Đang tải... (xem toàn văn)
This study developed and tested a theoretical framework of event quality, motivation, value, and destination image in order to investigate visitors’ behavioral intentions to revisit an international event, along with the moderating role of attachment avoidance. Results revealed that the relationships among event quality, motivation, value, image, and behavioral intentions were highly significant.
Journal of Travel & Tourism Marketing ISSN: 1054-8408 (Print) 1540-7306 (Online) Journal homepage: http://www.tandfonline.com/loi/wttm20 Factors affecting international event visitors’ behavioral intentions: the moderating role of attachment avoidance Myung Ja Kim, Choong-Ki Lee, James F Petrick & Sabena S Hahn To cite this article: Myung Ja Kim, Choong-Ki Lee, James F Petrick & Sabena S Hahn (2018): Factors affecting international event visitors’ behavioral intentions: the moderating role of attachment avoidance, Journal of Travel & Tourism Marketing, DOI: 10.1080/10548408.2018.1468855 To link to this article: https://doi.org/10.1080/10548408.2018.1468855 Published online: 07 May 2018 Submit your article to this journal Article views: 16 View related articles View Crossmark data Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=wttm20 JOURNAL OF TRAVEL & TOURISM MARKETING, 2018 https://doi.org/10.1080/10548408.2018.1468855 Factors affecting international event visitors’ behavioral intentions: the moderating role of attachment avoidance Myung Ja Kim a a , Choong-Ki Leea, James F Petrickb and Sabena S Hahna The College of Hotel & Tourism Management, Kyung Hee University, 26 Kyungheedae-ro, Dongdaemun-gu, Seoul 02447, Republic of Korea; Faculty of Department of Recreation, Park & Tourism Sciences, Texas A&M University, College Station, TX 77843-2261, USA b ABSTRACT ARTICLE HISTORY This study developed and tested a theoretical framework of event quality, motivation, value, and destination image in order to investigate visitors’ behavioral intentions to revisit an international event, along with the moderating role of attachment avoidance Results revealed that the relationships among event quality, motivation, value, image, and behavioral intentions were highly significant Attachment avoidance was additionally found to moderate the relationships between quality and value, motivation and value, value and destination image, and value and behavioral intentions Findings further provide specific implications for both theoretical insight and marketing practice in the context of annual and international events Received 15 January 2018 Revised 20 March 2018 Accepted 18 April 2018 Introduction International events are important as many generate major economic benefits for host countries and their residents (Han, Nelson, & Kim, 2015; Lee, 2000; Lee, Lee, & Wicks, 2004; Lee & Taylor, 2005) With competition growing for the attraction of visitors to events (Lee, Petrick, & Crompton, 2007), it is becoming ever more important to be able to identify the underlying reasons why people choose to attend Individual and/or group motives typically initiate the decision-making processes related to event visits (Crompton & McKay, 1997) Thus, an understanding of the underlying motivation factors which lead to the decision to attend an international event has been suggested to be crucial for event success (Formica & Uysal, 1998) Studies on event motivation have shown that exploration, socialization, family togetherness, novelty, and escape are key factors that motivate visits to cultural expositions (expos) (Kah, Lee, & Chung, 2010; Lee, 2000; Lee et al., 2004) Researchers have also asserted that quality plays a critical role in the success of both sports events (Jin, Lee, & Lee, 2013; Ko, Zhang, Cattani, & Pastore, 2011; Moon, Ko, Connaughton, & Lee, 2013) and traditional festivals (Yoon, Lee, & Lee, 2010) Further, in the context of worldwide events, Richards and Wilson (2004) revealed that the role of event motivations is significant in creating innovative event value Scholars have CONTACT Choong-Ki Lee, Professor, cklee@khu.ac.kr Dongdaemun-gu, Seoul 02447, Republic of Korea © 2018 Informa UK Limited, trading as Taylor & Francis Group KEYWORDS Event; quality; motivation; value; destination image; behavioral intentions; attachment avoidance; formative indicator; structural equation modeling; Oriental context further shown that event value influences destination image and intentions to revisit cultural events (Richards & Wilson, 2004) and recurring sports events (Kaplanidou & Gibson, 2010) In association with the literature cited above, the relationships between quality, motivation, value, image, and behavioral intention are recognized as important constructs for understanding the decisionmaking processes of event attendees However, little research has examined these relationships for international events, especially, while trying to understand the multiple dimensions of event motivation that could exist for worldwide expos in Asia Another variable that is likely important for understanding the decision-making processes of event attendees is attachment avoidance Attachment avoidance has been defined as being insecure with regard to and/ or having difficulty in having an attachment (Bowlby, 1969) One of the many reasons that has been found to cause attachment avoidance occurs when someone has felt rejected by their earlier caregivers, especially during times of need (Bowlby, 1969, 1977) Individuals who are highly avoidant have been found to have great difficulty in trusting and depending on others and typically become rigidly more self-reliant compared to their counterparts (Mikulincer & Shaver, 2007) Their goal of avoiding dependence often produces a unique style of regulating distress, involving suppressing attachment The College of Hotel & Tourism Management, Kyung Hee University, 26 Kyungheedae-ro, M J KIM ET AL needs and defensively disengaging from their partners (Mikulincer & Shaver, 2003, 2007) In an online context, Sabiote, Frías, and Casteda (2012) asserted that avoidance can have substantial moderating impact on overall perceived value of tourism products and services purchased online This implies that tourists’ behaviors are also affected by attachment avoidance One’s level of attachment avoidance has also been found to influence relationship quality (Mancini, Robinaugh, Shear, & Bonanno, 2009) Research has further found that attachment avoidance is associated with intimacy motivation (Edelstein, Stanton, Henderson, & Sanders, 2010), motivation for sex (Davis, Shaver, & Vernon, 2004), and academic motivation (Gore & Rogers, 2010) In addition, researchers have shown that attachment avoidance is linked to both behavioral intentions and actual behaviors (Fraley & Davis, 1997; Geller & Bamberger, 2009; Jin & Peña, 2010; Lee, 2013) In the event tourism domain, researchers have been considerably interested in event quality, perceived value, destination image, and attendees’ behavioral intention to revisit the event (Jin et al., 2013; Moon et al., 2013) Also, in studies on events visitors’ motivations such as exploration, family togetherness, novelty, escape, and socialization have been shown to lead to a better understanding of potential tourists’ behavior (Kah et al., 2010; Yoon et al., 2010) In addition, attachment avoidance research has identified differences between high and low levels of attachment avoidance as a moderating effect has been extensively studied (Lee, 2013; Mikulincer & Shaver, 2007) In line with the literature review above, research on the factors involved and how to influence visitors’ behavior in terms of quality, multidimensional motivations, value, image, behavioral intention, and attachment avoidance is believed to be timely In particular, even though attachment avoidance has consistently been found to play a moderating role in predicting individuals’ behaviors, it has yet to be utilized to understand the behaviors of event participants Therefore, the primary purposes of this research are to examine a proposed model including quality, motivation, value, image, and behavioral intention, and then to investigate the moderating effects of attachment avoidance Additionally, with an annual event taking place in China every year (Weifang China, 2016) and with annual visitor numbers to the International Expo in China being over 200,000 (China Chamber of International Commerce, 2016) it is believed that the use of this expo is a good case study for examining the study’s purposes Literature review Events Events have been suggested to include “world fairs/ expos, carnivals/festivals, sports events, cultural/religious events, historical milestones, commercial/agricultural events, or political personage events” (Ritchie, 1984, p 2) They have consistently been found to have the potential to increase a destination’s competiveness with regard to attracting visitors (Getz, 2008) Due to this, Richards and Wilson (2004) argued that cities are increasingly hosting events (e.g world fairs, expos, sporting events, and festivals) to enhance their image, encourage urban development, and to entice visitors and encourage investment Similarly, cultural events at historical sites are becoming increasingly important for promoting visits to destinations (Formica & Uysal, 1998) In particular, international events (Lee & Taylor, 2005) and world expos (Lee, 2000; Lee et al., 2004) have been found to be extremely effective at luring international and domestic tourists For the purposes of the current study, an international event refers to an event that attracts a considerable number of international participants and spectators, and generates significant short-term and long-term economic, social, and cultural benefits for the host country, providing international media coverage for tourism (Majorevents.govt.nz, 2016) Thus, this study uses international event visitors (i.e the China– Japan–Korea Industries (CJK) Expo annually held in the same place of Weifang China) as the population for understanding visitors’ revisit intentions to the event Motivation theory Motives have been defined as internal factors that arouse persons’ behaviors, and key tourism motivations have been suggested to include: personal competence, interpersonal diversion, positive interpersonal development, and avoidance of one’s daily routine and problems (Iso-Ahola, 1982) Motivations for pleasure travel have been found to include “socio-psychological motives (escape, exploration, relaxation, regression, prestige, kinship relationships, and socialization) and cultural motives (novelty and education)” (Crompton, 1979, p 408) Push motives are those internal to the tourists, which push them toward desiring to participate in an activity, while pull factors are the features of a destination/ attraction that are created to increase the desirability of a destination (Crompton, 1979) Crompton (1979) further found that push motivations include: the desire JOURNAL OF TRAVEL & TOURISM MARKETING for adventure/social interaction, escape/avoidance, rest/ relaxation, excitement, prestige, health/fitness, and family togetherness Pull factors can include the attractiveness of a destination: cultural attractions, recreation facilities, entertainment, natural scenery, and beaches (Yoon & Uysal, 2005) Motivations to attend an international cultural–historical event have been found to include: group togetherness, attraction/excitement, cultural/historical factors, family togetherness, socialization/entertainment, and novelty (Formica & Uysal, 1998) Similarly, it has been found that motivations to attend international expo events include: family togetherness, exploration, escape (recover equilibrium), novelty, attractions, known-group socialization, and external group socialization (Lee, 2000; Lee et al., 2004) For global expo events in Asia, motivations of visitors have been found to include: family togetherness, exploration, novelty, socialization, and escape (Kah et al., 2010) Based on the above cited literature, the current study operationalizes visitors’ motivations to include exploration, socialization, family togetherness, novelty, and escape Attachment avoidance From a psychological view, attachment has been defined as “the tendency of people to make secure affectional bonds to particular individuals and objects” (Bowlby, 1977, p 201) Attachment avoidance occurs when people deny and or avoid attachment to others or things (Bartholomew & Horowitz, 1991) Those who are attachment avoidant tend to feel subjective distress and discomfort when they become close to others (Bartholomew & Horowitz, 1991) Attachment avoidance has been found to be negatively related to the tendency to form attachment bonds with peers as well as romantic partners, as attachment avoidant individuals tend to think their partners are insecure (Fraley & Davis, 1997) Attachment avoidance has further been found to be negatively associated with emotional and sexual closeness (Davis et al., 2004) Much of the tourism literature has shown that seeking novelty (Lee & Crompton, 1992; Snepenger, 1987), and/or trying to escape one’s normal environment (IsoAhola, 1982; Kozak, 2001) are primary reasons people travel This suggests that new and/or spontaneous behaviors are important elements of touristic behaviors Yet, the attachment avoidance literature has consistently found that persons who are attachment avoidant have difficulty in performing spontaneous or new behaviors (Simpson, Rholes, & Nelligan, 1992) This has been suggested to occur because highly avoidant persons have different needs regarding seeking/giving emotional support, physical contact, and supportive comments (Simpson et al., 1992) Drawing upon the literature review, it is anticipated that attachment avoidance will have a moderating role in predicting mega-event visitors’ behaviors Thus, this study regards attachment avoidance as a potentially important moderator for explaining event consumers’ behaviors Hypotheses development Relationship between quality and value Event quality has been defined as “the quality of opportunity provided by the elements of a festival that are under the control of the promoting organization” (Crompton, 2003, p 306) Experience quality has been found to have significant impacts on the perceived value, satisfaction, and water park image for water park patrons, which in turn directly and indirectly influenced their behavioral intentions (Jin, Lee, & Lee, 2015) In the context of a major league baseball game, event quality has also been suggested to assist event managers in succeeding and maintaining a competitive edge within the event marketplace (Ko et al., 2011) Moreover, research has suggested that event quality is related to the physical environment and pertains to the spectators’ evaluations of physical facilities (Jin et al., 2013) Elements of the physical environment that have been found to be important include: ambience, equipment, facilitation, layout, and function (Bitner, 1992; Westerbeek, 2000) Thus, in association with the literature cited above, this research regards the quality of an event’s physical environment (e.g ambience, facility, design, cleanliness) as an important antecedent of perceived value Perceived value has been argued to be the comparison of what somebody gets for what they give (Zeithaml, 1988) Dimensions of what somebody gets have been found to include quality, reputation, and emotional response, while dimensions of what somebody gives can include both a monetary and a behavioral price (Petrick, 2002) Customer value has been argued to be a source of competitive advantage that directly leads to customer satisfaction, loyalty, and purchase (Parasuraman, 1997) Perceived value has been further suggested to not only influence customer selection behavior pre-purchase, but also satisfaction and intent to recommend and repurchase post-purchase (Lee, Yoon, & Lee, 2007; Parasuraman & Grewal, 2000) In an international sporting event, perceived value has been found to play an important role in understanding the relationships between quality, destination image, and behavioral intentions (Jin et al., 2013) Also, based on an international event study, perceived value M J KIM ET AL has been conceptualized as having hedonic, economic, social, and altruistic items (Moon et al., 2013) Thus, for the current study, event value includes the following factors: hedonic, economic, and overall value in the context of the CJK Expo From a travel and tourism marketing perspective, several studies have shown that quality perceptions are significantly related to perceptions of value (Jin et al., 2013, 2015; Moon et al., 2013; Petrick, 2004b; Yoon et al., 2010) Specifically, quality has been found to have a significant effect on perceptions of value for cruise passengers (Petrick, 2004b), festival attendees (Yoon et al., 2010), water park patrons (Jin et al., 2015), and bicycle racers (Moon et al., 2013) Hence, in line with the literature cited above, the following hypothesis (H) is proposed for an international event H1: Event quality has a positive effect on event value Relationship between motivation and value Crompton and McKay (1997) and Lee, Lee, Bernhard, and Lee (2009) asserted that motivations are closely related to satisfaction Thus, it is likely that festival managers should attempt to understand attendees’ motives They also argued that motivation is useful for segmenting target markets, which can help marketers develop effective promotion strategies Crompton and McKay (1997) specifically identified six motive domains for attending festivals including: cultural exploration, novelty/regression, recover equilibrium, known group socialization, external interaction/socialization, and gregariousness Similarly, in a study of worldwide expo visitors, Lee, Kang, and Lee (2013) identified the following motivations: cultural exploration, family togetherness, event attractions, socialization, novelty, and escape Multiple motivations have been found in various studies (Kim, Kim, & Wachter, 2013; Prebensen, Woo, Chen, & Uysal, 2013; Redondo-Carretero, CamareroIzquierdo, Gutiérrez-Arranz, & Rodríguez-Pinto, 2017; Woratschek, Durchholz, Maier, & Ströbel, 2017) that positively influence perceptions of value For example, hedonic and social motivations have been found to have a significant effect on perceived value to engage in using a smartphone that leads to satisfaction and mobile engagement intention (Kim, Kim, et al., 2013) Also, motives to learn a language have been found to be related to tourists’ perceived value (RedondoCarretero et al., 2017), and nature tourists’ motivations for relaxation and socialization have been found to positively influence their perceived value of destination experience (Prebensen et al., 2013) Further, Woratschek et al (2017) suggested that event visitors’ motivations for socializing and intercultural contact are key factors that can lead to value at sport events In association with the literature cited above, the following hypothesis is posited for an international expo: H2: Event motivation has a positive effect on event value Relationship between value and destination image Destination image has been defined as “the perception of individual attributes or destination features (e.g local cuisine) known as cognitive images and mental pictures or place imagery based on both cognitive (e.g safe for family) and affective images (e.g enjoyable experience)” (Prayag & Ryan, 2012, p 343) The image of an event destination has consistently been found to be a key determinant for attracting attendees (Richards & Wilson, 2004) Marketers, including those promoting events, thus attempt to develop promotions that aid in building an attractive image for a destination to draw potential tourists In addition, destination image has been found to significantly mediate the relationships between risks (i.e perceived sociopsychological and financial risks) and revisit intention in the context of post-disaster Japan tourism (Chew & Jahari, 2014) Destination image has been further found to partially mediate the relationship between service quality and behavioral intention, while destination image has been found to fully mediate the relationship between perceived value and behavioral intentions in an international event (Moon et al., 2013) In association with the literature review above, this study assumes that destination image will significantly mediate the relationship between value and behavioral intention in an international event context Perceived image of a destination has been suggested to be formed through the image projected by individuals’ own needs and values (Gartner, 1994) Moreover, perceptions of value have consistently been found to significantly influence tourists’ image of a destination (Baloglu & McCleary, 1999; Bramwell & Rawding, 1996; Moon et al., 2013; Um & Crompton, 1990) Thus, in association with the literature cited above, we posit the following hypothesis in an international event: H3: Event value has a positive effect on destination image JOURNAL OF TRAVEL & TOURISM MARKETING Relationship between value and behavioral intentions Behavioral intentions have been defined as an assertion of the likelihood of initiating a certain action (Oliver, 2010) Zeithaml, Berry, and Parasuraman (1996, p 34) referred to behavioral intentions by stating, “when customers praise the firm, express preference for the company over others, increase the volume of their purchases, or agreeably pay a price premium, they are indicating behaviorally that they are bonding with the company” Because actual behavior is not easy to measure, behavioral intentions have often been used as a proxy variable for predicting actual consumer behavior Two of the most common behavioral intentions examined in the tourism and event literatures are likely intention to visit/revisit an event and intention to spread positive word-of-mouth (Chen & Petrick, 2015; Li & Petrick, 2010; Petrick, 2004a) In event contexts, behavioral intentions can consist of coming back to the destination in the future, recommending the destination to friends or others, and telling other people positive things about this destination (Jin et al., 2013) Therefore, behavioral intentions in this study are operationalized as intentions to revisit, recommend the event to friends or others, and say positive things about the event The relationship between perceived value and loyalty has been found to be positive and significant for tropical island visitors (Song, Su, & Li, 2013) Among cruise passengers, perceived value has been found to be the most important indicator of future purchase intentions for repeat visitors (Petrick, 2004a) Perceived value has further been found to be related to the behavioral intentions of water park visitors (Jin et al., 2015), event attendees (Jin et al., 2013), and festival attendees (Lee, Petrick, et al., 2007) Thus, the following hypothesis is proposed in the context of an event: H4: Event value has a positive effect on behavioral intentions to revisit the event Relationship between image and behavioral intentions The tourism and event literature has also consistently found a significant relationship between the image that potential visitors have of a destination and their behavioral intentions to attend future events (Kaplanidou & Gibson, 2010) These include studies of sporting events (Kaplanidou & Gibson, 2010; Moon et al., 2013), visitors to a tropical international island resort (Song et al., 2013), city tour visitors (Kim, Holland, & Han, 2013), water park visitors (Jin et al., 2015), and Malaysian tourists to Japan (Chew & Jahari, 2014) Thus, in association with the literature cited above, this study posits the following hypothesis: H5: Destination image of an event has a positive effect on behavioral intentions to revisit the event Moderating role of attachment avoidance Even though attachment avoidance has consistently been found to be related to and/or moderate loyaltytype behaviors, it has received scant attention in the tourism literature Specifically, Mikulincer and Shaver (2007) found that highly avoidant persons tend to use more deactivating schemes to cope with relationship problem than individuals with low attachment avoidance Additionally, Jin and Peña (2010) found that in comparison with lower attachment avoidant persons, higher attachment avoidant persons were less likely to call or text others It has further been found that interpersonal tasks (i.e helping and caregiving) are different depending on individuals’ levels of attachment avoidance (Geller & Bamberger, 2009) Lower attachment avoidant persons have further been found to have more negative views of others (Bartholomew, 1990) and to fear intimacy (Bartholomew, 1990; Edelstein et al., 2010) Attachment avoidance has also been found to moderate coping with loss and marital adjustment (Mancini et al., 2009), the perceptions of value and quality of online tourism services (Sabiote et al., 2012), the relationship between body image and sexual activity (La Rocque & Cioe, 2011), academic motivations (Gore & Rogers, 2010), and social media behaviors (Lee, 2013) Moreover, individuals having high attachment avoidance have been found to be more likely to be distant from other people but to not feel as lonely as compared with their counterparts (Hazan & Shaver, 1987) Since the literature review above has consistently found attachment avoidance to moderate the relationships between antecedents and behaviors, this study postulates the following: H6: Attachment avoidance has a significant moderating effect on quality, motivation, value, image, and behavioral intentions to revisit the event Together with the study hypotheses, the proposed theoretical model is presented in Figure This model displays the proposed relationships between quality, motivation, value, image, and behavioral intentions realting to the event as well as the postulated moderating role of attachment avoidance 6 M J KIM ET AL The moderating role of attachment avoidance (H6) Destination image Quality H1 H3 H2 Value Motivation H5 H4 Behavioral intention First-order factor Exploration Family togetherness Novelty Escape Socialization Second -order factor Figure Proposed research model H: hypothesis Methods Study site As an annual global event, the second CJK Expo was held at the Shandong-Taiwan Convention and Exhibition Center in Weifang, China, from September 23–25, 2016 Located in the middle of the Shandong Peninsula and having a population of over nine million residents, Weifang is well-known for its international Kite Festival and for being a leading tourism destination in China (Weifang China, 2016) The city hosted the CJK Expo in 2016, which included 2000 booths, exhibitors from 14 countries including China, Japan, and Korea, and over 200,000 visitors from 41 countries (China Chamber of International Commerce, 2016) Exhibitors from foreign companies accounted for 55% of the total 1000 companies at the expo (China Daily, 2017) Measurements Each construct in this study utilized multi-item measures in order to avoid measurement errors related to the use of single-item measures (Churchill, 1979) Additionally, to help ensure the validity and reliability of measures, only previously validated measures were used after adapting them to the context of this study Ten constructs were included in this study as shown in Table Quality was evaluated with four items adapted from Crompton (2003), Ko et al (2011), Petrick (2004b), and Yoon et al (2010) Motivation was measured with 20 items adapted from previous research (Crompton, 1979; Crompton & McKay, 1997; Formica & Uysal, 1998; Kah et al., 2010; Lee, 2000; Lee et al., 2004; Yoon & Uysal, 2005): with four items each used to measure exploration, novelty, family togetherness, escape, and socialization Value was assessed with four items drawn from previous studies (Jin et al., 2015; Parasuraman, 1997; Petrick, 2002; Zeithaml et al., 1996) Image was assessed using four items derived from Kaplanidou and Gibson (2010), Moon et al (2013), Prayag and Ryan (2012), and Richards and Wilson (2004) To assess behavioral intentions, four questions were generated from Jin et al (2013), Lee, Petrick, et al (2007), Oliver (2010), and Zeithaml et al (1996) Finally, to measure attachment avoidance, four questions were derived from previous studies (Bartholomew & Horowitz, 1991; Davis et al., 2004; Hazan & Shaver, 1987; Lee, 2013; Mikulincer & Shaver, 2007; Simpson et al., 1992) All 40 items were assessed using a 5-point Likert-type scale ranging from strongly disagree (1) to strongly agree (5) In addition, some sociodemographics were included in the questionnaire The questionnaire was initially developed in English and then translated into Chinese by two professionals who are proficient in both languages To aid in establishing content validity, three academic experts familiar with the topic under study assessed the measurement items In addition, three field experts in events were asked to evaluate whether the measurement items were appropriately assessed in the context of the expo As a result, one item of behavioral intention was deleted as it was suggested to have a meaning overlapping with that of another item (i.e “I intend to come back to the expo in the future”), and one item of exploration was added to the questionnaire (i.e “To know about the expo”) Three general questions related to expos (i.e information source, the purpose of visiting the expo, and field of interest) were also added JOURNAL OF TRAVEL & TOURISM MARKETING Table Measurement sources and contexts Construct Quality (four items) Sources Crompton (2003) Ko et al (2011) Petrick (2004b) Yoon et al (2010) Motivation (exploration (four Crompton (1979) items); family togetherness (four items); Crompton & McKay (1997) novelty (four items); escape (four items); Formica & Uysal (1998) socialization (four items)) Kah et al (2010) Lee (2000) Lee et al (2004) Value (four items) Image (four items) Behavioral intention (four items) Attachment avoidance (four items) Yoon & Uysal, 2005 Jin et al (2015) Parasuraman (1997) Petrick (2002) Zeithaml (1988) Kaplanidou & Gibson (2010) Moon et al (2013) Prayag & Ryan (2012) Richards & Wilson (2004) Jin et al (2013) Lee, Petrick, et al (2007) Oliver (2010) Zeithaml et al (1996) Bartholomew & Horowitz (1991) Davis et al (2004) Hazan and Shaver (1987) Lee (2013) Mikulincer & Shaver (2007) Simpson et al (1992) Contexts/Delineated factors Event; ambience; source of information; comfort amenities; vendor interaction; parking Event; game; augment service; interaction; outcome; environment Cruise passenger; outstanding; reliable; dependable; consistent Festival; program; souvenirs; facilities; food Vacation; escape; exploration; relaxation; prestige; regression; kinship relationships; social interaction Event; cultural exploration; novelty/regression; recover equilibrium; known group socialization; external interaction/socialization; gregariousness Event; socialization/entertainment; attraction; excitement; group/family togetherness; cultural/historical; novelty Exposition; exploration; family togetherness; novelty; escape; socialization Exposition; exploration; family togetherness; escape, novelty; external/internal group socialization; attractions Exposition; exploration; family togetherness; novelty; escape; event attractions; socialization Tourist motivation; eight push factors; 10 pull factors Water park; reasonable price; good reputation; quality of service Customer value; utility; worth; benefits; quality Behavioral price; monetary price; emotional response; quality; reputation Low price; whatever I want in a product; quality I get for the price; what I get for what I give Event; overall image of the destination as a vacation destination Event; ease of communication; hospitality; friendliness; receptiveness; night-time and entertainment Attractions; diversity; accommodation; accessibility; exoticism Event; architecture; multicultural; dynamic; shopping; cozy; safe Event; come back to this destination; recommend; tell other people positive things Festival; say positive things; re-attend the festival; recommend to others; encourage friends and relatives to go the festival; the first choice among festivals Behavioral perspective; loyalty; long-term effects of satisfaction Loyalty; switch; pay more; and external/internal response A downplaying of the importance of close relationships; restricted emotionality; an emphasis on independence and self-reliance; a lack of clarity or credibility in discussing relationships Attachment-related avoidance was measured by representative subsets that comprise the experiences in close relationships Uncomfortable being close to others; difficult to trust others completely; difficult to allow myself to depend on others; nervous when anyone gets too close Uncomfortable having to depend on others; don’t like people getting too close to me; uncomfortable being too close to others; difficult to trust others; nervous whenever anyone gets too close to me Attachment; avoidance (structure, dynamics, and change); avoidance dimensions I find it relatively easy to get close to others; I’m comfortable having others depend on me; others often want me to be more intimate than I feel comfortable being A pilot test was conducted with 20 Chinese students who had visited international expos in the past This process resulted in the revision of several items on motivation to ensure clarity of meaning For example, escape motivation items were rewritten (i.e “I want a change from my ordinary life” and “To relieve daily stress” from “To avoid my ordinary life” and “To avoid my stress”) Also, some items of quality were found to be ambiguous and thus, were reworded to help ensure clarity For instance, “The ambience of the expo is what I’m looking for” and “I am impressed with the design of the facility” were rewritten from “The ambience of the expo is a spectator-setting” and “The design of the facility is impressive.” Data collection Due to the difficulty of conducting random sampling on site at the expo, convenience sampling of persons attending the CJK Expo 2016 in Weifang, China, between September 23 and September 25, 2016 was used Visitors were intercepted at a booth in the expo by four trained field researchers Before starting the survey, the field researchers were trained specifically on how to complete a questionnaire When intercepting potential respondents, a screening question was asked as to whether or not individuals were visiting the host city from out of town Individuals indicating “yes” were then asked to identify the primary purpose of their visit to Weifang, China The field researchers described the purpose of the research, guaranteed anonymity of all qualified responses, and distributed questionnaires only to respondents who agreed to take part in the survey In order to attempt to increase the response rate, a pack of cosmetic facial masks was given to attendees who finished the questionnaire as a token of gratitude Approximately 750 persons were approached to take M J KIM ET AL the survey, of which 450 passed the screening question, and agreed to participate Of these, 444 finished the survey Due to incomplete responses, 19 samples were removed from the data set during the data refinement stage (giving an estimated response rate of 56.2%) After an additional 19 responses were deleted due to outliers (Hair, Black, Babin, Anderson, & Tatham, 2006), a total of 406 questionnaires were coded for the analysis Data analysis Partial least squares–structural equation modeling (PLSSEM) was applied to analyze the data using a component-based approach This approach has been extensively utilized for theory confirmation and testing (Chin, 1998), and has been suggested to be more appropriate for complicated models or multi-group analysis than traditional SEMs based on covariance (Hair, Sarstedt, Ringle, & Mena, 2012) Further, PLS-SEM has been recommended when formative as second-order factor constructs are included (Chin, Marcolin, & Newsted, 2003) In this study, the proposed model utilized a formative construct of motivation, along with reflective constructs Additionally, a multi-group approach was needed in order to be able to assess differences in attachment between high- and low-avoidance groups Moreover, since the resultant data did not meet multivariate normality, the non-normality technique of bootstrapping of PLS was employed (Stevens, 2009) Thus, SmartPLS 3.2.6 was applied to analyze the measurement and structural models (Ringle, Wende, & Becker, 2015) Since respondents were asked to rate all survey questions at once, common method variance was a potential issue Thus, precautions were taken using several procedural remedies to address common method bias (Podsakoff, MacKenzie, Lee, & Podsakoff, 2003) First, the introduction section included a description of the study’s purpose, followed by a statement assuring all respondents of their anonymity Second, to decrease respondent apprehension, survey instructions noted that there were no right or wrong answers to the questions Third, the definition of each construct was clearly explained at the beginning of the survey to help ensure response validity Fourth, the questionnaire consisted of three parts: the first part included general information, the second included measurement items related to the research model, and the third included personal questions about demographic characteristics Harman’s single-factor test was performed to confirm if common method variance was present in the resultant data set (Podsakoff et al., 2003) That is, all self-reported survey items were entered into an exploratory factor analysis (EFA) Using this process, if a single factor emerges or one factor accounts for more than 50% of the variance in the variables, common method variance is deemed present (Harman, 1967) The EFA results showed that seven factors were delineated (eigenvalues >1), with the first factor accounting for 20.81% of the variance Subsequent factors explained 14.66%, 8.52%, 8.49%, 8.01% 9.91%, and 7.01% of the variance respectively Since the singlefactor test has been found to have some limitations (Podsakoff & Organ, 1986), a marker-variable approach was also employed For this procedure, a PLS algorithm was applied A marker variable (negative emotional response) was used to estimate the correlations on every theoretical construct in the PLS path model The corrections between the marker variable and all the constructs in the PLS path model were small and insignificant, including quality (−0.04), exploration (−0.03), family togetherness (0.15), novelty (0.01), escape (0.31), socialization (0.06), value (−0.02), destination image (−0.01), behavioral intentions (−0.03), and attachment avoidance (0.53) The resultant average of the squared multiple corrections was 0.04 for the 10 theoretical constructs Accordingly, both the traditional single-factor test and the marker-variable approach suggested that common method bias was not an issue in the study (Podsakoff et al., 2003) Multiple past studies have employed and suggested the use of second-order (formative constructs) factor analysis when research models are complicated (e.g Bock, Zmud, Kim, Lee, & Kong, 2005; Kim, Lee, & Bonn, 2016; Kim, Lee, & Chung, 2013) Formative measurement theory is based on the assumption that the measured variables cause the construct, where each indicator is a cause of the construct (Hair et al., 2006) This formative approach, with the use of PLS-SEM, has been argued to fit well when identifying multiple attributes in order to predict visitor behavior, including specific dimensions in the event tourism field (Ahrholdt, Gudergan, & Ringle, 2017) Therefore, in this study, motivation was measured as a formative construct of a composite factor as well as a second-order construct, including the five subdimensions of exploration, family togetherness, novelty, escape, and socialization among event attendees using PLS-SEM Results Profile of the respondents As shown in Table 2, the majority (66.2%) of respondents were female, and 37.0% of respondents were between 20 and 29 years old Further, respondents JOURNAL OF TRAVEL & TOURISM MARKETING Table Demographic and general characteristics of respondents Characteristics Gender Male Female Missing value Age (years) Less than 20 20–29 30–39 40–49 50–59 60 and over Missing value Educational level High school or less 2-year college University Graduate school or higher Missing value Marital status Single Married Missing value Monthly household income (CNY) Less than 2,999 * 3,000 – 4,999 5,000 – 5,999 6,000 – 6,999 7,000 – 7,999 8,000 – 8,999 9,000 – 9,999 10,000 and above Missing value N (406) % (100) 114 269 23 28.1 66.2 5.7 36 150 111 60 33 14 8.9 37.0 27.3 14.8 8.1 0.5 3.4 68 103 199 20 16 16.7 25.4 49.1 4.9 3.9 165 225 16 40.6 55.5 3.9 84 128 65 31 11 24 59 20.0 31.6 16.0 7.6 2.7 1.0 0.7 5.9 14.5 Characteristics Occupation Professional/Engineer Business owner/Self-employed Service worker Office worker Civil servant Home maker Student Retiree/Unemployed Other Missing value Information source** Through acquaintance TV/Newspapers/Radio Internet/Websites/Social network sites Public relation materials Other Missing value Purpose of visiting exposition** Curiosity/Fun Business trip Academic Trends of industries State of the art/Cutting edge New ideas for businesses Other Interesting fields** Smart manufacturing/Electronics New energy/Environment protection Modern agriculture/Food safety Healthcare and beauty Cultural–creative industries N (406) % (100) 40 44 21 102 38 23 81 11 29 17 9.9 10.8 5.2 25.0 9.4 5.7 20.0 2.7 7.1 4.2 147 58 84 55 26 36.2 14.3 20.7 13.5 6.4 0.3 161 30 47 41 120 37 34 39.7 7.4 11.6 10.1 29.6 9.1 8.4 52 77 80 242 126 12.8 19.0 19.7 59.6 31.0 *US $1 = 6.756 CNY; **multiple response who had attended a university and earned university degrees accounted for half of respondents (49.1%), and almost one third (31.6%) reported a monthly household income between CNY 3000 and CNY 4000 (US$ is equivalent to CNY 6756) Additionally, the majority of respondents were married (55.5%), and one fourth (25%) were office workers More than one third got information about the CJK Expo from acquaintances (36.5%) and had participated in the CJK Expo out of curiosity and for fun (39.7%) Furthermore, the majority of respondents were Chinese (93.2%), followed by minorities (e.g Tibet, Manchu, Hui) (3.4%) and unknown (3.4%) Group check In order to test H6, the sample (n = 406) was separated into high and low attachment avoidance groups The attachment avoidance construct (Cronbach’s α = 0.923) was assessed using four questions which can be found in Table When grouping cases have similar traits and the sample size represents 200 or more responses, the K-means cluster method has been suggested to be appropriate as K-means algorithms allow researchers to specify the number of clusters (Hair et al., 2006) The K-means clustering resulted in partitioning the 406 observations into two clusters in which each observation belonged to the cluster with the nearest mean The resultant groups were named high (n = 185; mean = 4.2) and low (n = 221; mean = 2.2) attachment avoidance Measurement model Confirmatory factor analysis (CFA) was conducted on the measurement model (Kline, 2011) One item of exploration was found to share residual variance with other items for being dropped (i.e the factor loading was below 0.5) In addition, one item for of exploration, family togetherness, novelty, escape, and socialization was dropped because of the low average variance extracted (AVE) of motivation That is, the AVE of motivation prior to dropping the items was 0.489 and then improved up to 0.503 The remaining 34 items were used for the analysis (see Table 3) Assessments in terms of reliability, convergent validity, and discriminant validity were subsequently performed (Hair et al., 2006; Stevens, 2009) As illustrated in Table 4, the Cronbach’s α and composite reliability of each construct was larger than 0.70, 10 M J KIM ET AL Table Results of confirmatory factor analysis Construct Quality 3 3 3 4 3 Exploration motivation Family togetherness motivation Novelty motivation Escape motivation Socialization motivation Value Destination image Behavioral intention Attachment avoidance Items The ambience of the expo is what I’m looking for The facility is clean and well-maintained I am impressed with the design of the facility The facility is safe To experience local customs and cultures To enjoy events and exhibition To increase my knowledge of the expo To spend time with my family together I thought the entire family would enjoy it To help my family learn more about the expo I am curious I seek novelty I seek adventure To escape from routine life To relieve boredom To relieve daily stress To be with people who are enjoying themselves To be with people who enjoy the same things I enjoy To see the event with a group together The time I spent at this event makes me feel good The quality of the event is outstanding The fees for this event are fair Overall, the CJK Expo is worth it The people in Weifang are friendly and interesting Weifang offers suitable accommodations I am not concerned about personal safety in Weifang The structure of the expo is well-built I would like to come back to the expo in the future I would recommend the expo to my friends or others I want to tell other people positive things about the expo I’m not comfortable when I have to depend on other people I don’t like people who get close to me I’m very uncomfortable when I am close to others I find it difficult to trust others completely Table Reliability and discriminant validity Correlation of the constructs Construct (1) (2) (1) Quality 0.913 (2) Motivation 0.633 0.709 (3) Value 0.689 0.658 (4) Destination image 0.768 0.633 (5) Behavioral intention 0.759 0.571 Cronbach’s alpha (α) 0.933 0.928 Composite reliability 0.952 0.937 Average variance extracted (AVE) 0.834 0.503 Mean 4.410 3.941 Standard deviation 0.691 0.731 (3) (4) (5) 0.906 0.648 0.609 0.927 0.948 0.821 4.214 0.777 0.890 0.798 0.912 0.938 0.792 4.323 0.721 0.925 0.915 0.946 0.855 4.388 0.743 All bold-faced diagonal elements appearing in the correlation of the constructs matrix indicate the square roots of AVEs All correlations are significant at the 0.01 level (two-tailed) indicating internal consistency and validating the reliability (Campbell & Fiske, 1959) Furthermore, convergent validity was confirmed as each item's factor loading and each construct’s AVE were larger than 0.5 (Hair et al., 2006) Discriminant validity was further confirmed because the square root of the AVE of each construct was larger than each correlation coefficient (Fornell & Larcker, 1981) The use of a formative approach has been argued to allow for the identification of multiple attributes, including specific dimensions (Ahrholdt et al., 2017) In the current study motivation was measured as a Factor loading 0.912 0.918 0.919 0.903 0.937 0.934 0.913 0.917 0.931 0.916 0.888 0.902 0.883 0.917 0.947 0.891 0.931 0.937 0.871 0.910 0.922 0.896 0.896 0.911 0.899 0.850 0.899 0.889 0.944 0.940 0.828 0.939 0.936 0.898 formative variable comprised of five dimensions Hence, five subconstructs of motivation (i.e exploration, family togetherness, novelty, escape, and socialization), each having three indicators, were used for the formative construct as a second-order factor The weights of the five submotivations, as explained by beta (β) coefficients derived from a standard regression, were used to verify the validity of their respective formative indicators (Kuan & Bock, 2007) The weights as well as t-statistics of the formative constructs are presented in Figure It was found that each indicator weight for the formative construct had significant t-statistics and was hence deemed valid Structural model Figure displays the results of the PLS-SEM, which evaluated the theoretical research model Each of the endogenous variables had ample variance explained (R2): value (55.7%), image (42.0%), and behavioral intentions (65.1%) The higher the value of R2, the better the prediction of the dependent variable (Hair et al., 2006) The path estimates as well as t-statistics were estimated for the hypotheses using a bootstrapping technique since the data were not found to have multivariate JOURNAL OF TRAVEL & TOURISM MARKETING Destination image R2=0.420 0.454*** (8.547) Quality 0.694*** (16.333) 0.648*** (15.592) 0.371*** (8.098) Motivation 0.293*** (28.912) 0.267*** (25.640) Exploration Family togetherness Value R2=0.557 Novelty 0.159*** (3.450) Behavioral intention R2=0.651 0.278*** (30.002) 0.169*** (12.328) 0.264*** (28.018) 11 Escape Socialization ***p < 0.001 Figure Results of path analysis p < 0.001 normality (Stevens, 2009) Results indicated that the relationships between quality and value (ɣ = 0.454, t-value = 8.547, p < 0.001), motivation and value (ɣ = 0.371, t-value = 8.098, p < 0.001), value and image (β = 0.648, t-value = 15.592, p < 0.001), value and behavioral intentions (β = 0.159, t-value = 3.450, p < 0.001), and image and behavioral intentions (β = 0.694, t-value = 16.333, p < 0.001) were all significant Thus, H1, H2, H3, H4, and H5, as presented in Figure 2, were supported In order to test the moderating role of attachment avoidance, H6a, b, c, d and e were assessed (Table 5) From the associated regression results, comparison of the explained variance (R2) revealed differences between two groups (Hair et al., 2006) It was found that more variance was explained for both value (13.4% more) and image (5.9% more) in the low attachment avoidance group in comparison with the high attachment avoidance group, while slightly more variance was explained for behavioral intentions (1.8%) in the high avoidance group than in its counterparts PLS-SEM was used to perform multi-group analysis in order to compare the five relationships across the high and low attachment avoidance groups As revealed in Table 5, with the exception of the association between image and behavioral intention (H6e), the coefficients of the other four paths showed significant differences between the two groups These findings supported H6a, H6b, H6c, and H6d The difference in magnitude of the coefficients between quality and value (ɣ = 0.505 > ɣ = 0.371) and value and image (β = 0.665 > β = 0.620) was significantly larger in the high avoidance respondents than in their counterparts Conversely, the difference in magnitude of the coefficients between motivation and value (ɣ = 0.317 < ɣ = 0.390) and value and behavioral intention (β = 0.163 < β = 0.182) was larger in the low avoidance group than in its counterpart Mediating effects An additional analysis was performed to investigate whether value and image mediate within the research framework PLS bootstrapping of the 500 re-samples was applied to examine the relationships between quality and image, quality and behavioral intention, motivation and image, motivation and behavioral intention, and value and behavioral intention As shown in Table 6, quality had significant positive indirect effects on image (ɣ = 0.294, t-value = 6.267, p < 0.001) and motivation had a significant and positive indirect effect on image (ɣ = 0.240, t-value = 6.920, p < 0.001) through the mediating role of value Also, quality was found to have a significant positive indirect effect on behavioral intentions (ɣ = 0.277, t-value = 5.862, p < 0.001) and motivation had a significant and positive indirect influence on behavioral intentions (ɣ = 0.226, t-value = 7.023, p < 0.001) through the meditating roles of value and image In particular, value had a highly and significantly positive indirect effect on behavioral intentions (β = 0.450, t-value = 10.419, p < 0.001) as image was found to partially mediate the relationship between value and behavioral intentions Thus, value and image played significant mediating roles in the research model Conclusion and implications The current research examined the theoretical foundations for understanding the determinants of event consumers’ behavioral intentions with inclusion of the 12 M J KIM ET AL Table Comparison of path coefficients between low and high attachment avoidance groups Low group (B) t-value (A–B) p-value (A–B) Test of hypothesis Quality → Value 0.505*** 0.371*** Motivation → Value 0.317*** 0.390*** Value → Destination image 0.665*** 0.620*** Value → Behavioral intention 0.163* 0.182** Image → Behavioral intention 0.686*** 0.690*** Exploration → Motivation 0.318*** 0.274*** Family togetherness → Motivation 0.292*** 0.251*** Novelty → Motivation 0.286*** 0.249*** Escape → Motivation 0.098*** 0.246*** Socialization → Motivation 0.286*** 0.275*** R2: Coefficient of determination (variance explained) The low group: value (57.3%), destination image (44.3%), and behavioral intention (64.6%) The high group: value (43.9%), destination image (38.4%), and behavioral intention (66.4%) 17.861 −11.054 8.529 −2.831 −0.668 28.540 24.276 25.658 −62.652 6.921