Perceptions of Risk and Travel Intentions: The Case of China and the Beijing Olympic Games

Transcription

1 Journal of Sport & Tourism ISSN: (Print) (Online) Journal homepage: Perceptions of Risk and Travel Intentions: The Case of China and the Beijing Olympic Games Christine Xueqing Qi, Heather J. Gibson & James J. Zhang To cite this article: Christine Xueqing Qi, Heather J. Gibson & James J. Zhang (2009) Perceptions of Risk and Travel Intentions: The Case of China and the Beijing Olympic Games, Journal of Sport & Tourism, 14:1, 43-67, DOI: / To link to this article: Published online: 27 Apr Submit your article to this journal Article views: 4154 View related articles Citing articles: 53 View citing articles Full Terms & Conditions of access and use can be found at

2 Journal of Sport & Tourism Vol. 14, No. 1, February 2009, pp Perceptions of Risk and Travel Intentions: The Case of China and the Beijing Olympic Games Christine Xueqing Qi, Heather J. Gibson & James J. Zhang The issue of risk associated with mega-event tourism has been receiving increased attention. The purpose of this study is to examine the relationship between risk perceptions and travel intentions associated with an Olympic Games host country, using the Beijing Games as a case study. To control for the influence of individual backgrounds, this study focused on a homogeneous population, students below 30 years of age, US born and raised, and enrolled at a university in the US. Research participants (N ¼ 350) responded to a questionnaire designed to measure perceived risks, travel intentions, international travel experience, tourist role, and demographics. A factor analysis revealed four perceived risk factors: Personal Safety, Cultural Risk, Socio-psychological Risk, and Violence Risk. Utilizing the factor scores, regression analyses revealed that Violence Risk and Socio-psychological Risk had significantly (p, 0.05) negative impacts on participants intention to visit China as a tourist. Although not statistically significant, the four risk factors were negatively predictive of the intention to attend the Olympic Games in China. The perceived risk factors were further found to be related to gender and tourist role types. Discussions are focused on the implications of these research findings to mega event management and promotions of host countries as tourism destinations. Keywords: Olympic Tourism; Perceived Risk; Travel Intentions; Event Management; Destination Promotion; China Christine Xueqing Qi is at the School of Professional and Continuing Education, University of Hong Kong. Heather J. Gibson and James J. Zhang are in the Department of Tourism, Recreation and Sport Management, University of Florida, USA. Correspondence to: Heather Gibson, Department of Tourism, Recreation and Sport Management, 304 Florida Gym, PO Box , University of Florida, Gainesville, FL , USA. hgibson@hhp.ufl.edu ISSN (print)/issn (online) # 2009 Taylor & Francis DOI: /

3 44 C.X. Qi et al. Introduction Successful tourism development is related to the reduction of risks associated with a destination and an event in addition to infrastructural improvements and heightened awareness on the world stage. The threat of terrorism has often been identified as a major risk associated with mega events (Toohey et al., 2003; Kim & Chalip, 2004; Taylor & Toohey, 2006), and the Olympic Games in particular are regarded as an attractive target (Cashman & Hughes, 1999; Neirotti & Hilliard, 2006). Over the last few years, risk has become an important factor when considering international travel (Sönmez, 1998; Sönmez & Graefe, 1998a, b; Lepp & Gibson, 2003, 2008; Reisinger & Mavondo, 2005, 2006; Kozak et al., 2007), especially for Americans who perceive themselves as particularly vulnerable. Tourists tend to avoid destinations they perceive as risky and choose ones they consider safe (Sönmez & Graefe, 1998a). Sönmez & Graefe (1998b) indicated that despite the tourism industry s worldwide economic strength, terrorism and political turmoil present major challenges to the industry. Potential tourists may change their travel plans because of terrorism, which in turn may lead to significant losses for a destination (Coshall, 2003). For example, because of the US Libya military confrontation in 1985, nearly two million Americans changed their foreign travel plans in 1986, which resulted in a 30% decrease in visitation compared with the previous year (Richter & Waugh, 1986; Sönmez & Graefe, 1998b; Edgell, 1990). The most significant disruption to tourist flows to date were the terrorist attacks on September 11, 2001 in the United States which resulted in 6.8% fewer international tourists visiting North America in 2001, compared with 2000 (Office of Travel and Tourism Industries, 2000, 2001, 2002). While, in recent years, risk has been largely associated with the threat of terrorism, travel-related risk is associated with a number of factors including political instability, health threats, crime, violence, war, natural disasters, and terrorism in or near a destination (Coshall, 2003; Lepp & Gibson, 2003; Kozak et al., 2007). Political instability, although not as blatant as terrorism, is a formidable barrier to international tourism (Sönmez & Graefe, 1998b), and it can also increase the perception of risk associated with a destination (Seddighi et al., 2001; Lepp & Gibson, 2003). For example, as a result of the incident in Tiananmen Square in China, hotel occupancy rates in Beijing dropped below 30%, and approximately 11,500 tourists canceled visits to Beijing in 1989 (Gartner & Shen, 1992). The total tourism earnings for China declined by $430 million in 1989 (Gartner & Shen, 1992; Hall & O sullivan, 1996; Sönmez, 1998). Subsequently, China made a great effort to attract international tourists in the following years. Although a 55% increase in foreign visitation was recorded in 1991 and 48% in 1992, memories of this event still remain (Hall & O Sullivan, 1996). Therefore, from a country s perspective, preventing political instability is necessary to develop a strong tourism industry (Teye, 1988). The reform-induced economic growth in China over the last two decades has led to the rapid development of the sport (Theodoraki, 2004) and tourism sectors (Witt & Turner, 2002). Indeed, the World Tourism Organization (UNWTO) predicts that China will become the number one tourist destination in the world by 2020 with

4 Journal of Sport & Tourism million international tourists arriving each year and an annual growth rate of 7.8% (WTO, 2000). As a part of its tourism development strategy, China has utilized sport tourism to attract visitors. It was anticipated that hosting the 2008 Beijing Olympic Games would boost China s tourism and sport tourism industry considerably. Beijing as the host city of the Games underwent a major renovation program that included restoring historical sites and architectural treasures such as the Hutongs and the Forbidden City. Indeed, Koldowski & Yoo (2006) identified the 2008 Olympic Games as one of the seven primary forces predicted to positively affect the Asia Pacific tourism industry over the next decade. Despite these positive predictions, in spring 2008 there were already indications of decreasing numbers of international visitors in Beijing and Shanghai (Barboza, 2008), experiences similar to those reported by many former Olympic host cities in the months leading up to the opening ceremonies (Steinmetz, 2008). China and Beijing were already receiving negative global coverage in the press about their relations with Tibet and the protests present along the torch relay route as well as fears over the air quality in Beijing (Streets et al., 2007). While it is not known the extent to which these news stories directly contributed to the decline in tourism at this time, media coverage can be linked to both positive and negative perceptions, including the image that an Olympic host city will be overcrowded (Faulkner et al., 1998) or full of potential risks (Neirotti & Hilliard, 2006). People cognitively appraise the severity and likelihood of being exposed to risk and evaluate their ability to cope with it. When their protection motivation is aroused, individuals change their behavioral intentions and attitudes (Rogers, 1975, 1983). Researchers have found that tourists perceptions of risk is also related to other factors such as previous international travel experience, life stage, gender, nationality, education, social class, and personality traits specifically related to the propensity for sensation seeking (Tremblay, 1989; Roehl & Fesenmaier, 1992; Sönmez & Graefe, 1998a, 1998b; Gibson & Yiannakis, 2002; Pizam et al., 2002, 2004; Lepp & Gibson, 2003, 2008; Reisinger & Mavondo, 2005, 2006). In reality, not all tourists perceive risk equally, in fact, some are attracted to risky situations while others are repelled (Lepp & Gibson, 2003, 2008; Pizam et al., 2002, 2004). Following the lead in the literature, particularly Lepp & Gibson s (2003) study design, life stage, nationality, and socio-economic status were held constant in this study, while gender, previous international travel experience, and tourist role preference were identified as independent variables in understanding risk perceptions and travel intentions associated with the Olympic Games, using China and the Beijing Olympic Games as a case study. International Olympic Committee (IOC) rules dictate that the promotion for the subsequent Olympic Games may not begin until the close of the current Olympic Games. Thus, data for this study were collected prior to the close of the Athens Summer Olympic Games. In so doing, this ensured that research participants had not been exposed to any media promotion for the Beijing Games, any of the debates over human rights (Gupta, 2006), or fears over the air quality in Beijing (Streets et al., 2007). The purpose of this study was to investigate the perceived risk that US-born college students held of an Olympic Games host country, specifically China as the

5 46 C.X. Qi et al. home of the 2008 Olympic Games, and how these risks were associated with their travel intentions. The research questions and variables included in the study are perennial issues to any Olympic host country, especially one that is hoping to grow its tourism industry. It is anticipated that the research findings will be pertinent to contemporary and future development of mega event sport tourism. Literature Review Perceptions of Risk and Tourism Perceived risk is related to a destination s image (Lepp & Gibson, 2003). While destination image has been studied since the early 1970s (Gunn, 1972; Hunt, 1975), in general the focus of this area of study has been on the cultural, natural, social, and infrastructural attributes of a destination (e.g. Echtner & Ritchie, 1993; Gartner, 1993; Baloglu & McCleary, 1999; Beerli & Martin, 2004). Indeed, an extensive body of knowledge has been amassed (Tasci & Gartner, 2007), although as Gallarza et al. (2002) observed, there is a lack of consistency in both the conceptualization and operationalization of destination image. Many destination image scales include items about such risk factors as safety and politics; however, to date a related but separate body of literature has grown up around the topic of perceived risk and tourism, even though perception of risk is inherently related to image. Certainly, identifying factors that influence perceptions of risk are likely to contribute to a better understanding of the relationship between destination image and intention to travel. This has become more pertinent in recent years as issues of tourist safety and risk have become particularly prominent (Sönmez & Graefe, 1998b), and a growing body of literature shows the negative impacts on travel and tourism accruing from a range of factors in, or around a destination (Coshall, 2003). This study follows the tradition of separating the mainstream destination image literature from perception of risk, although in the discussion and implications section a case is made for forging closer links between these two literature segments. Risk is an important factor when considering international tourism. Peace, calm, and safety are prerequisites to attracting tourists to any destination (Sönmez, 1998). In the consumer behavior literature, seven types of risk have been identified: equipment risk, financial risk, physical risk, psychological risk, satisfaction risk, social risk, and time risk. All of these have been used to examine tourism-related decisions. Equipment risk is the possibility of mechanical, equipment, or organizational problems while on vacation; financial risk is the possibility that the vacation will not provide value for the money spent; physical risk is the possibility of physical danger, injury, or sickness while on vacation; satisfaction risk is the possibility that the vacation will not provide personal satisfaction; social risk is the possibility that the vacation will affect others opinion of the person; and time risk is the possibility that the vacation will take too much time or be a waste of time (Roehl & Fesenmaier, 1992). Taking a lead from this work, Sönmez & Graefe (1998a) identified nine types of risk associated with international travel: financial, health, physical, political instability,

6 Journal of Sport & Tourism 47 psychological, satisfaction, social, terrorism, and time. For US travelers, terrorism, transportation reliability, political instability, and satisfaction risk were identified to be most often associated with international travel. In a similar study, Lepp & Gibson (2003) examined US-born young adults perceptions of risk associated with international travel. They identified seven risk factors: health, political instability, terrorism, strange food, cultural barriers, a nation s political and religious dogma, and crime. Using Cohen s (1972) tourist role typology, Lepp and Gibson found that tourists who seek more novelty in their travels, Cohen s explorers and drifters, tend to perceive less risk associated with international travel than those classifying themselves as the familiarity seeking organized mass tourists. Previous studies indicate that perceptions of risk are affected by personal characteristics (Roehl & Fesenmaier, 1992; Sönmez & Graefe, 1998b), tourist role or type (Roehl & Fesenmaier, 1992; Lepp & Gibson, 2003), previous travel experience (Sönmez, 1998; Sönmez & Graefe 1998a, b; Lepp & Gibson 2003), information search and sources (Sönmez & Graefe, 1998b; Pizam et al., 2004; Kozak, et al., 2007), life stage/ age (Gibson & Yiannakis, 2002; Lepp & Gibson, 2003), gender (Lepp & Gibson, 2003; Pizam et al., 2004), personality type (Pizam et al., 2002; 2004; Reisinger & Movondo, 2005; Lepp & Gibson, 2008), and nationality (Hurley, 1988; Tremblay, 1989; Sönmez, 1998; Seddighi et al., 2001; Pizam et al., 2004; Reisinger & Movondo, 2006; Kozak et al., 2007). Besides these, education as an indicator of social class has also been found to affect people s perception of risk (Sönmez & Graefe, 1998b). Some researchers have indicated that risk perception may be associated with personality traits. Sönmez & Graefe (1998b) identified two kinds of tourists, risk averse individuals and risk seekers. They found that risk averse individuals are likely to choose destinations perceived as safe, whereas, risk seekers are likely to show less concern about choosing destinations based on safety factors. Likewise, Roehl & Fesenmaier (1992) identified three types of tourists according to their differing levels of risk perception: risk neutral, functional risk, and place risk. Risk neutral tourists do not consider any risks associated with travel or their destination; functional risk tourists emphasize the risks associated with mechanical or equipment failure or organizational problems associated with their vacation; and place risk tourists are those individuals who perceive travel as fairly risky and the site of their last vacation as very risky. In relating these types of tourists to desired vacation attributes, Roehl & Fesenmaier suggested that risk neutral tourists tend to seek out more excitement and adventure in their vacation experiences. This line of thinking is certainly indicative of the idea that personality type might influence the degree of risk that individuals perceive in a destination. Certainly, the classic work of Plog (1974) supports this supposition. In his original study, Plog identified five types of traveler based on personality type categorized on a continuum from the pychocentrics, or those who prefer inclusive tours that give them a sense of safety and security, to the allocentrics who prefer spontaneity and adventure in their travels. More recently, Plog (2002) used the concept of venturesomeness to describe the propensity for different tourists to choose different travel styles and products. He argued that venturesomeness is a better predictor of travel intentions and choices than demographics such as income,

7 48 C.X. Qi et al. with those higher in venturesomeness traveling more frequently and choosing more adventure in their vacations. One of the best known tourist typologies was developed by Cohen (1972). Four types of international tourists were identified based on tourists preference for novelty (strangeness) or familiarity: the organized mass tourist, the individual mass tourist, the explorer, and the drifter. The organized mass tourists match most closely with the stereotypical image of tourists. They are risk averse and prefer the greatest amount of familiarity. They prefer package tours and stay mainly within their environment bubble of the familiar throughout the trip. Independent mass tourists also place a premium on familiarity and prefer the regular tourist routes. However, they travel independently and have more control over their time and itinerary. Explorers prefer a mix of familiarity and novelty. They try to get out of the environmental bubble and interact with the locals. However, compared with drifters, explorers are still careful and do not immerse themselves completely in the host society. Drifters represent the opposite extreme of the organized mass tourist and view novelty as a premium. They try to avoid the regular tourist route and totally immerse themselves in the host culture. They are the risk seekers. Pearce (1982, 1985) verified empirically Cohen s theoretical model that tourist roles could be distinguished from one another based on a number of characteristics, with the degree of familiarity and novelty they sought in a destination being one of them. Similarly, Lepp & Gibson (2003) found that the tourist role could be viewed as an indicator of the degree of novelty sought in a destination. Differences among tourists in terms of novelty seeking translate into differences in the level of risk they perceive to accompany international tourism. Therefore, novelty seekers may tolerate higher levels of risk. In their latest study, Lepp & Gibson (2008) investigated tourist role preference in conjunction with Zuckerman s (1979, 1994) Sensation Seeking Theory and found that preference for the novelty seeking roles (explorer and drifter) was associated with higher levels of sensation seeking, thereby adding some support for Cohen s hypotheses that individuals differ in the degree to which they are attracted to, or repelled by, novelty and strangeness in their travels. Tourist role preference is also related to socio-psychological needs (Gibson & Yiannakis, 2002). Gibson & Yiannakis applied Levinson s model (Levinson et al., 1978; Levinson, 1996) of the adult life course to understand stability and change in tourist role preference. They found that individuals in their 20s are most likely to prefer roles such as the drifter and explorer, choices that match their life stage characteristics and that include a desire for exploration, adventure, and experimentation. During the age 30 transition, travel decisions are increasingly made in the context of occupation, marriage, and family. The tendency for thrill-seeking type vacations diminishes (Gibson, 1996) and the type of novelty people are seeking tends to be more culturally and educationally oriented rather than physically risky. The role of the independent mass tourist also increases in popularity during the 30s and 40s. In late adulthood, tourist roles such as the seeker, the organized mass tourist, and the educational tourist are favored roles that are perceived as less risky.

8 Journal of Sport & Tourism 49 Past experience can also affect perception of risk. Experienced international tourists may perceive less risk (Sönmez & Graefe, 1998b; Lepp & Gibson, 2003; Kozak et al., 2007). However, although past international experience may provide confidence for future travel, negative experiences may make potential tourists nervous about future options. When perception of risk has a stronger influence on avoidance rather than likelihood of travel to a destination, past travel experience would overpower behavioral intentions (Sönmez & Graefe, 1998b). Gender also affects perception of risk. Although Sönmez & Graefe (1998a) failed to establish the influence of gender on perception of risk, many other researchers have found that gender does influence touristic choices (Carr, 2001; Lepp & Gibson, 2003; Pizam et al., 2004; Kozak et al., 2007). Lepp & Gibson found that women perceived a greater degree of risk regarding health and food. Kozak et al. found that older male travelers with experiences were less likely to change their travel plans when faced with potential health, terrorism, or natural disaster related risks. Pizam et al., using a combined risk and sensation seeking variable, found that males showed more propensity for adventurous and spontaneous vacations. Similarly, Carr found that among the young tourists who traveled in London, UK, there were gender differences in the perceptions of danger associated with the city at night, with more women perceiving greater risk. However, Carr also indicated that gender may not be the only influence on risk perception and tourist behavior, other factors such as personality type might also be influential. Perception of risk associated with international tourism has also been found to vary by country of origin. Hurley (1988) and Tremblay (1989) found that American tourists are more susceptible to the threat of international terrorism than European tourists. One explanation is Tremblay s suggestion that North American tourists have often been targets of terrorist violence and have been exposed to more intense media coverage of terrorist events. Kozak et al. (2007) used Hofstede s (2001) concept of Uncertainty Avoidance Theory to examine international travel related perceptions of risk. They found that residents from different countries could be characterized in terms of low, medium, and high risk avoidance, with residents of the US and Australia characterized as in the mid-range of uncertainty avoidance. Similarly, Reisinger & Movondo (2006) found that tourists from the US, in addition to those from Australia and Hong Kong, were likely to perceive more risk associated with international travel and had higher levels of anxiety. Indeed, in another related study, they found that anxiety levels had a significant impact on perceptions of safety and travel intentions (Reisinger & Movondo, 2005). Rogers (1975) Protection Motivation Theory has been used to explain travel intentions associated with perceived risk. This theory is an adaptation of expectancy-value theories and focuses on perception of risk and intention/attitude change. It postulates three crucial components of fear appeal: (i) the magnitude of the noxiousness of an environment, (ii) the probability of that event s occurrence, and (iii) the efficacy of a protective response. Protection motivation arises from the three components of fear appeal. Each of these three components initiates corresponding cognitive appraisal processes that affect attitude change. People cognitively appraise the severity and

9 50 C.X. Qi et al. likelihood of being exposed to the environment, and evaluate their ability to cope with the environment. Thus, protection motivation would be aroused, and then there would be change in behavioral intentions and attitudes. Using Roger s theory in a tourism context, Sönmez & Graefe (1998a) found that an increase in airplane accidents, crime, or terrorist activity targeting citizens of a potential traveler s nationality represents the magnitude of danger; recent occurrences involving a travel destination show the probability of occurrence; and selecting a safe destination, taking extra precautions while traveling to risky destinations, canceling travel plans, and so on can be effective actions to control consequences. After the three components have been cognitively appraised, the traveler s protection motivation will be aroused, therefore, his/her intent to travel may change. The degree of safety associated with international travel helps to determine a traveler s future travel behavior (Sönmez & Graefe, 1998b). Potential tourists tend to avoid destinations they perceive as risky and choose the ones they consider safe. Perception of risk and safety is also an important factor that helps to form an overall image of a destination. Likewise, hosting the Olympic Games is also associated with image construction, although as with general tourism work on destination image, the issue of risk perception in sport tourism has tended to be examined separately from destination image (Gibson et al., 2008). Perceived Risk and Sport Tourism Few studies exist regarding the relationship between perceived risk and sport tourism. Those studies that do exist tend to focus on hallmark and mega events in relation to the threat of terrorism. Terrorism has often been identified as the major perceived risk associated with mega events such as the Olympic Games. Due to the amount of international media attention, the Olympic Games are an attractive target and an ideal stage for terrorism (Cashman & Hughes, 1999). For example, the 1972 Palestinian attack during the Munich Olympic Games left 11 Israeli athletes dead and coverage of the incident reached a global audience of nearly 800 million viewers. The attack has been cited as a clear success by terrorists in terms of securing media attention (Schmid & degraaf, 1982). Terrorism occurred again at the 1996 Atlanta Olympic Games, where a pipe bomb exploded at Centennial Olympic Park and one person was killed and 111 were injured (Sönmez & Graefe, 1998b). The history of terrorist attacks on the Olympic Games is likely to increase levels of perceived risk among potential tourists considering attending future Olympic Games. The terrorist attacks of September 11, 2001 also increased the level of perceived risk associated with international travel in general and more specifically attending hallmark and mega events. The impact of these terrorist attacks has been highly visible and far reaching in their consequences (Toohey et al., 2003). For example, as a result of terrorism, events such as the Manchester 2002 Commonwealth Games, the 2004 Athens Olympic Games, and the 2008 Beijing Olympic Games had to employ more security personnel and implement more rigorous anti-terrorism measures. Indeed, since the 1996 Atlanta Olympic Games, the security budgets for successive

10 Journal of Sport & Tourism 51 Olympic Games have increased significantly with the budget for the 2008 Games estimated at US$2 billion (Lei, 2007). Another example of a prominent mega event that had to deal with increased security following 9/11 is the 2002 FIFA Football World Cup, which was jointly hosted by Korea and Japan. Although the event still attracted a significant number of spectators who were willing to travel, tourist numbers were less than anticipated before 9/11 (Toohey et al., 2003). Korea had to lower the anticipated number of foreign visitors to 460,000 from the original target of 640,000. In their study of the effects of terrorism on World Cup spectators in Korea, Toohey et al. found that 10% had considered not attending the World Cup because of security-related concerns and 15% had families concerned about their safety. In addition, many respondents reported that they were more anxious and nervous about attending major events because of the perceived threats to safety. Kim & Chalip (2004) in a study of US soccer club members and their interest and intent to attend the 2002 World Cup conceptualized risk as a constraint on travel in terms of threats to health and safety if they attended the event. While most of their respondents did not perceive risk to be a primary constraint, in a regression analysis Kim & Chalip found that individuals with higher incomes were motivated by escaping life s routines, had a high interest in the players, and perceived more risk associated with attending the World Cup than those who had attended previous World Cups. These individuals were interested in learning more about Korea and they liked the aesthetic appeal of soccer. Overall, similar to studies of perception of risk among general tourists, there are indications that sport-tourist-specific factors may also influence the degree to which event sport tourists perceive the risks associated with attending an event differently. Neirotti & Hilliard (2006) in a study of attendees at the Athens Olympic Games also reached this conclusion. Among their respondents, almost two-thirds reported that safety and security considerations were influential in their decisions about attending the Olympic Games. Just over a quarter reported that they knew of at least one person who had decided not to attend the 2004 Games because of safety concerns. The researchers recommend that future studies examine the diversity in Olympic tourists in terms of what factors deter some spectators but do not seem to influence others in their decisions to attend the Games. Likewise, Taylor & Toohey (2006), in a study of Rugby World Cup 2003 spectators, examined different types of rugby fans. They found that the committed rugby fans tended to report feeling safe at the matches although in general all attendees tended to diminish the threat of any danger from attending the event. Feeling safe was only one factor associated with enjoying the event among these rugby fans. Thus, taking a lead from the existing literature, the current study had three aims: (a) to investigate the perceived risks that US college students held of an Olympic Games host country, specifically China as a general tourist destination and as the host site of the Beijing Games; (b) to examine the effect of selected socio-demographic variables such as gender, previous travel experience, and tourist role on perceived risk; and (c) to examine the effect of perceived risk on intention to travel to the host country, namely China and the 2008 Olympics. To control for life stage, nationality, and education level

11 52 C.X. Qi et al. that potentially influence perceived risk and travel intentions (e.g. Lepp & Gibson, 2003), this study focused on a homogeneous population, students below 30 years of age, US born and raised, and enrolled at a university in the US. Three research questions were addressed: (1) How do US college students perceive the host country of an Olympic Games, namely China, in terms of travel-related risks? (2) Do perceptions of risk differ by gender, previous international travel experience, and tourist role? (3) What is the relationship between risk perception and intention to travel to the Olympic Games and the host country, using China and the 2008 Beijing Games as an example? Method Data Collection This cross-sectional study took place during July and August, 2004 in a large southeastern US university. Data collection was completed before the end of the 2004 Athens Games to ensure that research participants had not seen any coverage of the Beijing Games, which may have influenced their images of China. Spatial location sampling was used to identify four high foot-traffic areas on campus. Data were collected at each location on different days and times during the week. Participants were selected using systematic random sampling procedures at each location whereby every 10th person was asked to participate. Potential participants were screened for nationality and age before they were deemed eligible to participate. In total, 350 questionnaires were completed from the four locations, with a response rate of approximately 70%. Instrument This study was a part of a larger study that explored factors affecting the intention to travel to the Olympic Games and the host country. The questionnaire was developed based on a comprehensive review of the literature (e.g. Chalip et al., 1998; Lepp & Gibson, 2003; Kim & Chalip, 2004), which included five sections that are related to the current study: (a) international travel experience, (b) travel intention, (c) tourist role, (d) perception of risk, and (e) demographics. Previous international travel experience was measured by three questions: previous international travel, previous travel to Asia, and previous travel to China. Respondents were asked to report the number of trips in response to each of these questions in an ordinal scale (never, 1 2 times, 3 4 times, 5 or more times). Travel intentions were measured using questions adapted from Chalip et al. (1998). Questions included likeliness to attend the Beijing Games, intentions to travel to China in the next five years, past attendance at other Olympic Games, primary information sources for the Olympic Games, and awareness of the 2008 Beijing Olympic Games. Items were phrased in a Likert 5-point scale (1 ¼ very unlikely to 5 ¼ very likely). Tourist roles

12 Journal of Sport & Tourism 53 were measured by four statements based on Cohen s (1972) typology as developed by Lepp & Gibson (2003). Risk perceptions associated with traveling to the Beijing Olympic Games and China as the host country were measured with 19 items developed from the related literature, which covered a range of potential risks (Floyd et al., 2004; Lepp & Gibson, 2003). Items such as The threat of violence worries me about visiting China, Political stability in China is an important consideration, and Cultural difference could be a source of misunderstandings and problems were measured using a Likert 5-point scale (1 ¼ strongly disagree to 5 ¼ strongly agree). Two other questions asked participants how risky they thought the 2004 Athens Summer Olympic Games were for spectators, and how risky they thought the 2008 Beijing Olympic Games would be for spectators, which were also phrased in a Likert 5-point scale (1 ¼ very risky to 5 ¼ very safe). These two questions were based on Chalip et al. s (1998) study measuring perceptions of safety for the 2000 Sydney Summer Olympic Games among spectators. An additional question in a similar Likert 5-point scale asked respondents how they would rate the overall degree of risk associated with traveling to China. Demographic questions included gender, age, annual household income, racial background, and university class standing. Participants Of the 350 participants, 184 were male (52.7%) and 165 were female (47.3%). As would be expected on a college campus, 76.3% of participants were aged between 18 and 23 years, 15.2% were aged between 24 and 26, with the remaining 8.5% between 27 and 30 years. Just over two thirds (75.9%) of the participants had traveled internationally one or more times; however, most of them (82.9%) had never traveled to Asia or visited China (98%). According to Cohen s (1972) tourist role typology, 14.3% classified themselves as organized mass tourists, 35.7% as independent mass tourists, 40.8% as explorers, and 9.2% as drifters. Thirty-four participants (9.7%) had attended the Summer or Winter Olympic Games once before and an additional 0.6% participants reported that they had attended twice before; conversely, almost 90% had never attended the Summer or Winter Olympic Games. Among those with attendance experience at previous Games, 76% reported that they had been to the 1996 Atlanta Olympic Games. For awareness of the Beijing Olympic Games, 21.2% of the participants already knew Beijing would host the 2008 Games, while the remaining 78.8% either did not know or were not sure. When asked where the participants get most of their information about the Olympic Games, television was the primary information source, internet rated second, and newspapers ranked third. Data Analysis SPSS version 15.0 (SPSS, 2006) was used to analyze the data. Descriptive statistics were calculated for the demographic, tourist role, international travel experience, perceived risk, and travel intention variables. To understand the perceived risk that US college

13 54 C.X. Qi et al. students held of China as a tourism destination and the host of the 2008 Olympic Games, one-sample t-tests were conducted to examine the magnitude of perceived risk variables. For the purpose of conducting data reduction and examining the dimensionality of the perceived risk variables that were related to China as a tourist destination and a host site of the Olympic Games, a factor analysis with principal component extraction and varimax rotation was conducted. Conducting the factor analysis to explore the dimensionality of risk perceptions was based on the consideration that this study was exploratory in nature. Although general theories and literature of perceived risk were available to some extent, no systematic study had been found that was specifically related to China as a general tourist destination and as a host of the Olympic Games. Multiple regression analyses were conducted to examine the predictability of perceived risk factors on the travel intention variables. Independent group t-tests and analyses of variance (ANOVA) were conducted to examine the influence of gender, tourist roles, and previous international travel experience on the perceived risk factors. Results Research findings are presented in the following three sections: (a) descriptive statistics, (b) impact of perceived risk on intention to travel, and (c) impact of background variables on perceptions of risk. Descriptive Statistics Calculations of descriptive statistics revealed that respondents strongly agreed that proper sanitation and hygiene in China are very important (M ¼ 4.08; SD ¼ 0.86) and I prefer traveling to China if I knew something about it (M ¼ 3.95; SD ¼ 0.86). Yet, they strongly disagreed that I might be disappointed if I took a trip to China (M ¼ 1.99; SD ¼ 0.88) and there is a risk of friends/family/associates disapproving of my choice to travel to China (M ¼ 2.20; SD ¼ 1.01). The overall risk associated with traveling to China was a mean equal to 3.27 (SD ¼ 0.75) on a Likert 5- point scale, with 5 being very safe. Over 48% of the participants considered traveling to China was neither risky nor safe ; 38.1% considered it safe or very safe ; and 13.9% participants indicated it was risky or very risky (Table 1). Participants were also asked to rate the risk level associated with the Beijing Olympic Games. The overall mean risk of attending the 2008 Beijing Olympics perceived by the respondents was equal to 3.35 (SD ¼ 0.80) on a Likert 5-point scale, with 5 being very safe. Just under half (44.5%) rated the Beijing Olympic Games as safe or very safe. Almost 42% of them rated the Games as neither risky nor safe, while 14% considered them risky or very risky. As a point of comparison, participants were also asked to rate previous and future Olympic Games host sites. Canada was perceived to be the safest host country (M ¼ 4.32; SD ¼ 0.76), followed by Australia (M ¼ 4.14; SD ¼ 0.79). Korea was perceived as the riskiest Olympic host site (M ¼ 2.76; SD ¼ 0.96), whereas China was rated similar to Greece (M ¼ 3.34; SD ¼ 0.85) as a moderately risky country.

14 Table 1 US College Students Perceptions of Risk Associated with Traveling to China as the Olympic Games Host Country (N ¼ 350) Risk Items Mean a SD t b p Proper sanitation and hygiene in China are important I prefer traveling to China if I knew something about it I would not travel to China if one of its neighboring countries was at war. Language barriers could be a source of misunderstandings and problems. Political stability in China is an important consideration The threat of terrorism could influence my decision to travel to China. I would worry about pick-pockets and petty thieves I would not like to stand-out when traveling in China Cultural differences could be a source of misunderstandings and problems. Standards of health care in China concern me Drinking the water would not be a good thing to do when traveling in China. China s political orientation is a concern for me I would have concerns about rustic/primitive accommodations if I plan to travel to China. It is important to know about China s religious orientation before taking a trip there. The threat of violence worries me about visiting China HIV and other infectious diseases are a danger in China I prefer to eat food that is familiar to me when traveling in China There is a risk of friends/family/associates disapproving of my choice to travel to China. I might be disappointed if I took a trip to China a Measured using a Likert-type format where 1 ¼ Strongly Disagree and 5 ¼ Strongly Agree. b One Sample t-test with respect to 3.0 as the midpoint. Significant at the 0.05 level. Journal of Sport & Tourism 55 Conducting the factor analysis, four factors emerged when adopting the criterion of an eigenvalue equal to or greater than 1.0, with reference to the scree plot. A total of 55.8% of variance among the variables were explained by the factors. Following the criteria of a factor loading equal to or greater than.40 and without double loading (Nunnally & Bernstein, 1994), 17 of 19 perceived risk items were loaded on one of the four factors. Two items I prefer traveling to China if I knew something about it and Proper sanitation and hygiene in China are important were eliminated due to low factor loading values. The factors were labeled as Personal Safety (M ¼ 3.17; a ¼ 0.77), Cultural Risk (M ¼ 3.21; a ¼ 0.71), Socio-psychological Risk (M ¼ 2.14; a ¼ 0.60), and Violence Risk (M ¼ 3.32; a ¼ 0.72), where weighted factor mean scores and Cronbach alpha coefficients are included in the parentheses. All four factors displayed acceptable internal consistency (Table 2).

15 56 C.X. Qi et al. Table 2 Factor Analysis of the Perceived Risk Variables Risk Factor/Items Factor 1 Factor 2 Factor 3 Factor 4 Factor 1 Personal Safety I would have concerns about rustic/primitive accommodations if I plan to travel to China. Standards of health care in China concern me HIV and other infectious diseases are a danger in China. Language barriers could be a source of misunderstandings and problems. I would worry about pick pockets and petty thieves Drinking the water would not be a good thing to do when traveling in China Factor 2 Cultural Risk Cultural differences could be a source of misunderstandings and problems. I would not like to stand out when traveling in China. China s political orientation is a concern for me Political stability in China is an important consideration. It is important to know about China s religious orientation before taking a trip there. Factor 3 Socio psychological Risk There is a risk of friends/family/associates disapproving of my choice to travel to China. I might be disappointed if I took a trip to China I prefer to eat food that is familiar to me when traveling in China. Factor 4 Violence Risk I would not travel to China if one of its neighboring countries was at war. The threat of terrorism could influence my decision to travel to China. The threat of violence worries me about visiting China Eigenvalue Cronbach s Alpha Factor Mean % of variance explained Cumulative variance explained Impact of Perceived Risk on Intention to Travel Utilizing the weighted factor mean scores, the regression analyses revealed that the perceived risk factors were negatively (R 2 ¼ 0.05, p, 0.05) predictive of participants intention to visit the Olympic Games host country (i.e. China) as a tourist destination.

16 Table 3 Multiple Regression Analysis on the Impact of Perceived Risk on Intention to Travel to China as the Olympic Games Host Country in the Next 5 Years Risk Factor b Std. Error Standardized b t p (Constant) Health Risk Cultural Risk Socio-psychological Risk Violence Risk R 2 ¼.05 Adjusted R 2 ¼ F ¼ p ¼ Significant at the 0.05 level. Journal of Sport & Tourism 57 The Violence Risk factor (b ¼ 0.160, p ¼ 0.050) and the Socio-psychological Risk factor (b ¼.0202, p ¼ 0.052) negatively affected intention to travel to China in the next 5 years. Every unit of change in the Violence Risk factor translated into a change in intention to travel. As for the Socio-psychological Risk factor, every unit of change resulted in a change in intention to travel to China in the next 5 years. Interestingly, although not statistically significant, the Cultural Risk factor was positively related to intention to travel to China within the next 5 years (b ¼ 0.084, p ¼ 0.404). These findings on the impact of perceived risk factors are summarized in Table 3. Although not statistically significant, the four risk factors were found to be negatively predictive of the intention to attend the Olympic Games (R 2 ¼ , p. 0.05) (Table 4). Of them, the Health Risk factor (b ¼ 0.087, p ¼ 0.180) and the Sociopsychological Risk factor (b ¼ 0.073, p ¼ 0.258) were more predictive of intention to travel to the Olympics Games when compared with the other two factors, Cultural Risk (b ¼ 0.001, p ¼ 0.987) and Violence Risk (b ¼ 0.018, p ¼ 0.723). Impact of Background Variables on Perceptions of Risk Conducting independent group t-tests, significant (p, 0.05) difference was found between males and females in terms of mean Violence Risk scores (Table 5). Females Table 4 Multiple Regression Analysis on the Impact of Perceived Risk on Intention to Attend the Beijing Olympic Games Risk Factor b Std. Error Standardized b t p (Constant) Health Risk Cultural Risk Socio-psychological Risk Violence Risk R 2 ¼.023 Adjusted R 2 ¼ F ¼ p ¼ Significant at the 0.05 level.

17 58 C.X. Qi et al. Table 5 Independent Group T-Tests Examining Travel-Related Risks by Gender Males Females Risk Factor Mean a SD Mean SD Health Risk Cultural risk Socio-psychological Risk Violence Risk a Measured using a Likert-type format where 1 ¼ Strongly Disagree and 5 ¼ Strongly Agree. Significant at the 0.05 level. t p (M ¼ 3.42, SD ¼ 0.87) perceived Violence Risk higher than males (M ¼ 3.23, SD ¼ 0.95). For the other three risk factors, although not statistically significant, males had slightly higher scores on Health Risk (Males: M ¼ 3.20, SD ¼ 0.69; Females: M ¼ 3.15, SD ¼ 0.65) and Cultural Risk (Males: M ¼ 3.22, SD ¼ 0.77; Females: M ¼ 3.18, SD ¼ 0.69); whereas, males and females were not different in Sociopsychological Risk (Males: M ¼ 2.14; SD ¼ 0.75; Females: M ¼ 2.15; SD ¼ 0.73). ANOVA revealed that mean Socio-psychological Risk and Violence Risk factors were significantly (p, 0.05) different among tourist role types. Tukey HSD post hoc analyses revealed that organized mass tourists were higher in Socio-psychological Risk than independent mass tourists and explorers, and in turn higher than drifters. For Violence Risk, organized mass tourists and independent mass tourists were higher than explorers, and in turn higher than drifters (Table 6). Furthermore, ANOVA revealed no significant (p. 0.05) relationship between previous international travel experiences and the level of perceived risk (Table 7). Discussion and Implications Using China and the 2008 Beijing Olympic Games as an example, this study investigated the perceived risk factors that US-born college students held of an Olympic Games host country and their associations with travel intentions. The findings Table 6 ANOVA Examining Travel-Related Risks by Tourist Roles Risk Factor Organized Mass Tourists Independent Mass Tourists Explorers Drifters F p Health Risk Cultural risk Socio-psychological Risk Violence Risk Significant at the 0.05 level. Significant differences utilizing Tukey HSD post hoc analysis.