Factors affecting international students’ travel behavior

Article Factors affecting international students’ travel behavior Journal of Vacation Marketing 1–19 ª The Author(s) 2014 Reprints and permission: sagepub.co.uk/journalsPermissions.nav DOI: 10.1177/1356766714562823 jvm.sagepub.com Hanieh Varasteh, Azizan Marzuki and S Mostafa Rasoolimanesh Universiti Sains Malaysia, Malaysia Abstract This article attempted to find out important factors influencing international students’ travel behavior. A total of 409 international postgraduate students studying in five Malaysian research universities (Universiti Putra Malaysia, Universiti Malaya, Universiti Teknologi Malaysia, Universiti Sains Malaysia, and Universiti Kebangsaan Malaysia) participated in this quantitative study through a selfadministered questionnaire. A structural equation modeling–partial least squares using Warp PLS 3.0 was applied to analyze data. The study revealed that a number of demographic characteristics including age, marital status, nationality, and source of finance significantly affect preferred travel activities and preferences. In addition, travel behavior (as a third-order factor) was also affected by age, marital status, nationality, and source of finance. The moderating effect of information source on relationship between nationality and travel behavior has also been identified, with its main function being adjusting the strengths of relationships between nationality and travel behavior. Keywords International students, students’ market, travel activities, travel behavior, travel preferences Introduction Traveling for educational purposes is an ancient phenomenon experienced by the majority of nationalities over the past centuries (Gibson, 1998), and in the 21st century it has become a multibillion dollar industry due to huge numbers of people going outside of their country to study, who are called as international students (Payne, 2009). International students have a great tendency to travel while studying abroad in an effort to better understand the national culture and people, resulting in considerable revenue as well as employment opportunities for the host country (Payne, 2009). Information pertaining to their travel preferences and patterns are important to the host country due to the enormous financial potential and benefits that may accrue from tourists of this type. Without reliable and available information, improvement of this market segment would be impossible and the host country stands to lose enormous potential financial benefit derivable from this type of tourism (Arcodia et al., 2006; Chadee and Cutler, 1996; Kim, 2007; Kim et al., 2006). Travel behavior based on Recker et al. (1986) is generally understood to mean the way of scheduling activities in a particular manner by individuals. Therefore, complex travel behavior stems from complex scheduling activities. Identifying those sets of activity scheduling and decisions implemented by the individual considered as distinctive variables describing tourist preferences (Hu and Morrison, 2002) seems extremely urgent. Some of these variables such as travel preferences have been stated by Pearce (2005), including type of accommodation, preferred destinations, trip purposes, travel arrangements, and travel party. There are similarities between the attitudes expressed by Pearce (2005) and those Corresponding author: Hanieh Varasteh, School of Housing, Building and Planning, Universiti Sains Malaysia, Penang, Pulau Pinang 11800, Malaysia. Email: hani.varasteh@gmail.com Downloaded from jvm.sagepub.com at Universiti Sains Malaysia on December 15, 2014 2 Journal of Vacation Marketing described by Xu et al. (2009) who identified different attractions, activities, accommodations, and information sources as the travel behavior’s variables. A review of the literature pertaining to the student market revealed that although some studies have highlighted the importance of student market and examined the travel behaviors of domestic and/or international students, there is still a dearth of research and information on international students’ travel behavior while traveling domestically within their host country (Ryan and Zhang, 2007), particularly in the case of Malaysia. This kind of information is able to comprehensively explain or predict students’ travel decision (Kim et al., 2007), which is important for service providers of this market. Since the number of international students’ enrollment in Malaysia has increased dramatically over the past 10 years on account of various higher education reforms as a way to facilitate the entry of international students into higher education institutions (Yusoff and Chelliah, 2010), identifying this particular essential group’s travel behaviors is considered a crucial issue. Going by the latest statistics, there are more than 90,000 international students currently studying in the numerous institutions of higher learning in Malaysia (MOHE, 2010). The main goal of this article is to find out the important factors influencing international students’ travel preferences and activities, and it further attempts to investigate the relationships between international students’ demographic characteristics and travel behaviors including travel activities and preferences. This study attempted to include all the variables that have been mentioned in the previous studies and test the existing relationships as an integrated model. Information source moderating effect with regard to the relationship between country of origin and students’ travel behavior was also sought for the first time. Consequently, it contributes to the existing literature by developing a general framework for international students’ travel behavior considering previous research paucity. A majority of previous research have focused on international students from limited number of nationalities studying in one university; and owing to restrictions such as convenience, time, and money, respondents have been chosen from one geographical area. As travel behaviors depend on the student’s nationality (Chadee and Cutler, 1996; Field, 1999; Hsu and Sung, 1997; Michael et al., 2004; Weaver, 2003) as well as other variables (Chadee and Cutler, 1996; Hsu and Sung, 1997; Kim and Jogaratnam, 2003), the results of such studies cannot be applied to all international students. Hence, further investigations that segment responses based on nationalities and other variables were needed to address these shortcomings. This article will thus contribute to the existing literature by considering these variables in order to develop a general framework for international students’ travel behavior. A total of 409 international postgraduate students studying in five Malaysian research universities (Universiti Putra Malaysia (UPM), Universiti Malaya (UM), Universiti Teknologi Malaysia (UTM), Universiti Sains Malaysia (USM), and Universiti Kebangsaan Malaysia (UKM)) responded to a selfadministered questionnaire, and a structural equation modeling–partial least squares (SEMPLS) using Warp PLS 3.0 was applied to analyze data. Literature review Marketing theory suggests that businesses applying a market segmentation approach can develop their organizational performance (Kotler, 1997) by gaining excellent understanding of customers, leading to suitable marketing programs (Dibb et al., 2002). Market segmentation divides the target market into smaller groups to evaluate the target groups’ specific wants, needs, and behaviors (Jones et al., 2005; Kotler et al., 2002). Although there is no precise agreement on how the market should be segmented, often several segmentations will meet Kotler’s criteria. He divides market segmentation variables into four major areas, namely, geographic, demographic, psychographic, and behavioristic (Kotler, 1997). The validity of using demographic variables in segmentation studies has been supported over the years. Beane and Ennis (1987) emphasized demographic segmentation as the most prevalent form of market segmentation, possibly because consumers are positioned on clear scales of measurement, which are easily understood. The information is usually very easily interpreted, relatively easily gathered, and very easily transferable and collected from one research to another. Bass et al. (1968) also made good use of demographics in describing light and heavy users. Blattberg et al. (1976) stated that buyer behavior is closely related to their demographics. Frank et al. (1972), Brayley (1990), Kotler et al. (2002), Jones et al. (2005), Arcodia et al. (2006), Downloaded from jvm.sagepub.com at Universiti Sains Malaysia on December 15, 2014 3 Varasteh et al. Glover (2011), Bahng and Kincade (2014), and Kline et al. (2014) also discussed various demographic characteristics and their use in market segmentation. Students’ travels represent a distinct market with specific needs and preferences (Chadee and Cutler, 1996), and it is widely agreed by academics that the international student market needs to be further segmented into different clusters due to various characteristics that affect students’ preferences. Although there is no overall agreement regarding exactly how the market should be segmented, demographic characteristics often prove to be a good way to describe this identified segment’s desires, as recent research suggested that marketers must consider the influence of nationality, age, background, gender, and other classifications and construct their marketing strategies accordingly (Field, 1999; Oppermann, 1994; Sussmann and Rashcovsky, 1997). It is also indicated by Arcodia et al. (2006) and Chadee and Cutler (1996) that travel behaviors and preferences depend on the student’s nationality as well as other variables because each nationality has different travel characteristics and preferences. Supporting these researchers, Hsu and Sung (1997), who performed an exploratory study to examine travel behaviors of international students at a Midwestern American university, stated that travel preferences could vary because of the differing demographic characteristics like gender, age, degree sought, marital status, and source of income. In a comparative study of travel behaviors of international and domestic students at a Southeastern American university by Field (1999), gender, marital status, number of children, national origin, and degree program also proved to have significant effects on students’ travel behavior. Kim and Jogaratnam (2003) in another comparative study of activity preferences of Asian international and domestic American university students also suggested travel behavior being influenced by nationality, gender, age, source of income, and marital status, while key variables in the study of Michael et al. (2004), who investigated international students’ behavior as tourists in Australia, were country of origin, gender, and university attended. In the study of Shoham et al. (2004), it was found that neither marital status nor income played a role in explaining travel differences and that travel preferences of students certainly differ on the basis of their country of origin. One of the most important results from the study of Payne (2009) who also examined travel behaviors of New Zealand’s international students is the influence of nationality on certain activities’ participation and that international students’ preferences can be influenced by age and program of study. Glover (2011) further suggested that travel characteristics are affected by student status, faculty, level of study, and first language. Based on the previous studies, travel behavior in this study is divided into travel preferences and preferred activities that students choose to be involved during traveling. Travel preferences in this study are accommodation type used, style of eating, travel party, purpose of travel, and time of travel. Findings of the previous studies are summarized in Table 1. This study is based on the theoretical framework (Figure 1), which has been developed after an extensive review of literature based on previous studies regarding identifying travel behaviors. After a review of current approaches to complex travel behavior, the theoretical model was summarized, and its components and existing relationships are presented and discussed in the subsequent section. Based on the previous studies and for the purpose of this research, variables have been identified and selected accordingly. Key independent variables (IVs) based on previous studies are students’ demographic characteristics, including age (Chadee and Culter, 1996; Field, 1999; Hsu and Sung, 1997; Kim and Jogaratnam, 2003; Michael et al., 2004; Payne, 2009; Shoham et al., 2004), gender (Chadee and Culter, 1996; Field, 1999; Hsu and Sung, 1997; Kim and Jogaratnam, 2003; Michael et al., 2004; Shoham et al., 2004), level of study (Glover, 2011; Hsu and Sung, 1997; Payne, 2009; Shoham et al., 2004), country of origin (Chadee and Culter, 1996; Field, 1999; Hsu and Sung, 1997; Kim and Jogaratnam, 2003; Payne, 2009), marital status (Chadee and Culter, 1996; Field, 1999; Kim and Jogaratnam, 2003; Michael et al., 2004; Shoham et al., 2004), source of financial support (Field, 1999; Hsu and Sung, 1997; Kim and Jogaratnam, 2003; Shoham et al., 2004), and length of residency in host country (Glover, 2011; Hsu and Sung, 1997; Kim and Jogaratnam, 2003). Apart from Payne’s (2009) study, although previous studies usually surveyed respondents from one department or school from one university or only students of one university (Chen and Kerstetter, 1999; Hsu and Sung, 1997; Field, 1999; Limanond et al., 2011), the students’ current university has been identified as an IV in this study. Downloaded from jvm.sagepub.com at Universiti Sains Malaysia on December 15, 2014 4 Journal of Vacation Marketing Table 1. Factors affecting travel behaviors. Author(s) Subjects Findings Chadee and Insight into international travel by Cutler (1996) students in New Zealand; N ¼ 370 Hsu and Sung (1997) Field (1999) Chen and Kerstetter (1999) Pope et al. (2002) Kim and Jogaratnam (2003) Michael et al. (2004) Shoham et al. (2004) Weaver (2003) Arcodia et al. (2006) Payne (2009) Limanond et al. (2011) Glover (2011) Travel behaviors and preferences depend on the students’ ethnicity and culture Travel behavior could vary because of Travel behaviors of international the differing demographic students at a Midwestern university; characteristics. Age, gender, degree, N ¼ 278 and marital status found to have a significant influence on style of eating A comparative study of travel Gender, marital status, number of behaviors of international and children, national origin, and degree domestic students at a Southeastern proved to have more effects on university; N1 ¼ 509/N2 ¼ 1501 students’ travel behavior International students’ image of rural International students’ destination images are influenced by home Pennsylvania as a travel destination; country, gender, and household N ¼ 2537 status A relationship exists between country The role and economic impact of of origin and travel expenditure international student and family tourism within Western Australia Travel behavior is influenced by Activity preferences of Asian ethnicity, gender, age, source of international and domestic American income, length of stay, and marital university students; N ¼ 514 status Country of origin, gender, and The travel behavior of international university attended affect travel students: the relationship between behavior studying abroad and their choice of tourist destinations; N ¼ 219 Student travel behavior: a crossNationality and gender affect travel national study; N ¼ 558 preferences. Method Quantitative and empirical Quantitative and empirical Quantitative and empirical Quantitative and empirical Quantitative and empirical Quantitative and empirical Quantitative and empirical Quantitative and empirical Travel Preferences depend on ethnicity Quantitative The contribution of international and culture. and students to tourism beyond the core empirical educational experience: evidence from Australia International students an Australian Travel behaviors and preferences Conceptual tourism depend on the student’s ethnicity. Ethnicity plays an important role in International students as domestic Quantitative tourists in New Zealand. A study of and participation in certain activities and travel patterns, behaviors, empirical preferences can be influenced by age motivations and expenditure. N ¼ and program studying 221 Quantitative Travel behavior of university students Gender affects travel behavior and who live on campus: a case study of a empirical rural university in Asia University of Technology, Department of Transportation Engineering; N ¼ 130 A comparison between domestic and Travel aspects are affected by student Quantitative and status, faculty, level of study, and first international students’ trip empirical language characteristics: evidence from an Australian university; N ¼ 948 Source: compiled by the researcher for study. Students from five research universities from three important geographical areas (Kuala Lumpur, Johor Bahru, and Penang) have been chosen to participate in this study to test the existing relationships between the students’ studying areas and their travel preferences. It should be mentioned that Shoham et al. (2004) included monthly income and working Downloaded from jvm.sagepub.com at Universiti Sains Malaysia on December 15, 2014 5 Varasteh et al. Info. source preference Travel preferences Nationality -Time of Travel -Travel Party -Accommodation type used -Style of Eating -Travel purpose Age Gender Marital status Travel behavior Activities undertaken while Level of education traveling -Action - Leisure -Sport nature -Events -Touring -Recreation Source of finance Length of residency Current university Figure 1. Theoretical framework of the study. status (Field, 1999) in demographic characteristics but since most of the international students are not allowed to work in Malaysia during their study tenure; in the absence of regular income, these factors are thus not considered as affecting travel behavior in this study. In this study, dependent variables (DVs; travel behaviors) have been divided into two groups, namely, travel preferences (Field, 1999; Kim and Jogaratnam, 2003; Shoham et al., 2004) and travel-related activities (Field, 1999; Hsu and Sung, 1997; Kim and Jogaratnam, 2003). Travel preferences in this study, following the work of Glover (2011), include items of time of travel (Field, 1999; Michael et al., 2004; Payne, 2009), accommodation (Chadee and Cutler, 1996; Hsu and Sung, 1997; Field, 1999; Michael et al., 2004; Kim and Jogaratnam, 2003; Payne, 2009; Shoham et al., 2004), style of eating (Hsu and Sung, 1997; Field, 1999; Payne, 2009; Shoham et al., 2004), travel party (Glover, 2011; Shoham et al., 2004), and travel purpose (Kim and Jogaratnam, 2003; Payne, 2009; Richards and Wilson, 2004) as DVs (travel preferences), which are affected by demographic characteristics of travelers. Travel-related activities in this study based on previous studies (Hsu and Sung, 1997; Field, 1999; Michael et al., 2004; Kim and Jogaratnam, 2003; Payne, 2009; Shoham et al., 2004) include activities that have been undertaken by travelers while traveling, which consists of leisure-based activities (Field, 1999; Michael et al., 2004; Kim and Jogaratnam, 2003; Payne, 2009; Shoham et al., 2004), sport nature activities (Hsu and Sung, 1997; Field, 1999; Michael et al., 2004; Kim and Jogaratnam, 2003; Payne, 2009; Shoham et al., 2004), events (Hsu and Sung, 1997; Michael et al., 2004; Kim and Jogaratnam, 2003), touring activities (Hsu and Sung, 1997; Field, 1999; Michael et al., 2004; Kim and Jogaratnam, 2003; Payne, 2009), and action and recreation activities (Michael et al., 2004; Shoham et al., 2004). The current competitive condition of the tourism market means consumption of tourism products completely depends on the information sources used by the tourist (McIntosh and Goeldner, 1990; Moutinho, 1987), and as tourists have been also directly segmented based on their search behavior (Bieger and Laesser, 2004; Fodness and Murray, 1997; Um and Crompton, 1990), the need to understand clearly how tourists and student travelers obtain information about their destinations becomes a crucial issue Downloaded from jvm.sagepub.com at Universiti Sains Malaysia on December 15, 2014 6 Journal of Vacation Marketing in tourism marketing (Carr, 2003). According to Fodness and Murray (1997), tourists increase the quality of their travel by searching about preferred destinations, and understanding the way of obtaining information by tourists would enable destination marketers to effectively offer products. Moutinho (1987) also stated while information search employed as a descriptor to profile the behavior of tourists was segmented on some other basis, it has provided important and valuable information for planning and positioning of appropriate tourism marketing strategies. Information sources have been clarified in the previous studies (Crotts, 1999; Moutinho, 1987; Payne, 2009) as tourism offices/travel agents, friends, relatives/family, newspapers, magazines, radio, television, and the Internet. The relationships between information source preference and trip outcome are supported by previous studies as well. Andereck and Caldwell (1994) reported travel behaviors are related to ratings of information sources, while other researchers (Crompton, 1992; Gunn, 1988; Luo et al., 2005; Um and Crompton, 1990) indicated that information source of preferred destinations affected travel outcomes. Dawar et al. (1996) also stated information seeking is often coupled with a cultural background resulting in different patterns of behavior. It is also worth mentioning Bieger and Laesser (2004) found that tourists’ travel behavior could vary based on information sources; for instance, it was revealed in the case of destination choice, with the increase in travel distance, information increases not only with regard to importance but also with regard to professionalism and reliability. Further, it was also found that the higher the degree of professionalism and general importance of information sources, the earlier the final decision is placed ahead of departure. Based on the above discussions and statements, this considers information source preference as a moderator with its main function being to adjust the strength of relationships between nationality and travel behavior. Moutinho (1987) stated information search has provided valuable insights for planning marketing strategies when employed as a descriptor to profile the behavior of tourists segmented on some other basis as well. Research methodology This study adopted a quantitative method to examine the broad spectrum of international students’ travel behavior in five Malaysian universities from three important geographical areas of Malaysia (Kuala Lumpur, Penang, s Johor Bahru), which include UPM, UM, UTM, USM, and UKM. Since the number of international students represent a great proportion among the postgraduate students (masters and doctor of philosophy) in Malaysian universities, and this study is aimed at investigating international students’ travel behavior, postgraduate students have thus been identified to be surveyed in this study. A stratified random sampling was used, which was drawn up on the basis of universities. The total number of international postgraduate students studying in five research universities of Malaysia was 11,749 based on the data published by the Ministry of Higher Education of Malaysia in 2010. The appropriate sample size determined for this study is 386 respondents. Each stratum is taken in a number proportional to the stratum’s size when compared with the population. The survey instrument was administered to the target sample via online survey system, and the research focuses on travel behaviors of international postgraduate students studying in Malaysian universities. Study results centered on preferred time of travel, preferred accommodation, preferred food outlets, traveling party, main purpose of travel, activities undertaken, source of information about preferred destination, and the relationship between demographic characteristics, information source preference, and travel behaviors of students. A literature review of previous studies’ questionnaires was used to explain and justify questions in a valid and relevant manner (Brotherton, 2008). On the basis of literature reviews, a preliminary pattern of international students’ travel behaviors was constructed and a pilot study subsequently conducted. Before conducting the pilot test, an important step called content validity, required to establish the credibility of the research, was carried out. Content validity of the preliminary items was examined by a review panel consisting of seven academic faculty members who are experts in methodology, analysis, and tourism planning. They were asked to review the contents of the questionnaire and items and consider its suitability for the current study. A pilot test of the survey was conducted to ensure that instructions, wordings, explanations, and questions were clear and formatted properly and efficiently. Downloaded from jvm.sagepub.com at Universiti Sains Malaysia on December 15, 2014 7 Varasteh et al. Table 2. Sample size and valid replies in universities. University UM USM UKM UPM UTM Total Population Desired sample size Valid replies Response rate 2405 2035 2333 2728 2248 11,749 79 66 77 90 74 386 81 79 80 95 74 409 52% 60% 52% 53% 50% 53% Note: UM: Universiti Malaya; USM: Universiti Sains Malaysia; UKM: Universiti Kebangsaan Malaysia; UPM: Universiti Putra Malaysia; UTM: Universiti Teknologi Malaysia. In the pilot study, a small proportion sample (22 students) was chosen from USM, and the collected data were subjected to the same process and methods in the research. Based on the results of the pilot test and content validity conducted by the panel, some revisions to wording and layout were made to improve the questionnaire appropriately in order to achieve the research objectives and, finally, a three-section questionnaire was developed to obtain the required data. The first section captured the demographic information of respondents. The second section examined the travel behavior of students consisting of two subsections including travel preferences and activities undertaken by respondents while traveling, using a 5-point Likert-type scale to measure responses, with 5 ¼ almost always, 4 ¼ frequently, 3 ¼ sometimes, 2 ¼ seldom, and 1 ¼ never. The third section measured the level of information source usage by respondents, using a 5-point Likerttype scale to measure responses with 5 ¼ almost always, 4 ¼ frequently, 3 ¼ sometimes, 2 ¼ seldom, and 1 ¼ never. In this section, respondents were asked to indicate to what extent they used listed sources for obtaining information about their preferred destinations. At the beginning of the survey, there was one filtering question, which respondents were required to answer before they could continue to the next sections. If the respondents answered ‘No’ to the question, then they were not able to participate in the survey. This filtering question is: 1. Have you been on an overnight travel in the last 12 months while studying in Malaysia? international offices at universities and relevant student representative bodies of the overseas postgraduate student population in Malaysia. Some international students were also contacted and invited to participate in this survey through Facebook. To get a high response rate, the number of questionnaires distributed were twice the number of the determined sample size in each university. A total of 409 responses were returned for a 53% response rate (see Table 2). The duration of data collection was 3 months, beginning from September till November 2012. In order to analyze the collected data and address the research objectives, SEM-PLS technique was applied to find the relationships between social–demographic characteristics, including a set of personal characteristics and latent variables (LVs) with multiple indicators. In this study’s model, travel behavior is captured by travel preferences and travel activities, which include some LVs to be measured by some observed items. This study also desired to test the moderating effect of information sources on the relationship between demographic characteristics and travel behavior. Analysis and findings The assessment of the model by PLS analysis typically includes the assessment of the measurement model and the structural model (Chin, 2010; Hair et al., 2011, 2012). The assessment of the measurement model examines the validity and reliability of the relationship between the LVs and related observable variables, while assessment of the structural model examines the relationships between constructs (Chin, 2010; Hair et al., 2011). Data collection The online survey of this study was designed using Google Questionnaire Application. The URL link was made available to participants via e-mail obtained through personal contacts with Assessment of measurement model In the model used in this research, a number of reflective constructs were involved including travel behavior, preferences, activities, and Downloaded from jvm.sagepub.com at Universiti Sains Malaysia on December 15, 2014 8 Journal of Vacation Marketing source of information as a moderator. Travel behavior was considered as the third-order factor, and travel preferences and travel activities were the second-order factors. The travel preferences LVs included five first-order factors (time of travel, travel party, accommodation, eating style, and travel purpose) and travel activities LVs also included six first-order factors (action, touring, event, sport, recreation, and leisure). In this model, demographic characteristics were considered as IVs, including age, gender, marital status, nationality, level of study, source of finance, current university, and length of residency in host country, which were defined in the PLS model as a dummy LV. Information sources used in this study were considered as moderator and also as a first-order construct. The assessment of the measurement model has been conducted via a three-step analysis. In the first step, the first-order factors were analyzed together (Akter et al., 2011), and in the second step, after generating the second-order factors, they were also analyzed together; and at the end after generating the third-order factor (travel behavior), an analysis was run to complete the assessment of the measurement model. The reflective measurement model evaluates reliability and validity, and two key criteria for conducting such an evaluation are composite reliability (CR) and average variance extracted (AVE) (Chin, 2010; Hair et al., 2011). It should be noted that, in the assessment of measurement model, the sociodemographic variables were not included because in the model these variables were considered dummy variables and the CR and AVE equal one. Checking for reliability and validity of these variables was thus not required. Evaluating the reliability of the reflective measurement model for SEM included indicator reliability and constructs reliability. The loading of each indicator on its associated latent construct should be checked in order to assess indicator reliability. In order to gain acceptable indicator reliability, the loading of higher than 0.7 was needed (Gotz et al., 2010; Hair et al., 2011; Hulland, 1999). Table 3 shows the loading of indicators on their associated LVs before creating second-order LVs is higher than 0.7 and are thus acceptable. Furthermore, to assess construct reliability, two coefficients are typically considered, that is, CR and the more common coefficient Cronbach’s a (Bagozzi and Yi, 1988; Chin, 2010; Gotz et al., 2010). However, CR is more suitable for PLS-SEM (Hair et al., 2011); hence, CR has been mentioned in Table 3. Results of the measurement model for first-order factors. Construct Items Factor loading CR AVE Time of travel 0.74 0.5 A1 A2 A3 0.671 0.776 0.656 A4 A5 A6 0.656 0.672 0.696 A7 A8 A9 A10 A11 0.587 0.726 0.74 0.72 0.669 A12 A13 0.789 0.789 A14 A15 A16 A17 A18 A19 A20 0.637 0.684 0.792 0.827 0.765 0.657 0.663 B1 B2 B3 B4 B5 0.566 0.714 0.778 0.779 0.681 B6 B7 B8 B9 0.684 0.728 0.716 0.599 B10 B11 B12 0.841 0.867 0.611 B13 B14 B15 0.805 0.855 0.805 B16 B17 0.835 0.835 B18 B19 B20 B21 B22 0.735 0.788 0.67 0.599 0.618 C1 C2 C3 0.494 0.57 0.735 Travel party 0.5 0.71 Accommodation 0.5 0.82 Preferred meals 0.6 0.77 Travel purpose 0.5 0.88 Action 0.5 0.83 Touring 0.5 0.78 Event 0.6 0.82 Sport 0.7 0.86 Recreation 0.7 0.82 Leisure 0.5 0.81 Information source 0.5 0.89 Downloaded from jvm.sagepub.com at Universiti Sains Malaysia on December 15, 2014 (continued) 9 Varasteh et al. Table 3. (continued) Construct Items Factor loading CR AVE C4 C5 C6 C7 C8 0.8 0.811 0.804 0.715 0.672 Note: CR: composite reliability; AVE: average variance extracted. the tables that follow subsequently. Table 3 illustrates CR of first-order factors, which are higher than 0.7 for all the factors. Therefore, the measurement model has internal consistency and considered reliable. The validity of the reflective measurement model consists of convergent and discriminant validity (Gotz et al., 2010; Hair et al., 2011). To obtain an acceptable convergent validity, the AVE values of LVs should be higher than 0.5 (Bagozzi and Yi, 1988; Chin, 2010; Hair et al., 2011). AVE is used to measure the variance in an LV that is contributed from its indicators (Chin, 2010). Table 3 shows that the AVE values of all constructs of the measurement model are higher than 0.5, so the convergent validity is acceptable (Chin, 2010). Discriminant validity is the extent to which each construct is accurately distinct from the other constructs in the model. In order to test discriminant validity, the root square of AVE of each construct should be higher than the correlation of the construct with any other LV in the model, and an indicator’s loading with its associated LV must be higher than its loading with other LVs (Chin, 2010; Fornell and Larcker, 1981; Hair et al., 2011). Table 4 presents the comparison of the square root of AVE of each construct with the correlation of the other construct. This comparison demonstrates that for all the constructs, the discriminant validity is completely acceptable. The results of the assessment of the measurement model show the reliability, convergent validity, and discriminant validity are highly acceptable for measurement model consisting of first-order constructs. The second step of analysis of the measurement model was performed by generating two second-order factors (travel preferences and travel activities). Time of travel, travel party, accommodation, eating style, and travel purpose are considered as indicators for travel preferences, while action, touring, event, sport, recreation, and leisure are considered as indicators of travel activities. In this stage, the measurement model with second-order factors was thus assessed. The result shows CR of generated second orders is 0.75 and 0.82 for travel preferences and travel activities, respectively. Moreover, the AVE of secondorder constructs is 0.50 and 0.54 for travel preferences and travel activities, respectively. After generating the third-order construct, namely, travel behavior, the CR and AVE are 0.90 and 0.8, respectively. The reliability, convergent validity, and discriminant validity are highly acceptable for measurement model in three stages. Therefore, there were accurate relationships between LVs and related observable variables of the proposed model, and consequently, the validity and reliability of the adopted questionnaire are also confirmed. Assessment of structural model The following two criteria should be evaluated to obtain a preliminary assessment of the structural model (inner model) and hypothetical framework: R2 measure of endogenous constructs and the path coefficients (Chin, 2010; Hair et al., 2011). The path coefficients must be significant, and R2 is highly dependent on the research area. Since the objective of this study is to examine the relationship between demographic characteristics and travel behavior, R2 would not be appropriate and relevant to this study because R2 shows the predictive role of IV on DV. Therefore, R2 is not considered in this analysis, and this research did not consider the role of demographic variables on travel behavior as the previous studies also eliminated R2 from their studies (Field, 1999; Hsu and Sung, 1997; Kim and Jogaratnam, 2003; Shoham et al., 2004; Payne, 2009). As this research aims to investigate whether the effect of demographic variables are significant on travel preference and travel activities, the significance between demographic characteristics and travel behavior including travel preferences and travel activities needed to be assessed and reported. The analysis of the structural model was performed in three stages. In the first stage, the relationships between demographic characteristics and first-order factors of travel preferences (time of travel, travel party, accommodation, eating style, and travel purpose) and travel activities (action, touring, event, recreation, sport, and leisure) were examined. Tables 5 and 6 represent the results of hypotheses testing of this stage and show the significance levels of path coefficients. Downloaded from jvm.sagepub.com at Universiti Sains Malaysia on December 15, 2014 Downloaded from jvm.sagepub.com at Universiti Sains Malaysia on December 15, 2014 Table 4. Discriminant validity after modification. Time of travel Travel party Accommodation Meals Travel purpose Action Event Touring Sport Recreation Leisure Information source Time of travel Travel party Accommodation Meals Travel purpose Action Event Touring Sport Recreation Leisure Information source 0.703 0.488 0.401 0.198 0.627 0.471 0.289 0.321 0.278 0.335 0.367 0.424 0.675 0.362 0.426 0.484 0.476 0.351 0.227 0.334 0.39 0.389 0.666 0.691 0.169 0.336 0.355 0.283 0.149 0.456 0.423 0.547 0.343 0.789 0.271 0.331 0.123 0.186 0.18 0.185 0.272 0.424 0.722 0.53 0.352 0.443 0.283 0.41 0.306 0.466 0.708 0.536 0.46 0.522 0.542 0.508 0.513 0.781 0.386 0.396 0.377 0.455 0.438 0.684 0.19 0.226 0.302 0.212 0.822 0.662 0.616 0.381 0.835 0.462 0.382 0.686 0.442 0.709 Note: square roots of average variances extracted are shown on diagonal and in boldface. 11 Varasteh et al. Table 5. Relationships between age, gender, marital status, nationality and constructs (first-order factors). Hypothesis Path coefficient p value 0.069 0.143 0.055 0.117 0.041 0.085 0.027 0.017 0.102 0.168 0.113 0.03 0.014 0.077 0.109 0.035 0.016 0.01 0.103 0.156 0.062 0.036 0.121 0.071 0.205 0.059 0.076 0.113 0.175 0.013 0.099 0.048 0.113 0.079 0.205 0.095 0.068 0.089 0.138 0.025 0.016 0.05 0.101 0.068 0.14 <0.01 0.17 <0.05 0.262 <0.1 0.342 0.387 <0.05 <0.01 <0.05 0.275 0.38 <0.1 <0.01 0.226 0.372 0.428 <0.05 <0.01 0.109 0.202 <0.05 <0.1 <0.01 0.159 <0.1 <0.01 <0.01 0.414 <0.05 0.152 <0.05 <0.05 <0.01 <0.05 <0.05 <0.05 <0.01 0.34 0.385 0.138 <0.05 <0.1 Age ! Time of travel Age ! Travel party Age ! Accommodation Age! Preferred meal Age! Travel purpose Age ! Action Age ! Event Age ! Touring Age ! Sport Age ! Recreation Age ! Leisure Gender ! Travel time Gender ! Travel party Gender ! Accommodation Gender ! Meal Gender ! Travel purpose Gender ! Action Gender ! Event Gender ! Touring Gender ! Sport Gender ! Recreation Gender ! Leisure Marital status ! Travel time Marital status ! Travel party Marital status ! Accommodation Marital status ! Meal Marital status ! Travel purpose Marital status ! Action Marital status ! Event Marital status ! Touring Marital status ! Sport Marital status ! Recreation Marital status ! Leisure Nationality ! Travel time Nationality ! Travel party Nationality ! Accommodation Nationality ! Meal Nationality! Travel Purpose Nationality ! Action Nationality ! Event Nationality ! Touring Nationality ! Sport Nationality ! Recreation Nationality ! Leisure The tables illustrate the influence of IVs, including age, gender, marital status, nationality, source of finance, level of study, length of residency, and current university on first-order factors including preferred meal, accommodation, travel party, time of travel, travel purpose, and preferred activities. Table 7 shows the results of the second stage of assessment of the structural model. In this stage, the relationships between demographic characteristics and second-order factors including travel activities and travel preferences were Supported No Yes No Yes No Yes No No Yes Yes Yes No No Yes Yes No No No Yes Yes No No Yes Yes Yes No Yes Yes Yes No Yes No Yes Yes Yes Yes Yes Yes Yes No No No Yes Yes examined. Table 7 shows a highly significant relationship between age, marital status, nationality, source of finance, and factors of both travel activities and travel preferences. However, none of the other demographic characteristics including gender, level of study, years in host country, and current university has influence on either travel activities or travel preferences as the second-order factors. Figure 2 also illustrates the relationships between age, marital status, nationality, source of finance, and factors of both travel activities and travel. Downloaded from jvm.sagepub.com at Universiti Sains Malaysia on December 15, 2014 12 Journal of Vacation Marketing Table 6. Relationships between study level, source of finance, length of residency, current university, and constructs (first-order factors). Hypothesis Study level ! Time of travel Study level ! Travel party Study level ! Accommodation Study level ! Preferred Meal Study level ! Travel Purpose Study level ! l Action Study level ! l Event Study level ! Touring Study level ! Sport Study level ! Recreation Study level ! Leisure Source of finance ! Travel time Source of finance ! Travel party Source of finance ! Accommodation Source of finance ! Preferred meal Source of finance ! Travel purpose Source of finance ! Action Source of finance ! Event Source of finance ! Touring Source of Finance ! Sport Source of Finance ! Recreation Source of Finance ! Leisure Length of residency ! Travel Time Length of residency ! Travel party Length of residency ! Accommodation Length of residency ! Meal Length of residency ! Travel purpose Length of residency ! Action Length of residency ! Event Length of residency ! Touring Length of residency ! Sport Length of residency ! Recreation Length of residency ! Leisure University ! Travel time University ! Travel party University ! Accommodation University ! Meal University ! Travel purpose University ! Action University ! Event University ! Touring Table 8 shows the results of hypothesis testing after generating the third-order factor. The findings show the highly significant influence of age, marital status, nationality, and source of finance on travel behavior (see Figure 3 and Table 8). The moderating effect of information source A moderator construct is basically an IV, which modifies the relationship between two other variables. Moderator variables including specific factors (e.g., context information) are often assumed to reduce or enhance the influence that specific b Coefficient p Value Supported 0.03 0.137 0.045 0.038 0.044 0.028 0.013 0.127 0.067 0.119 0.0 0.045 0.14 0.055 0.20 0.064 0.105 0.021 0.012 0.1 0.07 0.039 0.126 0.01 0.007 0.06 0.068 0.073 0.094 0.045 0.01 0.014 0.009 0.068 0.014 0.054 0.068 0.068 0.033 0.071 0.012 0.301 <0.01 0.205 0.241 0.249 0.297 0.396 <0.05 0.138 <0.05 0.168 0.198 <0.01 0.119 <0.01 0.128 <0.05 0.346 0.4 <0.05 <0.05 0.2 <0.05 0.432 0.451 0.175 0.121 <0.1 <0.1 0.178 0.42 0.403 0.443 <0.1 0.373 0.143 <0.05 <0.1 0.25 <0.1 0.388 No Yes No No No No No Yes No Yes No No Yes No Yes No Yes No No Yes Yes No Yes No No No No Yes Yes No No No No Yes No No Yes Yes No Yes No IVs have on specific responses in question (DV). In this study, information source preference has been considered as a moderator with its main function in adjusting the strength of relationships between nationality and travel behavior. The result indicated the moderating effect of information source on relationships between nationality and travel behavior is highly significant. The results reveal the interaction effect is 0.164 and the p value of the interaction effect is significant at 0.01. Figure 4 shows the differences between respondents with high usage as well as low usage of information source. The relationship between Downloaded from jvm.sagepub.com at Universiti Sains Malaysia on December 15, 2014 13 Varasteh et al. Table 7. Relationships between demographic variables and constructs (second-order factors). Hypothesis Age ! Travel activities Age ! Travel preferences Gender ! Travel activities Gender ! Travel preferences Marital Status ! Travel activities Marital Status ! Travel preferences Nationality ! Travel activities Nationality ! Travel preferences Study Level ! Travel activities Study Level ! Travel preferences Source of Finance ! Travel activities Source of Finance ! Travel preferences Length of residency ! Travel activities Length of residency ! Travel preferences University ! Travel activities University ! Travel preferences Path coefficient p Value Supported 0.104 0.12 0.033 0.005 0.128 0.151 0.083 0.151 0.001 0.016 0.075 0.113 0.052 0.039 0.004 0.047 <0.05 <0.05 0.247 0.453 <0.05 <0.05 <0.1 <0.01 0.489 0.372 <0.1 <0.05 0.185 0.252 0.467 0.158 Yes Yes No No Yes Yes Yes Yes No No Yes Yes No No No No Figure 2. Relationships between demographic variables and travel preference and activity. nationality and travel behavior for respondents with low usage of information source is negative, whereas this relationship for respondents with high usage of information source is positive. This study thus confirms the moderating role of information source on relationship between nationality and travel behavior. Discussion This article can assist destination marketers and tourism organizers to gain useful information on the travel behavior of international students in Malaysia through a comprehensive investigation with the aim of achieving three objectives, Downloaded from jvm.sagepub.com at Universiti Sains Malaysia on December 15, 2014 14 Journal of Vacation Marketing Table 8. Relationships between demographic variables and travel behavior. Hypothesis Path coefficient p Value Supported 0.123 0.021 0.153 0.129 0.008 0.104 0.05 0.024 <0.1 0.315 <0.01 <0.01 0.438 <0.05 0.197 0.303 Yes No Yes Yes No Yes No No Age ! Travel Behavior Gender ! Travel behavior Marital Status ! Travel behavior Nationality ! Travel Behavior Study Level ! Travel behavior Source of Finance ! Travel behavior Length of Residency ! Travel behavior University ! Travel behavior Figure 3. Relationships between demographic variables and travel behavior. which were to examine the relationships between demographic characteristics and travel preferences of international students in Malaysia, to examine the relationships between demographic characteristics and travel activities of international students while traveling within Malaysia, and finally to investigate the moderating effect with regard to the relationships between country of origin and travel behavior of students in Malaysia. SEM was utilized to analyze the quantitative data and to explore the existing relationships among variables, followed by assessments of the measurement model and structural model, which were reported to describe reliability and validity of developed questionnaires as well as relationships between variables. The testing of assumptions before the performance of each statistical technique was all satisfied. The following sections discuss the objectives and the findings of the study in reasonable detail. Results revealed that preferred time for traveling was affected by marital status, nationality, years of residency in Malaysia, and location of current university. Significant relationships between preferred accommodation and gender, marital status, and nationality were also revealed. The study also found preferred meal to be significantly influenced by gender, age, nationality, source of finance, and current university, which is in agreement with Hsu and Sung (1997) who reported in their study that age and gender affect choices of food outlets, which is also Downloaded from jvm.sagepub.com at Universiti Sains Malaysia on December 15, 2014 15 Varasteh et al. Figure 4. Moderating effect of information source on the relationships between nationality and travel behavior. supported by Shoham et al. (2004) confirming the influence of gender on choices of meals. Hsu and Sung (1997) also reported the influence of degree and marital status on style of eating but is not confirmed by the findings of this study. The results also showed preferred travel party is significantly affected by age, marital status, nationality, level of study, and source of finance. Regarding the reasons behind international students’ main purposes for traveling, it was found that travel purpose is associated with marital status, nationality, and current university of international students. The study also found travel activities undertaken by students when they are traveling were affected by age, gender, nationality, marital status, source of finance, years in host country, level of study, and current university. Findings of Payne (2009) confirmed the influence of nationality on different travel activities and levels of participation. Relationships between demographic characteristics, information source preference, and travel behaviors of students were investigated as the main objective of the study. After generating secondorder constructs, including travel preferences and preferred activities, it was found that there is a highly significant relationship between age, marital status, nationality, source of finance, and factors of both travel activities and travel preferences. However, none of the other demographic characteristics including gender, level of study, years in host country, and current university has any influence on either travel activities or travel preferences as the second-order factors. There were some differences as well as similarities between these findings and the work of Hsu and Sung (1997). However, the current results support the findings of Hsu and Sung (1997) regarding the influence of age and marital status on travel activities. The influence of degree and gender which was found to be significant in the mentioned study was, however, not confirmed by the present study. After generating third-order construct termed travel behavior, results showed the highly significant influences between age, marital status, nationality, source of finance, and travel behavior. However, other variables including gender, level of study, years in host country, and current university have an insignificant effect on travel behavior, which is consistent with Hsu and Sung (1997) who noted travel preferences could vary because of the differing demographic characteristics and suggested that market of international students can be segmented in terms of not only nationality (Arcodia et al., 2006; Pope et al., 2002; Chadee and Cutler, 1996; Kim and Jogaratnam, 2003; Shoham et al., 2004; Weaver, 2003) but also other variables such as age (Giuliano, 2003; Giuliano and Dargay, 2006; Giuliano and Narayan, 2003; Kim and Jogaratnam, 2003), source of income (Kim and Jogaratnam, 2003), and marital Downloaded from jvm.sagepub.com at Universiti Sains Malaysia on December 15, 2014 16 Journal of Vacation Marketing status (Kim and Jogaratnam, 2003), as travel behavior is influenced by these variables. It should be noted that findings of the current study do not support the previous research by Shoham et al. (2004) who stated neither marital status nor income played a role in explaining travel differences. They also reported the significant influence of gender on travel preference, which is also not supported by the findings of this study. However, findings of the current study are in agreement with the findings of Michael et al. (2004) regarding identifying country of origin as one of the key variables but do not support the role of gender and university attended as the key variables. According to previous studies (Arcodia et al., 2006; Chadee and Cutler, 1996; Chen and Kerstetter, 1999; Glover, 2011; Field, 1999; Hsu and Sung, 1997; Kim and Jogaratnam, 2003; Limanond et al., 2011; Michael et al., 2004; Payne, 2009; Pope et al., 2002; Shoham et al., 2004; Weaver, 2003), it is believed that different demographic characteristics of travelers will result in different travel behavior, and it is also well documented that information source preference about destination in so many studies has been considered as an IV, which has significant effects on travel behavior (Andereck and Caldwell, 1994; Bonn et al., 1998; Dawar et al., 1996). Dawar et al. (1996) also stated information seeking is often coupled with cultural background resulting in different patterns of behavior. Based on these statements, information source preference has been considered as a moderator in this study with its main function being to adjust the strength of relationships between nationality and travel behavior. The results show that the moderating effect of information source on relationship between nationality and travel behavior is highly significant. This study therefore confirms the moderating role of information source on relationship between nationality and travel behavior. Implications of the study This research is the first attempt to investigate the travel behaviors of the international students who are currently studying in Malaysia. Important factors influencing travel behavior were identified, in a way that may potentially contribute to the development of a reliable travel behavior framework. This study revealed that different travel behaviors are strongly associated with demographic characteristics of international students. It was found that age, marital status, and nationality were the primary basis for segmentation because most of the students’ preferences vary predominantly based on their age, nationality, and marital status. The travel activities and preferences include travel party (affected by age, marital status, nationality, level of study, financial supports, and current university, while preferred meal affected by age, gender, nationality, and financial support), preferred accommodation (affected by gender, marital status, and nationality), travel purpose (affected by marital status and nationality), time of travel (affected by marital status, nationality, and length of residency), action activities (affected by age, marital status, nationality, and source of finance), sport activities (affected by age, gender, marital status, source of finance, and current university), recreation activities (affected by age, marital status, level of study, and financial support), leisure activities (affected by age, marital status, nationality, and current university), touring activities (affected by gender and level of study), and event activities (affected by marital status and level of study). These findings will facilitate tourism marketers when constructing their strategy, with the need to factor in the influence of students’ demographic characteristics in order to strategize and promote tourism products concerning different age-groups, nationalities, marital status, genders, and other classifications. Tourism marketers may, for instance, consider different age-groups’ interests and preferences in offering any action, sport, recreation, and leisure activities and restaurants in order to meet the students’ needs. They should also consider students’ level of study in offering touring and event activities in various tourism destinations. By sharing this relevant consumer information with all stakeholders and tourism operators, the strategic implications for tourism planners, managers, and policy makers would become more apparent. This area provides further research opportunities to identify applicable segments of the market and inform, so that tourism operators can consequently market appropriate products through approaches that are well structured and organized, allowing for the greatest return while better serving the needs of customers. When these significant differences and similarities are identified, tourism marketers would account for them through adapting their strategies, allowing for the vagaries in traveling students’ characteristics, rather than adopting a one-size-fits-all strategy. Downloaded from jvm.sagepub.com at Universiti Sains Malaysia on December 15, 2014 17 Varasteh et al. Conclusion and suggestions for further research This study revealed that although each component or indicator of preferred travel activities and preferences has been affected by some demographic characteristics of respondents, travel behavior (as a third-order factor) was only affected by four characteristics. The highly significant influences between age, marital status, nationality, and source of finance on travel behavior have been reported. Other variables including gender, level of study, years in host country, and current university have an insignificant effect on travel behavior, while moderating effect of information source on relationship between nationality and travel behavior has also been found by the current study for the first time. The results of this study can facilitate destination marketers and tourism organizers and contribute significantly to the development of marketing strategies in improving the markets and helping it to better serve the needs of its customers. The findings of this research, however, should be interpreted in light of some limitations. This study only focused on travel behaviors of international postgraduate students who were currently studying in five Malaysian universities in three cities, namely, Kuala Lumpur, Penang, and Johor Bahru; consequently, further studies on other university students’ travel behaviors are suggested to validate these research findings and their relevance to students of other universities. Further studies on undergraduate students’ travel behavior are also suggested to allow for more accurate comparison between these two groups and help develop deeper understanding of international students’ travel behavior. Additionally, further studies are also needed to investigate students’ preferences comprehensively in order to help tourism operators and planners to better understand students’ needs and desires because this study focused only on the existing relationships between their demographic characteristics and travel behavior, while more studies need to be conducted to investigate and describe preferences of different group ages, genders, nationalities, and other demographic groups. Funding This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors. References Akter S, D’Ambra J and Ray P (2011) Trustworthiness in mHealth information services: an assessment of a hierarchical model with mediating and moderating effects using partial least squares (PLS). Journal of the American Society for Information Science and Technology 62(1): 100–116. Andereck KL and Caldwell LL (1994) The influence of tourists’ characteristics on ratings of information sources for an attraction. Journal of travel and Tourism Marketing 2(2-3): 171–190. Arcodia C, Mei XY and Dickson C (2006) International students an Australian tourism: a critical review of the literature. Paper presented at the Trends, Impacts and Policies on Tourism Development, Crete, Greece. Bahng Y and Kincade DH (2014) Retail buyer segmentation based on the use of assortment decision factors. Journal of Retailing and Consumer Services 21(4): 643–652. Bagozzi RP and Yi Y (1988) On the evaluation of structural equation models. Academy of Marketing Science 16(1): 74–94. Bass FM, Tigert DJ and Lonsdale RT (1968) Market segmentation: group versus individual behavior. Journal of Marketing Research 5: 264–270. Beane T and Ennis D (1987) Market segmentation: a review. European Journal of Marketing 21(5): 20–42. Bieger T and Laesser C (2004) Information sources for travel decisions: toward a source process model. Journal of Travel Research 42(4): 357–371. Blattberg RC, Peacock P and Sen SK (1976) Purchasing strategies across product categories. Journal of Consumer Research 3: 143–154. Bonn MA, Furr HL and Susskind AM (1998) Using the Internet as a pleasure travel planning tool: an examination of the sociodemographic and behavioral characteristics among Internet users and nonusers. Journal of Hospitality and Tourism Research 22(3): 303–317. Brayley RE (1990) An analysis of destination attractiveness and the use of psychographics and demographics in segmentation of the within-state tourism market. Dissertation Abstracts International. A, Humanities and Social Sciences 51(5): 1772. Brotherton B (2008). Researching Hospitality and Tourism: A Student Guide. London: Sage Publications Limited. Carr N (2003) Use and trust of tourism information sources amongst university students. In: Ritchie B, Carr N, and Cooper C (eds) Managing Educational Tourism. Clevedon: Channel View Publications, pp. 215–220. Downloaded from jvm.sagepub.com at Universiti Sains Malaysia on December 15, 2014 18 Journal of Vacation Marketing Chadee DD and Cutler J (1996) Insights into international travel by students. Journal of Travel Research 35(2): 75–80. Chadee DD and Mattsson J (1996) An empirical assessment of customer satisfaction in tourism. Service Industries Journal 16(3): 305–320. Chen P-J and Kerstetter DL (1999) International students’ image of rural Pennsylvania as a travel destination. Journal of Travel Research 37(3): 256–266. Chin WW (2010) How to write up and report PLS analyses. In: Vinzi VE, Chin WW, Henseler J, and Wang H (eds) Handbook of Partial Least Squares. Berlin: Springer, pp. 655–690. Crompton J. (1992). Structure of vacation destination choice sets. Annals of Tourism Research, 19, 420–434. Crotts J. (1999). Consumer decision making and prepurchase information search. In: Pizam A and Mansfeld Y (eds) Consumer Behavior in Travel and Tourism. New York: The Haworth Hospitality Press, pp. 149–168. Dawar N, Parker PM and Price LJ (1996) A crosscultural study of interpersonal information exchange. Journal of International Business Studies 27: 497–516. Dibb S, Stern P and Wensley R (2002) Marketing knowledge and the value of segmentation. Marketing Intelligence and Planning 20(2): 113–119. Field AM (1999) The college student market segment: a comparative study of travel behaviors of international and domestic students at a southeastern university. Journal of Travel Research 37(4): 375–381. Fodness D and Murray B (1997) Tourist information search. Annals of Tourism Research 24(3): 503–523. Fornell C and Larcker DF (1981) Evaluation structural equation models with unobservable variable and measurement error. Journal of Marketing Research 18(1): 39–50. Frank RE, Massy WF and Wind Y (1972) Market Segmentation. Englewood Cliffs: Prentice-Hall. Gibson H (1998) The educational tourist. Journal of Physical Education, Recreation and Dance 69(4): 32–34. Glover P (2011) A comparison between domestic and international students’ trip characteristics: evidence from an Australian university. Journal of Vacation Marketing 17(4): 263–274. Gotz O, Liehr-Gobbers K and Krafft M (2010) Evaluation of structural equation models using the partial least squares (PLS) approach. In: Vinzi VE, Chin WW, Henseler J, and Wang H (eds) Handbook of Partial Least Squares. Berlin: Springer, pp. 691–712. Giuliano G (2003) Travel, location and race/ethnicity. Transportation Research Part A: Policy and Practice 37(4): 351–372. Giuliano G and Dargay J (2006) Car ownership, travel and land use: a comparison of the US and Great Britain. Transportation Research Part A: Policy and Practice 40(2): 106–124. Giuliano G and Narayan D (2003) Another look at travel patterns and urban form: the US and Great Britain. Urban Studies 40(11): 2295–2312. Gunn CA (1988) Vacationscape: Designing tourist regions. New York: Van Nostrand Reinhold. Hair JF, Ringle CM and Sarstedt M (2011) PLS-SEM: indeed a silver bullet. The Journal of Marketing Theory and Practice 19(2): 139–152. Hair JF, Sarstedt M, Ringle CM, et al. (2012) An assessment of the use of partial least squares structural equation modeling in marketing research. Journal of the Academy of Marketing Science 40(3): 414–433. Hu B and Morrison AM (2002) Tripography: can destination use patterns enhance understanding of the VFR market? Journal of Vacation Marketing 8(3): 201–220. Hulland J (1999) Use of partial least squares (PLS) in strategic management research: a review of four recent studies. Strategin Management Journal 20: 195–204. Hsu CH and Sung S (1997) Travel behaviors of international students at a Midwestern university. Journal of Travel Research 36(1): 59–65. Jones SC, Rees L, Hall D, et al. (2005) Using market segmentation theory to select target markets for sun protection campaigns. ANZMAC Conference, University of Western Australia. pp. 144–149. Kim K (2007) Understanding differences in tourist motivation between domestic and international travel: the university student market. Tourism Analysis 12(1-2): 1–2. Kim K and Jogaratnam G (2003) Activity preferences of Asian international and domestic American university students: An alternate basis for segmentation. Journal of Vacation Marketing 9(3): 260–270. Kim K, Jogaratnam G and Noh J (2006) Travel decisions of students at a US university: segmenting the international market. Journal of Vacation Marketing 12(4): 345–357. Kim K, Oh I-K and Jogaratnam G (2007) College student travel: a revised model of push motives. Journal of Vacation Marketing 13(1): 73–85. Kline CS, Greenwood JB, Swanson J, et al. (2014) Paddler market segments: expanding experience use history segmentation. Journal of Destination Marketing and Management 2(4): 228–240. Downloaded from jvm.sagepub.com at Universiti Sains Malaysia on December 15, 2014 19 Varasteh et al. Kotler P (1997) Marketing Management, Analysis Planning and Control. Englewood Cliffs: Perentice Hall International. Kotler P, Roberto N and Lee N (2002) Social Marketing: Improving the Quality of Life (2nd ed.). Thousand Oaks: Sage Publications. Limanond T, Butsingkorn T and Chermkhunthod C (2011) Travel behavior of university students who live on campus: a case study of a rural university in Asia. Transport Policy 18(1): 163–171. Luo M, Feng R and Cai LA (2005) Information search behavior and tourist characteristics: the Internet vis-à-vis other information sources. Journal of travel and Tourism Marketing 17(2-3): 15–25. McIntosh RW and Goeldner CR (1990) Tourism: Principles, Prattices, Philosophies. New York: Wiley. Michael I, Armstrong A and King B (2004) The travel behaviour of international students: the relationship between studying abroad and their choice of tourist destinations. Journal of Vacation Marketing 10(1): 57–66. MOHE (2010) Statistics of Higher Education of Malaysia. Putrajaya: Ministry of Higher Education, Malaysia. Moutinho L (1987) Consumer behaviour in tourism. European Journal of Marketing 21(10): 5–44. Oppermann DM (1994) Regional market segmentation analysis in Australia. Journal of travel and Tourism Marketing 2(4): 59–74. Payne K (2009) International Students as Domestic Tourists in New Zealand. A Study of Travel Patterns, Behaviours, Motivations and Expenditure. New South Wales: Southern Cross University. Pearce PL (2005) Tourist Behaviour: Themes and Conceptual Schemes (Vol. 27). Clevedon: Channel View Books. Pope J, Shanka T and Ali-Knight J (2002) The role and economic impact of international student and family tourism within Western Australia. Paper presented at the the12th International Research Conference of CAUTHE Western Australia. Recker WW, McNally MG and Root G (1986) A model of complex travel behavior: part I—theoretical development. Transportation Research Part A: General 20(4): 307–318. Richards G and Wilson J (2004) The international student travel market: travel style, motivations, and activities. Tourism Review International 8(2): 57–67. Ryan C and Zhang Z (2007) Chinese students: holiday behaviours in New Zealand. Journal of Vacation Marketing 13(2): 91–105. Shoham A, Schrage C and Eeden S (2004) Student travel behavior: a cross-national survey. Journal of travel and Tourism Marketing 17(4): 1–10. Sussmann S and Rashcovsky C (1997) A crosscultural analysis of English and French Canadian’s vacation travel patterns. International Journal of Hospitality Management 16(2): 191–208. Um S and Crompton JL (1990) Attitude determinants in tourism destination choice. Annals of Tourism Research 17(3): 432–448. Weaver DB (2003) The contribution of international students to tourism beyond the core educational experience: evidence from Australia. Tourism Review International 7(2): 95–105. Xu F, Morgan M and Song P (2009) Students’ travel behaviour: a cross-cultural comparison of UK and China. International Journal of Tourism Research 11(3): 255–268. Yusoff YM and Chelliah S (2010) Adjustment in international students in Malaysian public university. International Journal of Innovation, Management and Technology 1(3): 275–278. Downloaded from jvm.sagepub.com at Universiti Sains Malaysia on December 15, 2014