Abstract
It is widely recognized that tourism is becoming more and more international as a result of globalization, and since the 20th century, marketers have confronted increasingly multicultural marketplaces. This trend necessitates cross-cultural research in the hospitality and tourism area. The current study provides a state-of-art assessment of journal publications on cross-cultural tourist research. In this study, various aspects of cross-cultural tourist research are reviewed, addressing conceptual and methodological issues associated with the extant research. Existing studies are largely limited to English-speaking countries and data equivalence issues were not addressed properly, which may lead confounding explanations. It is suggested that future studies should be grounded in solid theoretical foundation and employ a more rigorous research design.
Marketers have confronted increasingly multicultural marketplaces since the 20th century. The growing number of migrants and transcontinental communication media are producing multicultural populations in domestic markets, and consumers are gradually exposing alternative needs and wants. This group of multicultural consumers, together with the globalization of markets and international competition, are forcing firms to operate in a multicultural environment. This change creates opportunities and challenges for business and government. One reason for these challenges is that consumer behavior is culturally bound. Cultural researchers argue that behavior differs from culture to culture because different cultural groups hold different values (Legoherel, Dauce, Hsu, & Ranchhold, 2009).
Culture encompasses elements as shared values, beliefs, and norms, which collectively distinguish particular groups of people from one another (Pizam, Jansen-Verbeke, & Steel, 1997). These widely shared values are subtly programmed into individuals from an early age (Otaki, Durrett, Richards, Nyquist, & Pennebaker, 1986) and prove resistant to change (Hofstede, 1997). A solid understanding of cross-cultural consumer behavior is, therefore, more crucial than ever. In addition, the study of cross-cultural consumer behavior helps assess the generalizability of empirical findings, assess if the findings differ from one cluster to another, understand the behavior of people living in a different culture (Costa & Bamossy, 1995), and identify the cultural dimensions or contextual factors that cause the differences (Earley & Singh, 1985). This backdrop necessitated cross-cultural management research and cross-cultural consumer research as early as the 1950s.
No one would challenge the fact that tourism is becoming more and more international as the world flattens. However, tourism and hospitality researchers were extremely slow to recognize this fact (Dimanche, 1994), and the study of cross-cultural consumer behavior in hospitality and tourism saw limited action in the late 1980s (e.g., Richardson & Crompton, 1988). Since then, a growing number of studies have discussed the similarities and/or differences in tourist behavior across nations and cultures. Despite the growing amount of cross-cultural tourist research, the paucity of systematic reviews of the body of knowledge has failed to synthesize previous studies and identify future research trends. The purpose of the study is, therefore, to systematically review and synthesize cross-cultural tourist research. The results will provide reference points and future directions for scholars interested in cross-cultural research in hospitality and tourism.
Cross-Cultural Consumer Research
Culture can be either etically or emically defined. From an etic approach, culture is defined by Hofstede (1997) as “the collective programming of the mind which distinguishes the members of one group or category of people from another” (p. 5). This definition focuses on the comparison of one culture with another and is typical for cross-cultural consumer behavior research. Researchers who follow an etic approach look for universal or culture-free theories by searching for variables and constructs common to all cultures to discover the differences and similarities between cultures.
From the emic perspective, culture is defined as
the “lens” through which all phenomena are seen. It determines how these phenomena are apprehended and assimilated . . . culture is the “blueprint” of human activity. It determines the coordinates of social action and productive activity, specifying the behaviors and objects that issue from both. (McCracken, 1988, p. 73)
Emic approaches promote a complete understanding of culture through thick description, instead of directly comparing two or more different cultures. Therefore, studies from an emic perspective provide culture-rich information rather than culture-free measures that can be directly compared.
The choice of etic versus emic approaches depends on the nature of the research question, the researcher’s resources and training, and the purpose of the study (Luna & Gupta, 2001). The component elements of culture include language(s), nationality, education (general), country of residence, profession (specialized education), groups (ethnicity), religion, family, gender, social class, and corporate or organizational culture (Legoherel et al., 2009; Usunier, 2000).
Hofstede (1997) suggested that culture influences behavior through four manifestations: values, heroes, rituals, and symbols. A value has been described by Rokeach (1968) as a “centrally held, enduring belief which guides actions and judgments across specific situations and beyond immediate goals to more ultimate end-states of existence” (p. 16). Values are the forms that store and express culturally determined knowledge. Values are regarded as the deepest of the four manifestations of culture (Hofstede & Hofstede, 2005), because they are the basis on which attitudes, cognition, emotions, and behavior evolve (Hills, 2002). Values are abstract forms of social cognition that serve as powerful explanations of, and influences on, human behavior (Homer & Kahle, 1988).
Value can also be defined etically or emically. McCracken’s (1988) interpretation of cultural principles demonstrated an emic understanding of value. Individual behavior embodies and expresses cultural principles, according to which phenomena are organized, evaluated, and constructed. Hofstede’s (1991) landmark study of the dimensions of culture exemplified an etic approach. Based on an enormous database of the behavior of residents from 85 countries, Hofstede (1991) portrayed the differences in national culture with five distinct culture elements. They are power distance, individualism versus collectivism, masculinity versus femininity, uncertainty avoidance, and long-term orientation (Hofstede, 1991). Because of his extensive examination of cross-cultural behavior, consumer behavioral researchers have frequently adopted his conceptual framework for empirical studies.
Rituals are “expressive, symbolic activity constructed of multiple forms of behavior that occur in a fixed, episodic sequence, and that tend to be repeated over time” (Rook, 1985, p. 252). Rituals are pervasive, being constantly performed by all members of a society. Rituals are important for consumer behavior. On the one hand, rituals originate the cognitive schemata and scripts of consumers, which can subsequently reinforce ritualistic behavior. On the other hand, rituals involve the consumption of goods and services (Luna & Gupta, 2001).
Heroes are “persons, alive or dead, real or imaginary, who possess characteristics which are highly prized in a culture, and who thus serve as models for behavior” (Hofstede, 1997, p. 8). The concept of hero can also include reference groups and opinion leaders. Heroes may influence consumer behavior through their association with certain products and brands.
Symbols are “words, gestures, pictures or objects that carry a particular meaning which is only recognized by those who share the culture” (Hofstede, 1997, p. 7). They are a broad category of processes and objects that carry meanings unique to a particular group of people (Geertz, 1973). Symbols can be easily developed and copied, and they are, therefore, the most superficial level of culture.
Culture in its various manifestations has significant impact on tourist behavior, and studying the topic of cross-cultural tourist behavior is pertinent because tourism is an international industry. The importance of examining cultural differences is twofold. On one hand, the increasing mobility and the growth of transcontinental communication media has accelerate the globalization of the tourism and travel industry in the past decade. The whole industry has gradually shifted from Western-dominant to be more internationalized (Reisinger & Turner, 2003). In this sense, the cross-cultural differences not only act as tourist attractions but also pose challenges to the practitioners. It is therefore imperative for the managers to understand cross-cultural issues in tourism and hospitality industry to respond promptly in policy making and compete successfully for market share.
On the other hand, from the perspective of tourists, their behavior is culturally bound. Individuals are programmed with the cultural values rooted in their normal environment and the cultural traits become part of their personality (S. Kim & McKercher, 2011), which will influence their behavior. It was generally assumed by some researchers and tourism practitioners that tourist holiday experience and satisfaction is determined by material and physical needs and derives from operational buying motives such as the level of service provision (Reisinger & Turner, 2003). However, the ability to attract and satisfy certain markets is actually highly dependent on the tourists’ cultural interpretation of the product, and different societies hold varied explanations of the product provided due to cultural diversity. Therefore, it is critical to analyze the cultural differences to understand which of the differences have the most influence.
In the realm of cross-cultural consumer behavior research, historical reviews of the methodological approaches have suggested that existing studies have yet to resolve many conceptual and methodological issues (Lenartowicz & Roth, 1999). Additionally, the assessment or identification of a proper cultural unit, which was identified as a vital issue of cross-cultural research by Rick, Toyne, and Martinez (1990), remains largely unexplored. Recognizing the aforementioned limitations, based on a review and evaluation of current culture assessment approaches, Lenartowicz and Roth (1999) identified four basic approaches to culture assessment: ethnological description, use of proxies (validated regional affiliation), direct values inference, and indirect values inference.
An ethnological approach generically refers to qualitative approaches used to identify and/or compare cultures. This approach, which guides emic studies of culture, provides a descriptive appraisal of cultures. The core concept of this approach is that cultures are so complex that they cannot be measured but merely observed and described. Through the thorough description of cultural groupings and cultural characteristics, ethnological description can provide rich information that is often extremely useful in formulating research hypotheses. This approach can help identify the unit of analysis of culture, which is fundamental to cultural studies. Examples of the ethnological approach include Hall’s classification of high- and low-context cultures and Gannon’s cultural metaphor approach (see, Soares, Farhangmehr, & Shoham, 2007). This approach has been rarely used in hospitality and tourism studies.
Grounded in the concept of national character and the premise that core cultural values are learned during childhood, the second approach, involving the use of proxies or validated regional affiliation, defines culture based on characteristics that reflect and resemble culture (Lenartowicz & Roth, 1999). The common proxies that are used include nationality, place of birth, and country of residence. This approach can be used at different levels of culture, including group levels, organizational levels, national levels, or a group of nations, such as the European Union, Asian nations, or Western nations. Using those proxies to operationalize culture is very common in business as well as tourism and hospitality literature because nationality, place of birth, or country of residence can be easily identified along clear geographic boundaries. For example, Sussmann and Rashcovsky (1997) studied the cross-cultural differences on general travel behavior between English-speaking and French-speaking Canadian using language as cultural proxy. Yu and Ko (2012) examined the perception of and possible participation in medical tourism among Chinese, Japanese, and Korean respondents with nationality as surrogate for culture.
However, this approach has been criticized for the absence of measures to test hypothesized relationships between the dependent variables and culture. The proxies can only serve as nominal data. Lenartowica and Roth (1999) suggested that the use of the regional affiliation should be granted by two conditions. First, the sociodemographic variables should be controlled through either sample design or the use of covariates. Second, the respondents should spend their childhood in the country of interest if the research includes individual-level measures.
Direct values inference assumes a values-based conceptualization of culture. This approach reflects the idea that culture is a set of learned characteristics shared by a particular group of people. There are a number of different values models in the literature to support the direct values inference approach to cultural assessment. In addition to Hofstede’s five value dimensions of culture, Rokeach (1973) developed the Rokeach Values Survey (RVS), which consists of 18 instrumental values (ideal modes of behavior) and 18 terminal values (ideal end-states of existence). Kahle (1983) developed a shorter version of RVS, called List of Values, to address the problem of ranking and relevance to daily life. RVS is the value model used the most in tourism and hospitality literature. For example, Luk, de Leon, Leong, and Li (1993), using RVS, examined tourist expectations for the quality of organized tour service. Visitors were grouped by three value factors: sociability, ability, and creativity.
Lenartowica and Roth (1999) suggested three methodological considerations with respect to direct values inference. First, sociodemographic variables should be controlled or large samples must be used to randomize sociodemographic effects. Second, value model instruments should be relevant and understandable to the subjects. Finally, due to the fact that empirically there might be multiple combinations of relatively homogenous groups, direct values inference is insufficient for defining cultural groups.
Indirect values inference uses secondary data to ascribe characteristics of cultural groupings. The cultural characteristics identified in other studies are extrapolated to the subjects after their culture is classified by one of the aforementioned proxies. Hofstede’s cultural dimension scores are the most commonly used benchmarks and are widely adopted in tourism and hospitality cross-cultural studies. For example, Quintal, Lee, and Soutar (2010) investigated the cross-cultural differences of tourists’ information search behavior along Hofstede’s cultural dimension of uncertainty avoidance. Tourists from Japan, Australia, and China were selected to represent high uncertainty avoidance, medium uncertainty avoidance, and low uncertainty avoidance, respectively.
The major concern with this approach is the potential measurement error arising from sampling: the characteristics of the group assessed by the benchmark study may differ from the surveyed sample. To address this problem, the sample size of the benchmark study and the undertaken study should be large enough to randomize the effects of variables influencing the values, or the research sample characteristics should be congruent with the benchmark studies. Because the samples of the existing benchmark studies (e.g., Hofstede, 1980) are not representative of the population, the first alternative validating means is not viable, and the remaining alternative is to follow the benchmark’s sampling design (Lenartowicz & Roth, 1999). Lenartowica and Roth (1999) summarized the strengths and weaknesses of the four approaches, as shown in Table 1.
Summary of Methods to Assess Culture
Source: Lenartowica and Roth (1999).
Note: ED = ethnographic description; VRA = validated regional affiliation; DVI = direct value inference; IVI = indirect value inference.
Another key methodological concern in cross-cultural research is data equivalence. The comparability of respondent data is critical, regardless of whether the research is conducted in one country or simultaneously in a number of countries (Bahalla & Lin, 1987). However, traditional approaches in marketing research have typically concentrated only on linguistic equivalence. Scholars in comparative social psychology have suggested four types of equivalence: construct equivalence, operationalization equivalence, scalar equivalence, and linguistic equivalence (e.g., Anastasi, 1982; Berry, 1976). In addition, some scholars (e.g., Aulakh & Kotabe, 1993) suggested that the sample equivalence, which is concerned with whether or not the sample from each culture is comparable, should be ensured. Linguistic equivalence refers to the consistent meanings and ideas of instruments across cultures. This type of equivalence is often satisfied by back-translation.
Construct equivalence is concerned with whether the studied construct exhibits similar meaning across different cultures, using three distinct components. The first is conceptual equivalence, which refers to the interpretation that people in different cultures place on the construct. The second component is functional equivalence, which depicts the similarity of behavioral goals across different cultures. The third one is category equivalence, which represents the grouping of objects, stimuli, and behaviors (Douglas & Craig, 1983). Construct equivalence could be satisfied through different techniques. For example, Mattila and Patterson (2004) used university sample to ensure the functional equivalence and in-depth interview to meet the requirement for conceptual equivalence.
Operationalization equivalence, also called measurement equivalence, is concerned with whether the construct is measured the same way across different cultures. Operationalization equivalence closely connects with item equivalence, which is a more concrete and microlevel perspective. To establish item equivalence, the construct should be measured by the same instrument. Gilbert, Veloutsou, Goode, and Moutinho (2004), in their study examining the service satisfaction of fast-food establishment in four English-speaking countries, checked the measurement equivalence by conducting four-factor analyses in different country samples. The factors extracted were compared among nations. Scalar equivalence, also called metric equivalence, examines whether the psychometric properties of data from various cultures exhibit the same coherence and structure. Cross-cultural comparison is meaningful only when the numbers on the response scales have the same meaning across cultures (Sin, Cheung, & Lee, 1999). Nonequivalence can be detected by structural equation modeling, optimal scaling, regression analysis, or by comparing the standard deviations and means of the subject responses over a large number of items across cultures (Sin et al., 1999). The structural equation modeling is the method used the most by scholars in hospitality and tourism (e.g., Quintal et al., 2010; Tsaur, Lin, & Wu, 2005).
Considering these aforementioned theoretical and methodological concerns, this study will first provide an overview of published articles, then examine the cultural assessment approach used and check the data equivalence, and, finally, provide recommendations for the future study of cross-cultural consumer behavior in hospitality and tourism.
Method
For the purpose of this study, cross-cultural tourist research is defined as any study that investigates tourist behavior in more than one culture/subculture. This research domain corresponds to “comparative articles” in Alder (1983) or “comparative international” studies in Alder and Bartholomew (1992). The articles were drawn from four major databases: Hospitality and Tourism Complete, Sage Journals Online, ScienceDirect, and Emerald Insights.
“Cross-culture/cultural,” “cross-nation/national,” “tourism,” “hospitality,” “hotel,” and “restaurant” were used as key words, and only the articles with “cross-culture/cultural” and/or “cross-nation/national” in the title, abstract, or key words were selected. After retrieving the articles from the database, two researchers first read the abstracts to ensure the appropriateness for the current study and then read the texts if the abstracts did not provide enough information. This procedure resulted in a total of 91 articles.
A preliminary coding sheet was then prepared to evaluate each article along the following dimensions: cultural assessment approach, topic studied, data equivalence, research design, and methods of analysis. The evaluative criteria were derived from a comprehensive review of the cross-cultural methodology literature (e.g., Alder, 1983; Brislin, Lonner, & Thorndike, 1973; Green & White, 1976; Malhotra, Agarwal, & Peterson, 1996; Triandis & Berry, 1980).
Cross-Cultural Tourist Behavior Studies
Number of Articles Published
A total of 91 articles were selected for this study. Detailed accounts of the articles selected are presented in Table 2. Selected articles were published in 21 refereed journals (Table 3). The number of articles published in each journal varied from a low of 1 article to a high of 21 articles. Tourism Management published the largest number of cross-cultural consumer studies in tourism and hospitality, followed by Journal of Travel & Tourism Marketing (10) and International Journal of Hospitality Management (8). The articles reviewed covered a period of 24 years, from 1988 to 2011 (Table 4). The 21st century has exhibited an increasing interest in cross-cultural study from academia. Among the 91 articles analyzed, 79% were published after 2000 (Table 4).
Summary of Cross-Cultural Consumer Research in Hospitality and Tourism
APJTR = Asia Pacific Journal of Tourism Research; ATR = Annals of Tourism Research; IJHM = International Journal of Hospitality Management; IJHTA = International Journal of Hospitality & Tourism Administration; IJQRM = International Journal of Quality & Reliability Management; IJTR = International Journal of Tourism Research; JBR = Journal of Business Research; JHLM = Journal of Hospitality and Leisure Marketing (JHMM = Journal of Hospitality Marketing & Management); JHTM = Journal of Hospitality and Tourism Management; JHTR = Journal of Hospitality and Tourism Research; JICM = Journal of International Consumer Marketing; JIHLTM = Journal of International Hospitality, Leisure & Tourism Management; JLR = Journal of Leisure Research; JPRA = Journal of Park and Recreation Administration; JST = Journal of Sport and Tourism; JQAHT = Journal of Quality Assurance in Hospitality & Tourism; JSM = Journal of Services Marketing; JTR = Journal of Travel Research; JTTM = Journal of Travel & Tourism Marketing; JVM = Journal of Vacation Marketing; LS = Leisure Sciences; MSQ = Managing Service Quality; SJHT = Scandinavian Journal of Hospitality and Tourism; T = Tourism; TG = Tourism Geography; TM = Tourism Management.
ED = ethnographic description; VRA = validated regional affiliation; DVI = direct value inference; IVI = indirect value inference.
LE = linguistic equivalence; CE = construct equivalence; OE = operationalization equivalence; ME = metric/scalar equivalence; SE = sampling equivalence.
Publication Journalstc
Year of Publication
Cultural Assessment Approach
The approach of cultural assessment is shown in Table 5. Except for the three conceptual studies, validated regional association has been widely employed to operationalize culture. A total of 70 out of 91 (76.9%) articles used nationality (49), region (7), country of residence (7), ethnicity (4), or language (3) as a surrogate for culture. The largest number of studies compared cultural differences as reflected in tourists’ behavior across Asian, European, and American countries (14), followed by between Asian countries and American countries (13), between European countries and American countries (8), and among European countries (7). Several studies investigated cultural differences among Asian countries (6).
Cultural Assessment Approach
Note: ED = ethnological description; VRA = validated regional affiliation; DVI = direct values inference; IVI = indirect values inference.
To establish a valid comparison of consumer behaviors between different cultures, cultures should be selected on a theoretical basis. However, the articles reviewed lacked an apparent process or underlying theory for the selection of countries. Countries were largely chosen due to convenience and availability of data. The most studied countries/regions were, not surprisingly, the United States (34), followed by Japan (25), the Greater China area (including Hong Kong and Taiwan; 24), the United Kingdom (23), and South Korea (16).
Table 6 chronologically summarizes the number of articles for each country/region. Scrutinizing the correlation between countries/regions studied and year of publication reveals that academic inquiries into cross-cultural consumer in hospitality and tourism started from comparisons between highly industrialized cultures (e.g., the United States and the United Kingdom) and Asian culture, represented by Japan. The curiosity about the “Eastern Others” from Western society further expanded to South Korea and the Greater China area in late 1990s. After 2000, more and more studies investigated the differences between Chinese culture (as represented by Mainland Chinese, Taiwanese, or Hong Kongese) and other cultures. All but nine articles included English-speaking countries as study objects.
Countries/Regions Studied
Note: China = includes Mainland China, Hong Kong, and Taiwan; NA = North America; A = other Asian countries, including Indonesia, Singapore, Jordan, Israel, Turkey, Philippines, Thailand, and Malaysia; SA = South American countries, including Brazil and Jamaica.
Five articles used direct values inference (5.5%) to assess culture. Two articles used Hofstede’s cultural dimensions (Bowden, 2006; Liu, Furrer, & Sudharshan, 2001), and another three applied the RVS (Luk et al., 1993; Reisinger & Turner, 1998a, 1998b). Eleven studies (Crotts & Erdmann, 2000; Jo & Sarigollu, 2007; C. Kim & Lee, 2000; Litvin & Goh, 2003; Litvin, Tan, Tay, & Aplin, 2004; Lord, Putrevu, & Zheng, 2005; Mattila & Patterson, 2004; Money & Crotts, 2003; Quintal et al., 2010; Tsaur et al., 2005; You, O’Leary, Morrison, & Hong, 2000) employed Indirect Values Inference, and they all adopted Hofstede’s cultural dimensions to assess culture. Among the articles that used Hofstede’s cultural dimensions, the most widely adopted dimension was individualism/collectivism (6), followed by uncertainty avoidance (4). The ethnography approach has yet to be adopted to assess culture.
Range of Topics
A majority of the studies explored cross-cultural impacts on tourist behavior in the context of tourism, and only seven studies were conducted in the context of hotels and/or restaurants (Table 7). A wide range of research topics were investigated in a cross-cultural context. The most popular topic was service quality. Twenty-four out of 91 articles (26.4%) investigated cross-cultural impact on tourist expectations of, evaluation of, and satisfaction with service quality. Other topics included tourism motivation (9), tourist information search behavior (8), travel behavior (e.g., trip characteristics, expenditure and trip planning; 8), cross-cultural impact on destination/restaurant selection (7), perception/image (7), attitude/behavioral intention (7), and tourist behavior as perceived by employees (5).
Range of Topics
Research Method
Cross-cultural tourist behavior researchers have most widely used survey research (n = 60 or 65.9% of the studies). Other designs, used less frequently, include secondary data (n = 17 or 18.7% of the studies), experiment (n = 6 or 6.6%), and interview (n = 3 or 3.3%). The overreliance on survey research is consistent with the general methodological trend in tourism marketing and tourist behavior research.
Table 8 shows the data equivalence check. Of the 67 studies requiring translation, 24 studies used back-translations in the instrument design (26.4% of the studies), whereas the remaining studies either employed direct translation (n = 7 or 7.7%) or did not report a language equivalence procedure. Seven studies (7.7%) examined sampling equivalence, two studies (2.2%) used techniques to test for metric equivalence, and another two studies (2.2%) checked construct equivalence. There were two articles that checked operationalization equivalence.
Data Equivalence Check
Note: LE = linguistic equivalence; CE = construct equivalence; OE = operationalization equivalence; ME = metric/scalar equivalence; SE = sampling equivalence; N/A = not applicable.
With regard to the analytical techniques employed in empirical research, a wide range of statistical techniques were adopted in the studies reviewed, varying from cross-tabulation to ANOVA to covariance structural analysis. Most of the studies employed more than one statistical technique for analysis. The most widely used analytical techniques were factor analysis and ANOVA.
Key Contributors
Although a large number of scholars dabbled in the area of cross-cultural tourist research, only a few of them made continuous effort to the subject and demonstrated leadership in this area. Table 9 showed the list of scholars who led more than one article, as well as the research topics and year of publication. Three types of contributions can be identified from Table 9. The first is to replicate similar research in different cross-cultural contexts. Articles included in this group are Chen’s (2000, 2000) and Gursory’s (2000, 2000) studies on external information sources, Pizam’s (1995, 1996, 1997) studies on tour guides’ perception on tourists behavior, and Reisinger’s (1997, 1998) comparative studies on cultural differences between tourists and Australian hosts. Scholars in the second category extensively explored the cross-cultural differences of varied behavioral constructs. Most of the key contributors belong to this category, including M. Kozak (2001, 2002, 2004), Hudson (2001, 2010, 2011), S. Kim (2002, 2005, 2011), Bowden (2003, 2006), and Litvin (2003, 2004). In the third category, scholars (D. Kim & Park, 2007; D. Kim, Wen, & Doh, 2010; Mattila, 1999, 2000; Mattila & Patterson, 2004) intensively explored different aspects of certain behavioral construct, and therefore made significant and incremental contributions to the area of interest.
Analysis of Lead Author
Discussion and Conclusion
Recognizing the increasing attention to cross-cultural tourist behavior from academia and industry, this study reviewed 91 cross-cultural tourist behavior articles. Given the growing importance of understanding the effects of culture in tourist behavior as well as cross-cultural comparison, it was expected that there would be an upward trend in the number of cross-cultural tourist behavior studies appearing in major hospitality and tourism journals. However, the findings of this study did not support that assumption. Although research on generic marketing and consumer behavior has shifted significantly over the past decade toward global or international topics, cross-cultural consumer research in hospitality and tourism remains largely neglected in the scholarly journals, despite the fact that travel and tourism is an international phenomenon. The findings of the current study revealed that the first cross-cultural tourist behavior study was published in 1988, and there have been only 91 articles published since, contributing to a very small percentage of the overall publications in hospitality and tourism journals.
Cross-cultural tourist behavior research has received more attention from academia since 2000, with more than 79% of the articles published since that year. Tourism Management published the most cross-cultural tourist studies. Among the 91 studies, 76.9% operationalized culture by regional affiliations, including nationality, region, language, country of residence, and ethnicity. Thirty-four articles included the United States as one of the study objects. Hofstede’s landmark study of the cultural dimensions was widely adopted in studies using Direct Value Inference or Indirect Value Inference to assess culture. Of the 11 articles that adopted direct values inference to assess culture, seven employed Hofstede’s work.
Service quality received the most attention from cross-cultural tourist behavior researchers. A total of 96.7% of the published are empirical studies, although theoretical work remains at a consistently low level of 3.3% of published articles. Questionnaire surveys were the most widely adopted means of data collection. This finding is consistent with trends in generic cross-cultural consumer behavior research. Despite the fact that more than 40% of the studies in consumer research employed experimentation, only 9.4% used experiments, whereas the majority (72.7%) used questionnaire surveys. The high percentage of surveys in cross-cultural consumer research may be attributed to the efficiency of the method (Sin et al., 1999). Data equivalence was not extensively addressed in the current literature.
Despite contributions to the understanding of tourist behavior in different cultural contexts, existing studies suffer from the absence of theoretical framework and a lack of methodological rigor. Most of the cross-cultural tourist behavior studies did not mention which definition of the term culture was adopted, showing that the concept of culture has yet to be properly operationalized. Although there are many categorizations of culture or cultural dimensions, only two (i.e., Hofstede’s cultural dimensions and the RVS) have seen limited application in cross-cultural tourist research. Cross-cultural tourist research has been hampered by the common use of nationality as a surrogate for cultural affiliation, and the terms culture, country, nation, and society are often used interchangeably. The use of those collective cultural proxies as discriminating variables for explaining differences in tourist behavior assumes that cultural homogeneity exists within national or ethnic boundaries. Such operationalization, however, overlooked layers of culture and shortchanges the richness of the cultural concept. Observed differences may be due to many effects other than culture, such as contextual factors, which leads to erroneous conclusions (Nakata & Pokay, 2004).
In addition, as pointed out by Craig and Douglas (2006), culture is no longer a phenomenon defined by geographic or political boundaries because the world is becoming increasingly deterritorialized and penetrated by elements from other cultures. These forces result in cultural contamination, cultural pluralism, and hybridization. It is therefore suggested that future study must carefully specify and define the appropriate unit of analysis and move beyond national culture to incorporate other important cultural components. These factors include the abstracts of intangible elements of culture, such as values and belief systems; the communication links that bind and perpetuate a cultural system; and the material aspects of culture, such as symbols and rituals.
Cross-cultural studies are often subject to severe manifestations of ethnocentrism (Boyacigiller & Adler, 1991), and cross-cultural tourist studies are no exception. Despite the attention paid to linguistic issues in instrument development via back-translation, the instrument may yet measure different attributes across cultures due to nonequivalent conceptualizations in the construct. Existing studies include an a priori assumption that the measures used will be universally applicable to other cultures (van Raajj, 1978). Currently, many measurement scales used in cross-cultural studies were developed in the United States and translated into local languages to measure the construct in culturally diverse groups. However, in addition to the linguistic problems in the translation process (Ellis, 1989), whether the measurement used was interpreted in the same way among culturally diverse groups is still questionable. Because meaningful comparisons across groups must be based on common measurements, validity of the instrument and measurement equivalence must be established in each culture before conducting cross-cultural studies. Otherwise, similarities and differences across cultures are open to alternative discussions.
Only two studies in this review addressed metric or scalar equivalence. The responses of participants to rating scales tend to be culturally bounded and affected by social norms of responding and the cultural background (Hui & Triandis, 1985). The same rating scales could be interpreted differently, and the same numerical value on the measurement scale may represent different levels of the constructs across cultures. Sampling equivalence is another area ignored by cross-cultural tourist researchers, with only seven studies assessing sampling equivalence. The cross-cultural differences may therefore be attributed to dissimilar samples.
Cross-cultural research is not merely an extension of domestic research, and researchers face several complex methodological issues, with nonequivalence as the most critical. Failure to address the equivalences adequately can lead to confounding explanations and severely limit the validity and reliability of the study. Future study on cross-cultural tourist research should therefore make the cross-cultural equivalence as prerequisite for comparisons across cultural and ethnic boundaries and employ varied techniques to address different types of data equivalence. For example, the measurement equivalence could be assessed by multisample confirmatory factor analysis, as suggested by Steenkamp and Baumgartner (1998), and the sampling equivalence could be assured by using the same sampling frame (e.g., student samples) or employing more rigorous sampling methods.
Current studies covered a wide array of topics with an emphasis on service quality mainly in the context of destinations, whereas hotels and restaurants have been largely ignored with only few exceptions (e.g., D. Kim & Park, 2007; D. Kim et al., 2010; Mattila, 1999, 2000). Future studies are therefore suggested to carry out in the settings of hotels and/or restaurants where intensive cross-cultural host–guest and guest–guest encounters exist. In addition, there is a lack of continuous effort of investigation and research lines that examine similar topics from different perspectives have yet been established as most of existing studies were conducted based on convenience and data availability. Future studies are warranted to address this limitation with more rigorous research design particular for the cross-cultural issues. With regard to the research topics, it was found from the review that previous studies focused more on tourists’ experience in the destination/hotel/restaurant, whereas the previsit decision-making process has only been investigated by a handful of researchers (e.g., Kay, 2009; M. Kozak, 2002; Laing & Crouch, 2005). Investigations into previsit behavioral constructs such as motivation or expectation are therefore called for.
The current study highlighted and discussed theoretical and methodological issues in cross-cultural tourist study, also making recommendations for future research regarding the interplay of culture and tourist behavior. This study, to the best knowledge of the author, is one of the first attempts to provide an overall review of this increasingly important topic and herein the contribution to the body of knowledge of cross-cultural tourist research. There are, of course, some limitations in this study that could be addressed by future research of similar nature. First of all, this review study was limited to journal articles, and book chapters were excluded. In addition, although four major databases containing hospitality and tourism journals were consulted, the search for relevant journal articles may not be complete because some journal articles may not indicate the nature of study in their titles, keywords, or abstracts. Given the limited resources, however, it is difficult to implement a journal-by-journal and article-by-article search to include all relevant articles. Second, this study limits its scope to English publications. However, due to the nature of the topic, many insights might be found in publications in other languages.
With globalization and technology innovation, the world is diversifying culturally. In addition, cultural boundaries are being deterritorialized and blurred. These phenomena necessitate cross-cultural research in tourism and hospitality because human behavior is culturally biased. However, the review of the extant literature shows that cross-cultural tourist research is still in its infancy. Investigations into how tourists behave differently across cultures and how previously developed theoretical models are interpreted by tourists from different cultural background will have tremendous potential for developing insights into the tourist behavior literature. However, future research should be more conscious of adequately addressing the absence of a theoretical framework and the lack of methodological rigor, as identified by this review study.
