Journal of Travel Research

http://jtr.sagepub.com/ Predicting Souvenir Purchase Intentions

Soyoung Kim and Mary A. Littrell Journal of Travel Research 1999 38: 153 DOI: 10.1177/004728759903800208 The online version of this article can be found at: http://jtr.sagepub.com/content/38/2/153

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

On behalf of:

Travel and Tourism Research Association

Additional services and information for Journal of Travel Research can be found at: Email Alerts: http://jtr.sagepub.com/cgi/alerts Subscriptions: http://jtr.sagepub.com/subscriptions Reprints: http://www.sagepub.com/journalsReprints.nav Permissions: http://www.sagepub.com/journalsPermissions.nav Citations: http://jtr.sagepub.com/content/38/2/153.refs.html

Downloaded from jtr.sagepub.com at SCD Universit de Savoie on April 28, 2011

NOVEMBER JOURNAL OF1999 TRAVEL RESEARCH

Predicting Souvenir Purchase Intentions

SOYOUNG KIM AND MARY A. LITTRELL

Structural equation modeling was used for predicting souvenir purchase intentions of 277 U.S. female tourists traveling to Mexico. The respondents evaluated five textile products and described their travel behaviors. Findings suggested a sequence of significant relationships between hedonic values, world-mindedness, recreational and ethnic tourism, attitude toward souvenirs, and purchase intentions. United States tourists expenditures on both domestic and international tourism grew a dramatic 75% during the 10 years between 1985 and 1995 (Travel Industry Association of America 1996). Of these expenditures, approximately one-third are devoted to shopping (Littrell et al. 1994). Ross and Iso-Ahola (1990) identified six major motives that were important to sightseeing tourists and found that shopping was one distinct motive. As Kent, Shock, and Snow (1983) commented, To be able to peruse, to examine, to feel and think of the joys derived from purchasing certain merchandise is indeed pleasurable to millions of people, and for them is a minor, if not a major reason for travel (p. 2). Shopping is becoming increasingly important on trips of a leisure nature; it is correlated with other tourism styles as well (JansenVerbeke 1991; Littrell et al. 1994; Timothy and Butler 1995). Shopping is also considered a primary motive for undertaking a trip across international borders (Jansen-Verbeke 1991; Timothy and Butler 1995). Particularly, shopping opportunity is a major attraction that draws tourists to many developing and less developed countries where the prices of goods are generally low (Keown 1989). Even though some people may not give a high priority to shopping as a significant criterion when initially selecting a travel destination, they do shop as they travel (Travelers Notes 1995). In a national survey conducted by the Travel Industry Association of America, 44% of all respondents reported shopping as one of their travel activities (Cook 1995). The main products purchased by tourists are local handicrafts, clothing, and books (Travelers Notes 1995). The possession of travel souvenirs is very important to tourists as tangible evidence of the travel experience (Littrell et al. 1994, p. 3). Tourists may also purchase nonsouvenir items, including apparel and durables, during travel. Given the attention that tourists devote to shopping, it is important that tourism marketers expand their understanding of souvenir-purchasing behavior. Empirical evidence on the usefulness and applicability of prominent consumer behavior models such as the Engel, Kollatt, and Blackwell model (Engel, Blackwell, and Miniard 1993) suggests that tourists souvenir-purchasing behavior may be formulated through a process similar to

general consumer behavior models. That is, as with consumer behavior more generally, tourists souvenirpurchasing behavior may be influenced by such variables as values and attitudes. A number of studies have examined tourists psychological characteristics, such as personal values, and their travel activities, including physical involvement and level of interaction with host communities. In general, these studies were directed at describing tourism styles based on such variables. However, no studies were found that focus on predicting the relationships between tourists psychological characteristics, travel activities, and souvenirpurchasing behavior. The present study was intended to explore such predictive linkages. International tourism is ripe for growth (U.S. Congressional House Committee on Small Business 1994). The World Tourism Organization estimates that international tourist arrivals worldwide will reach 661 million in the year 2000 and 937 million by 2010 (Vellas and Bcherel 1995). Despite the significance of international tourism, few studies have dealt with souvenir-purchasing behavior within an intercultural context. This research seeks to explore international tourists souvenir-purchasing behavior by studying U.S. tourists in a situation of traveling to another culture Mexico. In many developing countries, tourism is a major economic factor (Vellas and Bcherel 1995). During recent decades Mexico has been one of the major destination countries for international tourism. The 1992 international tourism expenditure in Mexico reached $6.1 billion, with an average annual growth rate of 3.2% in the period 1980 to 1992 (Latham 1995). International tourists bring into Mexico approximately $1.5 billion more than Mexicans spend abroad (van den Berghe 1995), placing tourism consistently among the top three earners of foreign receipts in Mexico (U.S. Department of Commerce 1993). In addition, the hospitality industry in Mexico employs about one-tenth of the labor force (van den Berghe 1995). Considering that more than 90% of its visitors are from the United States (Latham 1995), it is critically important for the Mexican tourism
Soyoung Kim is an assistant professor of human environments at Utah State University in Logan. Mary A. Littrell is a professor of textiles and clothing in the Department of Textiles and Clothing at Iowa State University in Ames. This research was part of the first authors doctoral dissertation at Iowa State University. The authors wish to thank Dr. Frederick Lorenz for his assistance with the data analysis. Partial support for the project came from the Louise Rosenfeld International Fund and the College of Family and Consumer Sciences, Iowa State University.
Journal of Travel Research, Vol. 38, November 1999, 153-162 1999 Sage Publications, Inc.

Downloaded from jtr.sagepub.com at SCD Universit de Savoie on April 28, 2011

154

NOVEMBER 1999 FIGURE 1 PROPOSED MODEL FOR PREDICTING SOUVENIR INTENTIONS

industry to understand U.S. tourists and to develop appropriate tourism promotions, activities, and products. The purpose of this research was to better understand international tourists travel and souvenir-purchasing behavior in order to predict their intention to purchase souvenirs. Specifically, we proposed a model incorporating hypothesized relationships between the variables that may affect international tourists travel and souvenir-purchasing intentions. The model is depicted in Figure 1. The exogenous variables selected for the study were hedonic values (as an indicator of personal values) and world-mindedness (as an indicator of attitude toward other cultures). The endogenous variables were tourism styles, attitude toward souvenirs, and intention to purchase souvenirs. In this study the model was tested for U.S. tourists traveling to Mexico.

n A* = bEeE i = 1, where A* = attitude toward the behavior, bi = the belief that performing behavior B leads to consequence i, eE = the evaluation of consequence i, and n = the number of salient consequences. Fishbein (1967) proposed that an individuals attitude toward performing a behavior is the summed set of the beliefs about the consequences of engaging in the behavior weighted by the evaluation of these consequences. In marketing, these consequences are typically represented in the form of product attributes. In a specific purchase situation, therefore, a consumers attitude is defined as being a function of the presence (or absence) and evaluation of a number of product attributes. Finally, Fishbeins (Fishbein and Ajzen 1975) subsequent general model of behavior and intentiontheory of reasoned actionprovided a conceptual framework relating attitude toward souvenirs, intention to purchase souvenirs, and actual purchase behavior. The model suggests that consumer behavior can be predicted by using intentions as the mediator between attitude and overt behavior. Hence, in the present study, the relationship between intention and actual purchase behavior was not examined, but it was assumed that purchase behavior is determined by, and predicted from, the individuals behavioral intention.

THEORETICAL FRAMEWORK Organization Framework

The Engel, Kollatt, and Blackwell (EKB) (Engel, Blackwell, and Miniard 1993) model of consumer behavior served as a theoretical framework for proposing relationships between the variables expected to influence tourists souvenir-purchasing behavior (Figure 2). The EKB model proposes that the product evaluation process leading to purchase decision consists of four key components: beliefs, attitudes, evaluative criteria, and intention. In addition, product evaluation is influenced by individual characteristics, including personality, lifestyles, motives, values, norms, and reference groups (Engel, Blackwell, and Miniard 1993). On the basis of the EKB model, a model was developed for this study that incorporated the souvenir evaluation process and variables influencing souvenir evaluations. Consumers general characteristics such as personal values and attitude toward other cultures were expected to influence their travel-specific characteristics (tourism styles). It was anticipated that these, in turn, would influence souvenir evaluation, including attitude toward souvenirs (a combination of evaluative criteria and beliefs) and intention to purchase souvenirs. In addition to the EKB model, Fishbeins (1967) attitude model was used to define the concept of attitude toward souvenirs. The Fishbein model can be expressed in the following mathematical form:

Exogenous Variables
Hedonic values. Values are centrally held evaluative beliefs that provide the criteria for guiding actions across lifes situations and for judging personal action or the action of others (Rokeach 1969). Studies of values have been concerned with questions of how personal values might influence and guide individuals attitudes and behavior. Few studies have examined the role of personal values in determining travel-related attitudes and behavior. Using a sample of travelers to in-state weekend travel attractions, Pitts and Woodside (1986) observed that seeking pleasure was an important

Downloaded from jtr.sagepub.com at SCD Universit de Savoie on April 28, 2011

JOURNAL OF TRAVEL RESEARCH FIGURE 2 A PARTIAL MODEL OF ENGEL, KOLLATT, AND BLACKWELLS CONSUMER BEHAVIOR MODEL

155

Source: Based on the Shim and Drake (1990) model.

value in differentiating between visitors and nonvisitors to a resort known for sports. In the present study, tourists hedonic values, or the seeking of pleasure, were anticipated to influence their tendency to participate in recreational travel activities. World-mindedness. Attitudes that tourists hold toward other cultures may play a critical role in influencing their international travel experiences. In this research, worldmindedness was included as one dimension of attitude toward other cultures. Sampson and Smith (1957) defined worldmindedness or global-mindedness as a frame of reference, or value orientation, favoring a world-view of the problems of humanity, with mankind, rather than the nationals of a particular country, as the primary reference group (p. 105). Hett (1991) identified five dimensions of world-mindedness including responsibility, cultural pluralism, efficacy, globalcentrism, and interconnectedness. Smith (1955) compared the attitudes and personalities of exceptionally world-minded individuals and exceptionally nationalistic individuals. He found that the world-minded individual was less masculine, more impulsive, easygoing, emotionally dependent, inclined toward introspection, and more likely to belong to organizations with international interests. In addition, world-mindedness had a highly negative association with ethnocentrism. Although the relationship between international tourists level of worldmindedness and their travel behavior in other cultures has not

been studied, world-mindedness may be an important determinant of international tourists tendency to actively interact with local people and show interest in indigenous cultures.

Endogenous Variables
Tourism styles. Typologies of tourism styles were developed on the basis of various travel-related variables such as tourists exposure to novelty during travels, shopping behavior, travel activities, and levels of interaction with host communities. Cohen (1972) was one of the first to suggest the existence of different patterns of tourist behavior. He theorized that there are four tourist roles based on the extent of the tourists exposure to the novelty of the host culture. In his later work, Cohen (1979) proposed that tourists can also be categorized into different types according to their travel experiences, specifically whether they are in pursuit of mere pleasure or in quest of meaning and authenticity. Recreational tourists are not concerned with learning and experiencing other societies or cultures, although they find their travel experience interesting. Diversionary tourists are alienated from their own society but do not seek meaning or authentic experiences elsewhere. For them, travel is a mere escape from boredom or a meaningless pleasure. Experiential tourists are modern, alienated individuals who seek authentic experiences elsewhere. Experimental tourists view authentic experiences as essential but do not become fully immersed in other cultures. Finally, existential tourists, who are the oppo-

Downloaded from jtr.sagepub.com at SCD Universit de Savoie on April 28, 2011

156

NOVEMBER 1999

sites to the recreational tourists, are deeply immersed in other cultures or societies. Smith (1989) projected that, using the different types of leisured mobility, tourism can be categorized into the following styles: ethnic, cultural, historic, environmental, and recreational tourism. Based on Graburns (1989) hypothesis that varying types of tourists may be attracted by different types of souvenirs (Graburn 1989), a typology similar to Smiths (1989) but incorporating souvenirs, was empirically generated by Littrell and her colleagues (1994). Tourists preference for travel activities, souvenir products, and shopping were integrated in four distinct patterns. Ethnic, arts, and people orientation involved travel to ethnic communities where tourists visited local residents, participated in ritual events, visited art galleries and museums, and attended concerts or theaters. Tourists with a strong ethnic, arts, and people orientation perceived tourism as an opportunity for personal development and as a time for seeking authenticity in a community and among its residents. Products native to an area were favored souvenirs. History and parksoriented tourists were engaged in activities focusing on history or natural beauty. Favorite crafts displayed a regional, ethnic, country, or nature-related theme. Urban entertainmentoriented tourists were continuously involved in recreational activities such as shopping, attending a professional sports event, visiting a recreational theme park, or going dancing or to a nightclub. Finally, tourists who enjoyed outdoor activities such as hiking, backpacking, camping, and boating were referred to as active outdoororiented tourists. Tourists in the final two groups purchased T-shirts and sweatshirts with a name or logo symbolizing the vacation destination. The Littrell et al. typology established correlates, but not predictive relationships, among tourism variables. In the present research, two tourism stylesethnic and recreationalwere chosen for further study using a predictive model. Cultural and recreational themes were selected as they are (1) prominently featured in Mexican tourism promotions and (2) common across typologies developed by tourism scholars (Cohen 1979; Littrell et al. 1994; Smith 1989). Attitude toward souvenirs. Two important concepts in Fishbeins attitude model are evaluative criteria, that is, delineation of product attributes that the consumer finds to be important and perception of whether products possess the valued attributes. Several studies have sought to identify the dimensions by which consumers evaluate souvenirs or ethnic textiles and apparel (Littrell 1990; Slaybaugh, Littrell, and Farrell-Beck 1990). Attributes related to physical properties, cultural distinctiveness, and use or versatility are common. Littrell et al. (1994) found that tourists travel activities were closely related to evaluative criteria considered important for selecting souvenirs. Although souvenir evaluative criteria have been identified in several studies, little is known about consumers perception of the relative importance of attributes in souvenirs. Intention to purchase souvenirs. The ultimate dependent variable examined in this study was intention to purchase souvenirs. Intention provides a link between consumers reactions to products and their acquisition or use of the products. Thus, intention has been used as an alternative measure to purchase behavior. While a number of studies in tourism

have described souvenir-purchasing behavior, no studies have examined tourists intention to purchase souvenirs in relation to their attitude toward souvenirs. In the present study, it was expected that tourists favorable attitude toward souvenirs would predict their intention to purchase the souvenirs.

Hypotheses
Based on review of the existing literature, the following hypotheses were proposed and tested: Hypothesis 1: Hedonic values will positively affect recreational tourism. Hypothesis 2: World-mindedness will positively affect ethnic tourism. Hypothesis 3: Ethnic tourism will significantly affect tourists attitude toward souvenirs. Hypothesis 4: Recreational tourism will significantly affect tourists attitude toward souvenirs. Hypothesis 5: Tourists attitude toward souvenirs will positively affect their intention to purchase the souvenirs.

METHOD Sample and Procedure

The population of interest was made up of female adults who had traveled to Mexico or were planning to travel to Mexico in the near future. Female adults were chosen as respondents for this study because women are the more frequent purchasers of souvenirs such as apparel and household textiles (Anderson and Littrell 1995; Littrell et al. 1994). Another reason for choosing female respondents was to reduce the influences or differences related to gender in this predictive study with its array of other exogenous and endogenous variables. Six Mexico government tourism offices and one marketing office in the United States were contacted to provide a list of names for individuals who had requested travel information concerning potential travel in Mexico. After eliminating male names and travel agencies from the list, 900 randomly selected women were mailed a self-administered questionnaire and color photographs of five Mexican textile products. The Dillman (1978) Total Design Method was used in collecting the data, except that a fourth mailing was not used. Of the 900 questionnaires mailed, 51 were returned due to an inaccurate address or because the recipient was of minor age. From the remaining 849 questionnaires, 336 were returned for a response rate of 40%. Of these, 277 surveys were completed by adult women who had traveled to or were planning to travel to Mexico within 1 year. These 277 surveys provided the data for the factor and structural equation modeling analysis. Respondents were between 18 and 76 years old; the average age was 41 years. The sample was comparable in age to the U.S. female population (43 years) (U.S. Bureau of the Census 1997). A large number were Caucasian (83%), followed by Hispanics (6%) and African Americans (3%). Married women represented 63% of the sample, while 31% were single. More than 47 % of the respondents reported an annual household income of $50,000 or above; approximately 26%

Downloaded from jtr.sagepub.com at SCD Universit de Savoie on April 28, 2011

JOURNAL OF TRAVEL RESEARCH

157

had an annual household income between $25,000 and $49,999. The average income for the total U.S. population was $39,255; about 37% had an annual household income of $50,000 or more (U.S. Bureau of the Census 1997). Overall, the sample possessed somewhat higher income levels than the total U.S. population. The majority (83%) of the sample completed education beyond high school and 30% had at least some graduate education. Seven out of 10 (71%) were employed.

Questionnaire: Operationalization of Variables

Hedonic values. Two items measuring hedonic values (pleasure and comfortable life) were selected from Dickson and Littrells (1997) study of consumers of handcrafted apparel. A 99-point rating scale ranging from 1 (Nothing I do is ever based on this guiding principle) to 99 (Everything I do is based on this principle) was adopted ( = .78). World-mindedness. Nine items measuring worldmindedness were adopted from previous studies (Bhawuk and Brislin 1992; Wiseman, Hammer, and Nishida 1989) as well as developed by the researchers (1 = strongly disagree, 7 = strongly agree). Factor analysis using the principal component method of extraction and varimax rotation was used to reduce the number of the items. An eigenvalue of one was used for the purpose of extraction. In addition, an examination of scree tests and conceptual clarity of items included within each factor determined the decision for the number of factors. This procedure for item reduction and determination of number of factors was also employed for tourism styles and attitude toward souvenirs. Factor analysis of the worldmindedness items generated two factors that accounted for 46% of the total variance. The first factor included items indicating general interest in other cultures (e.g., I like to decorate my home or office with artifacts from other countries and I am interested in the history and culture of other countries). This factor was titled Interest in Other Cultures ( = .73). The second factor, called Open-Mindedness, contained items representing global worldview and cultural pluralism (e.g., Culturally mixed marriages are not wrong and Our country should permit the immigration of foreign peoples) ( = .65). Open-Mindedness was an outer-directed perspective on other cultures, while Interest in Other Cultures represented a more emotional, inner-directed way of viewing other cultures. Tourism styles. Twelve items designed to measure ethnic and recreational tourism were adapted from the scale used by Littrell and her colleagues (1994). Respondents rated the level of importance of various travel activities when traveling in other countries (1 = very unimportant, 7 = very important). Factor analysis of 12 travel activities resulted in three factors, explaining 63% of the total variance. The first factor, Ethnic, was composed of items favoring engrossment in the authentic life of communities. Activities included visiting ethnic communities, attending ethnic or community festivals or fairs, and meeting interesting people different from themselves ( = .75). Recreational activities such as shopping, visiting recreational theme parks, bringing home souvenirs from the trip, participating in night entertainment, and taking a complete package tour were loaded on the second factor, labeled Recreational ( = .69). The third factor included items such as attending sports events or concerts and looking for antiques. This factor did not seem to have a common theme and, thus, it was not included in further analysis. Attitude toward souvenirs. Items measuring evaluative criteria for souvenirs were developed by Littrell et al. (1994). Respondents were asked to rate the importance of 14 criteria when selecting souvenirs as they travel (1 = very unimportant, 7 = very important). Respondents then viewed color photographs of five Mexican souvenir items and indicated

Stimuli
Photographs of five Mexican textile products served as the stimuli for the study. Textile products were selected because they are small and easy to pack (Graburn 1989) and are common shopping items among tourists visiting Mexico (Adair 1996). Initially, 13 Mexican souvenirs representing seven different product categories were chosen for pretesting. Selection was based on the researchers observations of markets and shops in Oaxaca, Mexico and consultation with a cultural specialist in Oaxaca. Oaxaca is the destination of choice for tourists seeking colonial surroundings, archaeological sites, and indigenous creativity (Adair 1996); Oaxaca is recognized as a major folk art center for Mexico. Products from Oaxaca are marketed to tourists throughout Mexico. The city has become Mexicos eighth most popular travel destination, following Mexico City and beach destinations such as Cancun and Acapulco (If You Go 1997); tourism is now Oaxacas leading source of economic income (Mader 1998). Criteria for souvenir selection were adopted from previous studies of tourists (Littrell 1990; Littrell et al. 1994; Slaybaugh, Littrell, Farrell-Beck 1990). These included that (1) items be currently on the market and available to tourists and that (2) they be diverse in technique, color, function, design, theme, and price. For 8 out of 13 products, photographs as well as information about fiber content, size, and price were obtained from the 1996 catalog of FONART (Fondo Nacional para el Fomento de las Artes). FONART is a government organization that promotes craft development and marketing in Mexico and abroad. FONART has a network of stores that are located in the largest cities of Mexico and features a wide selection of crafts from all over the country. Although T-shirts and woven belts were among major souvenir items in Mexico, these were not available in the catalog. Therefore, three T-shirts and a woven belt were purchased from stores or markets in Oaxaca. Photographs of these five items not included in the catalog were taken by a professional photographer in a manner to resemble the photographic style of the catalog. The 13 stimuli photographs were pretested with a convenience sample of 24 women employed in faculty and staff positions at the researchers university. Those souvenirs that elicited diverse responses from participants related to purchase intention were selected for the main survey. In addition, the researchers ensured that product variety be maintained for the final stimuli. Five Mexican textile souvenirs representing five different product categories (T-shirt, woven rug, place mat, cushion cover, and woven belt) were selected. Information on the fiber content, size, and price was presented in the questionnaire so respondents would have the same information as would be available in an actual purchase situation.

Downloaded from jtr.sagepub.com at SCD Universit de Savoie on April 28, 2011

158

NOVEMBER 1999 TABLE 1 STATEMENTS MEASURING CONSTRUCTS Construct Hedonic values Interest in other cultures Pleasure Comfortable life I like to decorate my home or office with artifacts from other countries. I love to travel abroad if I have time and money. I am interested in the history and culture of other countries. Item

their perception of the presence of 11 attributes for each souvenir (1 = strongly disagree, 7 = strongly agree). Factor analysis of evaluative criteria resulted in three factors: Aesthetics, Uniqueness, and Care and Travel. The factors accounted for 57% of the total variance. Factor analysis was also performed on perceived attributes separately for five Mexican souvenirs. A three-factor solution was employed to produce factors directly comparable with the three evaluative-criteria factors. As a result of factor analysis, similar factors to those of evaluative criteria were revealed for all five souvenirs, indicating the relevance of the factors across various product types. Only items loading on the same factor across five souvenirs and on the identical factor of evaluative criteria were chosen to create a measure of attitude toward souvenirs, resulting in two items for each factor. Items for the Aesthetics factor were The design is appealing and The colors are appealing, while the items for the Uniqueness factor included It is new, innovative and The product is unique or one of a kind. The items for the Care and Travel factor covered It is easy to care for or clean and It is easy to pack or carry with me on the trip. The three dimensions of attitude toward souvenirs were measured using indices derived from a combination of the evaluative-criteria factor and the perceived-attributes factor. Attitude scores for each dimension appeared highly consistent across the five souvenirs. Therefore, attitude scores were summed across the souvenirs and one model, rather than separate models for the five souvenirs, was run. Intention to purchase souvenirs. Researchers have used the intention-to-purchase scale as an attitudinal measure of likely purchase behavior. Gruber (1971) found a high correlation (r = .95) between purchase intent and purchase probability. To ascertain behavioral intentions with regard to purchasing Mexican souvenirs, respondents were asked to indicate the likelihood that they would purchase each souvenir item (1) for themselves and (2) for others as a gift (1 = very unlikely, 7 = very likely). Intention scores appeared highly consistent for the five souvenirs and, thus, were summed across the five souvenirs. Demographic information. Information on demographic characteristics was obtained for sample description purposes. Respondents were asked about their age, ethnic group, marital status, education, occupation, and income.

Open-mindedness Culturally mixed marriages are not wrong. It would be better to be a citizen of the world than of any particular nation. Any healthy individual, regardless of race or religion, should be allowed to live wherever he or she wants in the world. Our country should permit the immigration of foreign people. Recreational tourism Shopping Visiting recreational theme parks with rides and sports Bringing home souvenirs from the trip Participating night entertainment such as dancing or nightclubs Taking a complete package tour Visiting ethnic communities Attending ethnic or community festivals or fairs Meeting interesting people different than myself The colors are appealing. The design is appealing. It is new, innovative. It is unique or one of a kind. It is easy to care for or clean. It is easy to pack or carry with me on the trip. How likely is it that you would buy this product for yourself? How likely is it that you would buy this product for others as a gift?

Ethnic tourism

Aesthetics Uniqueness Care and travel

Analysis of the Proposed Model

Because the measures used in this study demonstrated good internal consistency, it was decided to use summated scores as single item indicators (Brown, Cron, and Leigh 1993; Good, Page, and Young 1996). Scores for individual factor items were added and then divided by the number of items in the factor. Summated scores were created for all factors, with the exception of factors for souvenir evaluative criteria and perceived attributes. Three measures of attitude toward souvenirs (Aesthetics, Uniqueness, and Care and Travel) were created using Fishbeins (1967) multiattribute mathematical model for combining souvenir evaluative criteria and perceived attributes. Table 1 indicates the statements of the summated scales of each of the nine constructs in the model.

Purchase intentions

The maximum-likelihood estimation procedure was used via LISREL VII (Jreskog and Srbom 1989) to analyze the proposed causal model. Maximum-likelihood techniques estimate the population parameters most likely to have produced the observed data. Compared to path analysis, structural equation modeling is based on less-restrictive assumptions. Structural equation modeling can be applied where there is measurement error, correlated residuals, and reciprocal causation (Pedhazur 1982). In this research, the dimensions under the same variable, such as ethnic and recreational tourism styles, were hypothesized to have no direct

Downloaded from jtr.sagepub.com at SCD Universit de Savoie on April 28, 2011

JOURNAL OF TRAVEL RESEARCH TABLE 2 INTERCORRELATION, MEANS, AND STANDARD DEVIATIONS FOR VARIABLES IN THE PROPOSED MODEL (N = 250) Variable 1: Recreational tourism 2: Ethnic tourism 3: Aesthetics (souvenir attitude) 4: Uniqueness (souvenir attitude) 5: Care and travel (souvenir attitude) 6: Purchase intentions 1: Hedonic values 2: Interest in other cultures 3: Open-mindedness 1 1.0 .04 .26 .38 .28 .24 .28 .06 .13 2 1.0 .27 .04 .16 .21 .14 .48 .35 3 4 5 6 1 2 3

159

M
4.3 5.4 5.4 4.2 5.2 4.0 82.8 5.7 5.2

SD
1.2 1.1 0.8 1.0 0.9 1.2 16.0 1.1 1.2

1.0 .35 .36 .58 .07 .23 .18

1.0 .40 .41 .17 .01 .02

1.0 .34 .06 .08 .05

1.0 .00 .23 .25

1.0 .00 .09

1.0 .43

1.0

association; instead, the residuals were allowed to covary with each other. The analysis was performed with the covariance matrix of the variables. Table 2 shows the intercorrelations, means, and standard deviations for variables in the model. Analysis of the causal model involved three steps. First, the overall fit of the causal model was evaluated using a chisquare statistic, goodness-of-fit index (GFI), adjusted goodness-of-fit index (AGFI), and root mean square residual (RMS). Second, the overall fit of the model was compared with those of alternative models, including a null model and a less restrictive model. Finally, the parameter estimates associated with the research hypotheses were examined.

indicator is the RMS, a measure of the average residual variances and covariances. Generally, models with good fit are characterized by a p-value for a chi-square test > .05, GFI > .95, AGFI > .90, and RMS < .05 (Jreskog and Srbom 1989). The value of chi-square for the proposed model in this study was 38.56 (df = 17, p = .002). The GFI, AGFI, and RMS were .969, .917, and .056, respectively. The 2/df ratio was 2.27. According to Carmines and McIver (1981), ratios of 3, 2, or less represent a good fit. Taken together with these indicators, the fit of the proposed model to the data appeared satisfactory.

Comparison of Models A MODEL OF SOUVENIR PURCHASE INTENTION Proposed Model

The proposed model was tested for its overall fit to the data. Various measures are available to judge overall fit of the structural equation model and, among those, a chi-square statistic, GFI, AGFI, and RMS are most often used. A chisquare statistic measures the difference between the sample variance-covariance matrix and the one reproduced through model estimation; accordingly, a large chi-square value represents a poor model fit. The chi-square statistic alone, however, is not adequate for testing model fit, because it increases with sample size. Therefore, when sample size is fairly large, a model can be rejected even when the model fits the data well; if the sample size is small, a model can be falsely accepted even when it fits the data poorly (Pedhazur 1982). According to Bagozzi and Yi (1988), the chi-square statistic is not a good indicator of fit when N > 200. A chisquare statistic is also sensitive to departure from the multivariate normal distribution assumed by the maximumlikelihood estimation technique (Jreskog and Srbom 1989). Other measures of the overall fit of the model such as GFI and AGFI are independent of sample size and nonnormality. These are indicators of the extent to which the variance and covariance are explained by the model. The AGFI adjusts for the degrees of freedom of a model relative to the number of variables, while the GFI does not. Another important Comparing the fit of the proposed model with alternative models provides added understanding of the acceptability of the proposed model (Anderson and Gerbing 1988). Three nested models, including the baseline model (i.e., the most restrictive model), proposed model, and less-restrictive model, were compared with each other. The baseline model in this study posited complete independence of all observed variables. In the less-restrictive model, five additional paths (i.e., hedonic values [1] ethnic tourism [2], interest in other cultures [2] recreational tourism [1], open-mindedness [3] recreational tourism [1], recreational tourism [1] purchase intentions [6], and ethnic tourism [2] purchase intentions [6]) were freed so that the five direct relationships were newly established. Chi-square comparisons indicated improvements in total fit obtained by one model over another. The difference in chi-square values between the baseline model and the proposed model was significantly large (2 [19] = 439.85, p < .001), suggesting that the proposed model explained the relationships between variables significantly better than the baseline model. The difference between the proposed model and the less-restrictive model was not significant (2 [5] = 9.85, p > .050), indicating that the five additional paths did not provide a significant increase in explanatory power.

Test of Hypotheses
Figure 3 presents parameter estimates of the proposed model. Results showed that three of the five hypothesized relationships were confirmed. Consistent with the first

Downloaded from jtr.sagepub.com at SCD Universit de Savoie on April 28, 2011

160

NOVEMBER 1999 FIGURE 3 PROPOSED MODEL AND STANDARDIZED PARAMETER ESTIMATES

Note: Values in parentheses represent t-values. 2 = 38.56; goodness-of-fit index (GFI) = .97; adjusted goodness-of-fit index (AGFI) = .92; root mean square residual (RMS) = .06.

hypothesis, hedonic values had a direct positive causal influence on recreational tourism (11 = .30, t = 4.98). Tourists who placed greater importance on hedonic values, such as pleasure and comfortable life, were more likely to favor recreational travel activities. Interest in other cultures and open-mindedness were found to have strong positive direct effects on ethnic tourism (22 = .41, t = 6.82; 23 = .18, t = 3.04), in agreement with the second hypothesis. Tourists with greater interest in other cultures and a stronger global and open mind were more likely to participate in ethnic tourism when traveling. As predicted by the third hypothesis, recreational tourism was significantly related to all three dimensions of attitude toward souvenirs. Those actively engaged in recreational tourism held more favorable attitudes toward aesthetic qualities (31 = .25, t = 4.27), uniqueness (41 = .38, t = 6.45), and portability (51 = .28, t = 4.61) of souvenirs. Among the three dimensions of souvenir attitudes, uniqueness was most strongly affected by recreational tourism. By comparison, ethnic tourism significantly affected two dimensions of attitude toward souvenirs. Tourists actively involved in ethnic tourism were more likely to hold favorable attitudes toward the aesthetic qualities (32 = .26, t = 4.41) and portability of souvenirs (52 = .15, t = 2.46). It should be noted that attitude toward the aesthetic aspects of souvenirs was nearly equally influenced by both recreational and ethnic tourism. The final hypothesis predicted the positive effects of attitude toward souvenirs on purchase intentions. Attitude toward the aesthetic qualities and uniqueness of souvenirs exerted strong positive effects on purchase intentions (63 = .47, t = 8.61; 64 = .21, t = 3.76); however, there was no relationship between attitude toward the portability of souvenirs and purchase intentions. Tourists with more favorable attitudes toward the aesthetic aspects and uniqueness of souvenirs tended to have stronger purchase intentions.

CONCLUSIONS AND APPLICATION

The present research represented the first attempt to test and validate a model of tourists souvenir-purchasing intentions. The variables in this study have been examined extensively in past research, but mostly in isolation or through correlation. Studies that explain the structural relationships among the variables do not currently exist. The advantage of incorporating a number of variables within one model is that it facilitates understanding of the relative and disparate effects of the variables. The results, which generally supported the EKB (Engel, Blackwell, and Miniard 1993) model of consumer behavior, suggest continued use of the theoretical frameworks by tourism researchers. Through analysis of the causal model, this research recognized the existence of two distinct dimensions of travel experience, that is, recreational tourism and ethnic tourism, and explored the two styles in relation to their determinants and outcomes. Researchers in tourism have long been interested in developing a typology of travel experiences; however, relatively little empirical research has been done to explore why and how tourists differ. Empirical results in this study suggested that personal values and attitude toward other cultures can help to predict these two distinct patterns of travel experiences. That is, tourists desire for pleasure significantly predicted their tendency to participate in recreational travel activities, such as visiting recreational theme parks and participating in night entertainment. On the other hand, tourists general interest in other cultures explained their importance on such travel activities as visiting ethnic communities and meeting interesting people different from oneself. This finding confirms the attitude-behavior linkage in the context of tourism and suggests that traveling can be viewed as a way of fulfilling individual desires and manifesting personal values.

Downloaded from jtr.sagepub.com at SCD Universit de Savoie on April 28, 2011

JOURNAL OF TRAVEL RESEARCH

161

Findings regarding relationships between attitude toward souvenirs and purchase intention suggest several applications for retailers that are marketing souvenirs to tourists or for tourism promoters who develop print, media, and internet advertising. First, tourism marketers should stress the aesthetic qualities and uniqueness of products from their destinations as a means of attracting international tourists. While tourists favor souvenirs that can be easily transported, portability does not appear to be a final predictor of intent to purchase. Product aesthetics and uniqueness warrant greater attention in tourism promotions, visual merchandising through in-store displays, and by retail sales associates when visiting with tourists. Second, at travel destinations where both ethnic and recreational tourists are attracted, a single strategy focusing on product aesthetics and uniqueness is appropriate. Retailers need not develop separate strategies for marketing souvenirs to each group. Finally, designing shopping environments that offer pleasure to tourists while also broadening their cultural experiences would enhance their travel experiences.

LIMITATIONS
The results of this study should be evaluated in the light of certain limitations. First, respondents evaluations of souvenirs were limited to five stimulus items. Although the five souvenirs were considered representative of Mexican textile souvenirs, they do not reflect the full range of textile souvenirs available in a real shopping situation. In addition, photographs were used in this study instead of actual products. Respondents evaluated the souvenirs based only on visual assessment. It was assumed that other senses such as touch and smell are also important in making purchase decisions for textile souvenirs. Second, the study was limited to U.S. women; the specific setting was the Mexico souvenir market. Thus, findings may not be generalizable to male tourists or to other international tourist populations. Third, methodological issues should be noted. The single summated item, selfmeasure of a variable, is potentially capable of measurement error. In this study, however, the model was analyzed under the assumption of no such error.

Testing the model with Korean tourists would, therefore, reveal meaningful information on how the same variables interrelate for tourists from collectivist cultures. Fourth, in the present study, purchase intention was measured by asking respondents about their likelihood of buying a souvenir for oneself and their likelihood of buying the souvenir for others as a gift. Scores for the two questions appeared strongly correlated with each other across five souvenirs, and thus the summated score was used for the measure of purchase intention. Future research, however, may differentiate between the two types of purchase intention to provide information on whether the relationship between attitude toward souvenirs and purchase intention is influenced by purchase situations of buying for oneself versus buying for others. Finally, while it was beyond the scope of this study to develop hypotheses regarding demographic variables, future research should examine the effects of demographic variables on travel and souvenir-purchasing behavior. The effects of age on tourist behavior were evidenced by several studies (Anderson and Langmeyer 1982; Anderson and Littrell 1995), which found distinctly different trip-planning styles and travel activities among tourists in different age-groups; however, studies that examine the predictive relationship between age and souvenir-purchasing behavior do not currently exist. Therefore, the effects of age along with the effects of other demographic variables should be explored in more detail.

REFERENCES
Adair, M. (1996). Mexico. New York: Macmillan. Anderson, B., and L. Langmeyer (1982). The Under-50 and Over-50 Travelers: A Profile of Similarities and Differences. Journal of Travel Research, 21: 20-24. Anderson, J. C., and D. W. Gerbing (1988). Structural Equation Modeling in Practice: A Review and Recommended Two-Step Approach. Psychological Bulletin, 103: 411-23. Anderson, L. F., and M. A. Littrell (1995). Souvenir-Purchase Behavior of Women Tourists. Annals of Tourism Research, 22(2): 328-48. Bagozzi, R. P., and Y. Yi (1988). On the Evaluation of Structural Equation Models. Journal of the Academy of Marketing Science, 16 (1): 74-94. Bhawuk, D.P.S., and R. Brislin (1992). The Measurement of Intercultural Sensitivity Using the Concepts of Individualism and Collectivism. International Journal of Intercultural Relations, 16: 413-36. Brown, S. P., W. L. Cron, and T. W. Leigh (1993). Do Feelings of Success Mediate Sales Performance-Work Attitude Relationships? Journal of the Academy of Marketing Science, 21 (2): 91-100. Carmines, E. G., and J. P. McIver (1981). Analyzing Models with Unobserved Variables: Analysis of Covariance Structures. In Social Measurement: Current Issues, edited by G. W. Bohrnstedt and E. F. Borgatta. Beverly Hills, CA: Sage, pp. 65-115. Cohen, E. (1972). Toward a Sociology of International Tourism. Social Research, 39: 164-82. (1979). A Phenomenology of Tourist Experiences. Sociology, 13: 179-201. Cook, S. D. (1995). 1996 Outlook for Travel and Tourism Basics for Building Strategies. In Proceedings of the Travel Industry Association of Americas Twenty-First Annual Outlook Forum, edited by S. D. Cook and B. McClure. Washington, DC: Travel Industry Association of America, pp. 5-18. Dickson, M. A., and M. A. Littrell (1997). Consumers of Clothing from Alternative Trading Organizations: Social Attitudes and Purchase Evaluative Criteria. Clothing and Textiles Research Journal, 15 (1): 20-33. Dillman, D. A. (1978). Mail and Telephone Surveys: The Total Design Method. New York: John Wiley. Engel, J. F., R. D. Blackwell, and P. W. Miniard (1993). Consumer Behavior. 7th ed. Fort Worth, TX: Dryden. Fishbein, M. (1967). A Behavior Theory Approach to the Relations between Beliefs about an Object and the Attitude toward the Object. In Readings in Attitude Theory and Measurement, edited by M. Fishbein. New York: John Wiley, pp. 389-400.

FUTURE DIRECTIONS
On the basis of the findings of this study some suggestions for future research are offered. First, additional research is needed to further refine measures for the research variables. Although the measures possessed acceptable reliability and construct validity, they should be viewed as preliminary measures of the constructs. Second, future studies should test alternative models proposing different relationships and other constructs not addressed by this research, to provide a more complete understanding of tourists travel and souvenir behavior. Third, future studies should test the model with a different sample of international tourists, for example, Korean tourists who are traveling to Mexico. Previous studies (Pizam and Sussmann 1995; Ritter 1987; Woodside and Lawrence 1985) have found differences between travel behavior patterns of East Asian tourists and U.S. tourists.

Downloaded from jtr.sagepub.com at SCD Universit de Savoie on April 28, 2011

162

NOVEMBER 1999
Ross, E. L., and S. E. Iso-Ahola (1990). Sightseeing Tourists Motivation and Satisfaction. Annals of Tourism Research, 17: 226-37. Sampson, D. L., and H. P. Smith (1957). A Scale to Measure WorldMinded Attitudes. Journal of Social Psychology, 45: 99-106. Shim, S., and M. F. Drake (1990). Consumer Intention to Purchase Apparel by Mail Order: Beliefs, Attitude, and Decision Process Variables. Clothing and Textiles Research Journal, 9 (1): 18-26. Slaybaugh, J., M. A. Littrell, and J. Farrell-Beck (1990). Consumers of Hmong Textiles. Clothing and Textiles Research Journal, 8 (2): 56-64. Smith, H. P. (1955). Do Intercultural Experiences Affect Attitudes? Journal of Abnormal and Social Psychology, 51: 469-77. Smith, V. (1989). Introduction. In Hosts and Guests: The Anthropology of Tourism, 2d ed., edited by V. Smith. Philadelphia: University of Pennsylvania Press, pp. 1-17. Timothy, D. J., and R. W. Butler (1995). Cross-Border Shopping: A North American Perspective. Annals of Tourism Research, 22 (1): 16-34. Travel Industry Association of America (1996). 1995 Travel Market Report: Full Year Results of the National Travel Survey. Washington, DC: Travel Industry Association of America. Travelers Notes. (1995). Consumer Reports Travel Letter, October: 234. U.S. Bureau of the Census (1997). Statistical Abstract of the United States. Washington, DC: Government Printing Office. U.S. Congressional House Committee on Small Business (1994). The Effect of International Tourism and International Commerce on Local Economies and Small Businesses. Washington, DC: Government Printing Office. U.S. Department of Commerce (1993). World Tourism at the Millennium: An Agenda for Industry, Government and Education. Washington, DC: Government Printing Office. van den Berghe, P. L. (1995). Marketing Mayas: Ethnic Tourism Promotion in Mexico. Annals of Tourism Research, 22 (3): 568-88. Vellas, F., and L. Bcherel (1995). International Tourism. New York: St. Martins. Wiseman, R. L., M. R. Hammer, and H. Nishida (1989). Predictors of Intercultural Communication Competence. International Journal of Intercultural Relations, 13: 349-70. Woodside, A., and J. Lawrence (1985). Step Two in Benefit Segmentation: Learning the Benefits Realized by Major Travel Markets. Journal of Travel Research, 24 (1): 7-13.

Fishbein, M., and I. Ajzen (1975). Belief, Attitude, Intention, and Behavior. Reading, MA: Addison-Wesley. Good, L. K., T. J. Page, and C. E. Young (1996). Assessing Hierarchical Differences in Job-Related Attitudes and Turnover among Retail Managers. Journal of the Academy of Marketing Science, 24 (2): 148-56. Graburn, N. H. (1989). Tourism: The Sacred Journey. In Hosts and Guests: The Anthropology of Tourism, 2d ed., edited by V. Smith. Philadelphia: University of Pennsylvania Press, pp. 21-36. Gruber, A. (1971). Purchase Intent and Purchase Probability. Journal of Advertising Research, 10: 23-27. Hett, E. J. (1991). The Development of an Instrument to Measure GlobalMindedness. Dissertation Abstracts International, 52 (6): 2099-A. If You Go. (1997). Milwaukee Journal Sentinel, November 23, p. 8. Jansen-Verbeke, M. (1991). Leisure Shopping: A Magic Concept for the Tourism Industry? Tourism Management, 12: 9-14. Jreskog, K. G., and D. Srbom (1989). LISREL VII: A Guide to the Program and Applications. Chicago: SPSS Inc. Kent, W. E., P. J. Shock, and R. E. Snow (1983). Shopping: Tourisms Unsung Hero(in). Journal of Travel Research, 21 (4): 2-4. Keown, C. E. (1989). A Model of Tourists Propensity to Buy: The Case of Japanese Visitors to Hawaii. Journal of Travel Research, 27 (3): 31-34. Latham, J. (1995). International Tourism Flows. Progress in Tourism and Hospitality Research, 1: 63-71. Littrell, M. A. (1990). Symbolic Significance of Textiles Crafts for Tourists. Annals of Tourism Research, 17 (2): 228-45. Littrell, M. A., S. Baizerman, R. Kean, S. Gahring, S. Niemeyer, R. Reilly, and J. Stout (1994). Souvenirs and Tourism Styles. Journal of Travel Research, 33 (1): 3-11. Mader, R. (1998). Tourist Guide of the State of Oaxaca. Business Mexico, 8 (5): 45. Pedhazur, E. J. (1982). Multiple Regression in Behavioral Research. 2d ed. New York: Holt, Rinehart & Winston. Pitts, R. E., and A. G. Woodside (1986). Personal Values and Travel Decisions. Journal of Travel Research, 25: 20-25. Pizam, A., and S. Sussmann (1995). Does Nationality Affect Tourist Behavior? Annals of Tourism Research, 22 (4): 901-17. Ritter, W. (1987). Styles of Tourism in the Modern World. Tourism Recreation Research, 12 (1): 3-8. Rokeach, M. (1969). Beliefs, Attitudes and Values. San Francisco: JosseyBass.

Downloaded from jtr.sagepub.com at SCD Universit de Savoie on April 28, 2011