lifestyle, this study modified the Taiwanese vacation lifestyle instruments of Cheng and Lio's (2001) study that mirrored the AIO scales of Perreault, Darden, and Darden's (1977), Gladwell (1990), and Zins (1998). The lifestyle attributes by Cheng and Lio—along with new attributes acquired from in‐depth interviews with tourism scholars, travel agents, and tourism managers in Taiwan—were screened and purified by the research team. A multi‐attribute measurement reflecting Taiwanese vacation characteristics was developed afterwards. The face validity of the lifestyle measure was further enhanced through a review by a group of tourism professors and graduate students. A set of 37 vacation lifestyle items was used in this study and the AIO measurements were based on a Likert‐type scale from strongly disagree to strongly agree (Definitively disagree  =  1 to Definitively agree  =  6). The following examples are representative of the AIO statements to which respondents were asked to respond: I prefer to camp on vacation; I take time to enjoy nature on vacation; on vacation, it is most important to relieve the stress and strain of normal life; I prefer to take several short vacations rather than a few larger ones; I spare no expense in making vacation travel arrangements; and vacations are a chance to develop close relationships with colleagues.

DATA ANALYSIS

Demographics and Trip Profiles

The demographic traits of the samples were as follows: 62.6% of the respondents were male; the mean age of the sample size was 35‐years‐old; nearly 74% had at least a college degree; the family life cycle of the respondents were 30% young single and 32% family with preschool children (Full Nest I); and the mean household income of the respondents was Taiwan Dollars (TWD) 60,000–100,000. Trip profiles of the respondents indicate that the average frequency of domestic trips for these travelers was 4.2 times per year. In addition, these travelers took an average of one or more international trips in the past one year. The average expenditure of accommodations of these tourists is TWD 3,113 per day. Nearly 48% of the respondents usually chose luxury hotel accommodations. They primarily obtained sources of vacation information from newspapers/magazines, and the next frequently used sources were family and colleagues. The respondents were most often accompanied by family members on their trips.

Lifestyle Market Segments

The 37 original vacation lifestyle (AIO) items were subjected to factor analysis using principal components analysis with orthogonal rotation (varimax) to reduce a large number of variable sets of underlying factors. Following the deletion of some variables due to cross loadings or poor fit to the factor solution, the factor loading of variables higher than 0.5 were retained for further research. The Kaiser‐Meyer‐Olkin Measure of Sample Adequacy was conducted to ensure the adequacy for factor analysis. The Kaiser‐Meyer‐Olkin Measure for the data was 0.73, which suggests these data are suitable for analysis as it exceeds the recommended value of 0.60 (Coakes & Steed, 1999; Kaiser, 1974). Furthermore, the Bartlett's Test of Sphericity (Bartlett, 1954) reached statistical significance (p < .001). In addition, after the initial factor analysis was performed, the 37 AIO items were reduced to 17 items and were grouped into five dimensions that exhibit the best sample structure: First Class, Family Orientation, Social Orientation, Outdoor Recreation, and Relaxation. Moreover, the five dimensions used in subsequent analysis explained 56.99% of the variance, with resulting eigenvalues ranging from 1.77 to 2.17. Nunnallly (1967) pointed out that the Cronbach's alpha coefficient at 0.5 or 0.6 is modest reliability that will suffice for basic research. Reliability analysis was performed to test internal consistency of each factor obtained from factor analysis. Cronbach's alpha coefficients ranged from 0.59 to 0.67. The overall Cronbach's alpha coefficient of five vacation lifestyle dimensions was 0.70. Table 1 presents the result of the factor analysis including AIO items, factor loading, Cronbach's alpha, eigenvalues, and explained variance.

Table 1. Factor Analysis of Vacation Lifestyle Variable

Dimensions and AIO Items	Factor Loading	Cronbach's α	Eigenvalue	Explained variance %
Relaxation		0.67	2.17	12.79
On vacation, most important is to relieve stress and strain of normal life	.76
On vacation, I look forward to relieving stress and being renewed	.69
I prefer vacations that help me relax	.69
I like to visit the area that could enjoy nature view	.63
Outdoor Recreation		0.65	1.96	11.53
I prefer to camp on vacation	.77
I like to barbecue on vacation	.69
I like hiking and climbing on vacation	.69
First Class		0.59	1.93	11.37
I would not vacation in an area where first class accommodations were not available	.72
I spare no expense in making vacation travel arrangements	.71
On vacation, good food and beverage are important for me	.69
I like to visit the area that has health spa facilities	.51
Family Orientation		0.67	1.85	10.88
Our family travels together quite a lot	.81
Vacation should be planned for the children	.74
I like to gather photos that family travel together	.69
Social Orientation		0.64	1.77	10.44
On vacation, social interaction is important	.81
Vacations are the chances to develop close relationship with colleagues	.76
I usually participate in an association trip	.56
Total		0.70		56.99

In order to segment respondents into groups with similar vacation‐specific lifestyles, cluster analysis was performed on the factor score to group individuals according to vacation lifestyle. The use of factor scores derived from factor analysis is widely used as input for cluster analysis (Singh, 1990). As Johnson (1967) observed, “there is an increasing interest in identifying groupings or clustering of the objects under study that best represent certain empirically measured relations of similarity” (Gladwell, 1990, p. 17). Therefore, cluster analysis could be helpful in placing objects into relatively homogeneous groups (Gladwell, 1990). For purposes of this study, K‐mean cluster analysis was conducted to adjust the cluster identification process and the best number of clusters was determined by examining a series of solutions. All five factor scores made a significant contribution to differentiating the three groups (p < .001). It was evident upon the analysis that three clusters provide the best solution for partitioning the respondents into homogeneous groups in this study. Therefore, three clusters were classified into groups entitled “Family Oriented Travelers,” “Stylish Travelers,” and “Social Oriented Travelers.” Table 2 illustrates the final cluster solution based on a derived factor score of vacation lifestyle and the percentage of each cluster. A description of the three classifications follows.

Table 2. Profile of Vacation Lifestyle Segments

Vacation Lifestyle Dimensions	Family Oriented Travelers (I)	Stylish Travelers (II)	Social Oriented Travelers (III)	F Ratio		Scheffe Post‐Hoc Test^b
Relaxation	−.44557	.17584	.31787	26.06	***	II > I, III > I,
Outdoor Recreation	−.19459	.21820	.02850	5.53	**	II > I
First Class	−.53347	.91779	−.17119	105.4	***	II > I, II > III, III > I
Family Orientation	.61848^a	−.03412	−.60502	72.9	***	I > II, I > III, II > III
Social Orientation	−.15536	−.65919	.67293	80.9	***	I > II, II > I, III > II
Total respondents 395 (100%)	144 (36.5%)	110 (27.8%)	141 (35.7%)
Note. ^aThe cluster descriptions are based on factor scores, which were derived from vacation lifestyle data and have a mean of zero, and standard deviation of one.^bUse Scheffe post‐hoc test to examine the significant difference of lifestyle dimensions in the 3 groups (p<.05).**p < .01; ***p < .001.

Group 1: The Family Oriented Travelers

The people in this group comprised 36.5% of the total respondents. They exhibited a strong interest in travel with family members and they think family vacations were happier. The individuals in this group demonstrated that their family travels together quite a lot and likes to accumulate families' vacation photos as their recollection. They think the vacation should plan for the children. They tended to like family‐oriented vacations but their major vacation interests were more economy‐oriented than the other two groups.

Group 2: The Stylish Travelers

Representing 27.8% of the total sample, these people were not concerned with the expenses of their vacation. They were interested in finding good food and beverage no matter how much it will cost. When they plan a trip, they will consider whether the vacation areas have luxury hotels or not. The people in this group desired to stay in hotels that have health spa facilities. This group had the lowest cluster score on social orientation and it showed that they do not prefer to vacation for social interaction.

Group 3: Social Oriented Travelers

This group of individuals made up 35.7% of the total sample. They were characterized by their strong interest in social interaction. The people in this group preferred to vacation with the colleague groups. Their interests in travel were reflected in the fact that vacations were the chances to develop close relationships with colleagues. They like to join trips organized by association groups. The cluster score also showed that they were seeking stress relief on vacation but they did not incline to travel together with family.

The Social‐Demographic Profiles of the Lifestyle Market Segments

Results of the ANOVA and chi‐square analysis showed there were significant differences among the three groups on social‐demographic characteristics. For the social‐demographic variable, results revealed that there were significant differences across the clusters on age, monthly household income, marital status, and family life cycle.

In addition, after a significant difference was found by the ANOVA, a Fisher's (1935) LSD multiple‐comparison procedure was performed to detect the mean difference between the groups. The LSD test showed that Family Oriented Travelers (mean  =  36.31) were significantly older than Stylish Travelers (mean  =  34.40) and were also significantly older than Social Oriented Travelers (mean  =  33.90). Moreover, the result of the LSD test showed that the household income of Stylish Travelers (mean  =  106,415) was statistically significantly higher than the other two groups—Family Oriented Travelers (mean  =  92,517) and Social Oriented Travelers (mean  =  94,857). In terms of education in years, the Stylish Travelers (mean  =  14.90) were also significantly higher than Social Oriented Travelers (mean  =  14.10). In general, the data supported the concept that vacation lifestyle is pervasive. The socio‐economic characteristics were substantially consistent with the group's attitudes toward an interest in vacation activities. Table 3 and 4 present the social‐demographic differences among the groups.

Table 3. Differences in Age, Income, and Education Among the Segments

Social‐demographic	Family Oriented Travelers	The Stylish Travelers	Social Oriented Travelers	F Ratio	LSD Test
Age	36.30 (6.8)^a	34.40 (6.7)	33.90 (7.8)	4.49**	I > II, I > III
Monthly household income (TWD)	92,517 (39,565)	106,415 (41,933)	94,857 (38,896)	4.04**	II > I, II > III
Education in years	14.4 (1.02)	14.9 (1.04)	14.1 (0.95)	2.7*	II > III
Note. 1 U.S. Dollar  =  30.53 Taiwan Dollar as of May 28, 2008.Use LSD Test to examine the significant difference in the 3 groups (p < .05).^aStandard deviation.**p < .05; *p < .10.

Table 4. Differences in Gender, Marital Status, Family Life Cycle, and Occupation Among Segments

Social‐demographic	Family Oriented Travelers	The Stylish Travelers	Social Oriented Travelers	χ ^2 Test
Gender				0.63^a
Male	90 (62.5)^b	72 (65.5)	84 (59.6)
Female	54 (37.5)	38 (34.5)	57 (40.4)
Marital Status
Married	119	77	77	31.90**
Unmarried	23	33	64
Family life cycle
Young single or Young couple (no children)	34 (21.9)	43 (39.1)	72 (51.4)	31.67**
Full nest I (family with pre‐school children)	53 (37.1)	38 (34.5)	30 (21.4)
Full nest II (school‐age children)	49 (34.3)	25 (22.7)	29 (20.7)
Full nest III (children over 18 but nondependent) and Empty nest (children all dependent)	11 (7.7)	4 (3.6)	9 (6.4)
Occupation				12.78
Managers and professionals	39 (31.0)	25 (29.8)	24 (20.2)
Technicians and Associate professionals	43 (34.1)	26 (31.0)	34 (28.6)
Clerks and salesmen	23 (18.3)	21 (25.0)	40 (33.6)
Machine operators and Skilled agricultural and fishery workers	12 (9.5)	9 (10.5)	17 (14.3)
Elementary workers	9 (7.1)	3 (3.6)	4 (3.4)
Note. ^aPearson chi‐square ratio.^bSample size and percentage of cluster.**p < .05.

The Family Oriented Travelers

The individuals in this group tended to be older than those in the other two groups. In addition, this group displayed a relatively larger percentage of married individuals than the other two groups. The people in this group had more pre‐school children and school‐age children than those in other two groups. They had the lowest average monthly household income among three groups. These characteristics may help explain why this group traveled together with family quite a lot and why this group was more economy‐oriented in their travel vacations. These findings support the name “Family Oriented Travelers” and reinforce their characteristic of strong desire for travel together with family and more concern about the cost of their vacation.

The Stylish Travelers

The stylish travelers had higher monthly household income levels than the other two groups. They also had the highest education levels among the three groups. In addition, these individuals had moderate average age among the three groups. These results showed that this group was more financially optimistic than the other two groups and also reflected why this group is not concerned about the expense in planning a vacation. In addition, based on their financial optimism, these individuals had strong interests in seeking delicacies. These findings seem consistent with the name “Stylish Traveler” in their vacations.

The Social Oriented Travelers

These individuals were characterized as having moderate household incomes and education levels. The average age in this group was significantly lower than the other two groups, which may account for why these individuals like to vacation in groups and interact with others. In addition, the result of the family life cycle showed that they are a larger percentage of young singles or young couples, who do not have children, among the three groups. It may also explain that the social oriented traveler had little or no interest in family oriented vacations. Overall, these patterns of response reflect their vacation interests and the findings support the name “Social Oriented Travelers.”

The Vacation Travel Behavior Profiles of the Lifestyle Market Segments

In addition to the demographic and preferential characteristics explained above, there were also differences among the three groups based on vacation travel behaviors. For the results of vacation travel behaviors, there were significant differences across the three groups: (a) frequencies of international travel, (b) the average expenditure of accommodations per night, and (c) the hotel choice preference. Fisher's test found the significant pair mean differences of international travel frequencies that Stylish Travelers (mean  =  1.12) and Social Oriented Travelers (mean  =  1.08) were higher than Family Oriented Travelers (mean  =  0.72). Furthermore, Stylish Travelers (mean  =  3,558) spent significantly more on the average expenditure of accommodation than Family Oriented Travelers (mean  =  3081) and Social Oriented Travelers (mean  =  2,773). Tables 5 and 6 present the group mean and the percentage of cluster for five vacation travel behaviors. An overall test of significance showed that these measures provided significant discriminations among the three vacation lifestyle segments.

Table 5. Differences in Frequencies of Domestic and Foreign Trips, and Expenditure of Accommodation Among Segments

Vacation Travel Behavior	Family Oriented Travelers	The Stylish Travelers	Social Oriented Travelers	F Ratio	LSD Test
Frequencies of international travel	0.72 (0.96)^a	1.12 (1.79)	1.08 (1.35)	3.46**	II > I, III > I
Frequencies of domestic travel	5.89 (5.02)	5.11 (4.15)	4.89 (4.27)	1.88
The average expenditure of accommodation per night (TWD)	3,081.6 (1336.6)	3,558.3 (1401.30)	2,773.9 (1297.1)	10.461**	II > I, II > III
Note. According to http://finance.yahoo.com, the exchange rate is 1 U.S. Dollar  =  32.28 Taiwan Dollar.^aStandard deviation.**p < .05.

Table 6. Differences in Hotel Preference, and Vacation Information Among Segments

Vacation Travel Behavior	Family Oriented Travelers	The Stylish Travelers	Social Oriented Travelers	χ ^2 Test
Hotel preferences				30.512^a
Luxury/resort hotel	56^b (40.9)	74 (69.8)	52 (38.8)
Moderate Hotel	16 (11.7)	11 (10.4)	14 (10.4)
Economy hotel or B&B	65 (47.4)	21 (19.8)	68 (50.7)
Vacation source of information
Newspaper and magazine	47 (35.3)	11 (22.4)	45 (32.6)	5.91
Television program	11 (8.3)	8 (16.3)	14 (10.1)
Internet information	41 (30.8)	15 (30.6)	35 (25.4)
Friends and relatives	34 (25.6)	15 (30.6)	44 (31.9)
Note. ^aPearson chi‐square ratio.^bSample size and percentage of cluster.**p < .05.

The Family Oriented Travelers

Family oriented travelers take the fewest international trips among three groups but they are more likely to have higher frequencies of domestic travel. The majority of respondents sought vacation information from the newspapers and magazines. The type of preferred accommodation was economy hotel. The findings of these results appear congruent with an economic orientation in their vacation activities and reflect that they did not have much budget for foreign vacations.

The Stylish Travelers

The respondents in the stylish travelers cluster were substantially higher than the other two groups in frequencies of international vacation. The respondents in this group sought the source of vacation information mostly from the Internet or their friends and relatives. The average expenditure of accommodation in this group was significantly higher than the other two groups. In addition, the result of hotel choice preference showed that the stylish travelers mostly tend to choose luxury or resort hotels as their accommodations. These findings support the pattern of their vacations, which is that they are not concerned about the expenses of their vacations.

The Social Oriented Travelers

The social oriented travelers had the fewest frequencies of domestic travel and moderate frequencies of international vacation. In terms of the average expenditure of accommodation, they were significantly the lowest group among the three groups. The respondents of this group significantly tend to choose economy hotel or B&B as their accommodations among the three clusters. The vacation sources of information primarily came from newspapers/magazines and friends.

CONCLUSION

This research attempts to purify the vacation lifestyle scale and further utilizes it as the base to classify leisure travelers. The results of this research found five vacation lifestyle dimensions among Taiwanese travelers: Relaxation, First Class, Family Orientation, Social Orientation, and Outdoor Recreation. In comparison with prior studies, the above lifestyle dimensions are identical to the work of Cheng & Lio (2001). In respect to the cluster analysis, vacation lifestyle dimensions could be used to develop a classification to identify travelers within three distinct groups: “Family Oriented Travelers,” “Stylish Travelers,” and “Social Oriented Travelers.” A further comparison with social‐demographics revealed that the three groups have significant differences in age, monthly household income, marital status, and family life cycle. The results show that Family Oriented Travelers have the lowest income level among the three groups. In addition, they had a relatively larger percentage of married individuals and the highest proportion had pre‐school children or school‐age children. Furthermore, Stylish Travelers had higher income and education level than the other two groups, and it shows this group is more financially optimistic. Moreover, Stylish Travelers are the youngest among the three groups and have the highest proportion of young singles or young couples.

In regard to the travel behavior of the lifestyle segments, the results of comparing vacation behavior among the three groups show that there were significant differences in frequency of international travel, the average expenditure on accommodations per night, and hotel choice preference. The results show that Family Oriented Travelers had higher frequencies of domestic vacation than the other two groups but the lowest frequencies of foreign travel. Besides, Stylish Travelers were more likely to have overseas vacations and tend to choose luxury hotels as accommodation. The average accommodation expenditure was also significantly highest among the three groups. In contrast, Social Travelers prefer to choose Economy Hotel or B&B, spending the least amount of money on accommodations. Overall, these results of comparison of social‐demographic and travel behavior support the classification of vacation lifestyle and are substantially consistent with the group's attitudes toward an interest in vacation activities. These differences appear to be patterned rather than random. The results of the study showed that the modification of previous vacation lifestyle instruments can be useful in helping Taiwan tourism industries better segment the market and explain tourists' trip behavior. Furthermore, the vacation lifestyle construct could be employed by tourism marketers to identify the heterogeneous and homogenous characteristics of distinct groups of domestic tourists, thus making it viable for tourism marketers to predict traveler preference and vacation behavior. In addition, the combination of the vacation lifestyle variable and social‐demographic information could provide better general understandings of travel behavior. The findings from this study imply that separate service products and marketing strategies should be developed for different vacation lifestyle market segments. Service planners may provide different combinations of services consistent with different vacation lifestyle groups and focus their efforts on those targets that will be most important in terms of revenues.

Lastly, the following limitations need to be considered. The first limitation is pertaining to the generalizability issue. Because the samples were collected in one of the most visited destinations, it may not well represent the overall Taiwanese travel public. It is suggested that further study be conducted to cross‐validate the resultant segments. Secondly, it is important to note that some lifestyle attributes were derived from a series of studies based on American culture that may well fit in the current study since the exploratory factor analysis reveals that the subscales of lifestyle measure show reliability scores below .70. Due to the exploratory nature of the study, the size of sample is not sufficient for conducting cross‐validation by using confirmatory factor analysis. The literature suggested that new samples need to be drawn if a confirmatory factor analysis is to cross‐validate the dimensionality suggested by an exploratory factor analysis. For future studies, it is suggested that the scale be validated by larger samples to statistically purify the scale and secure the construct validity. Lastly, the readers may be cautious with the data used in the study that was collected in 2002. Nevertheless, it is expected that the vacation lifestyle is unlikely to be altered significantly since the economic growth rate is rather slow with an average rate of 5.01% from 2002 to 2007 (Taiwan Executive Yuan, 2007). Certainly, it may furnish new marketing insights on the change of vacation lifestyle, as a new market survey will be completed by 2012.