Impact of the Perceived Image of Celebrity Endorsers on Tourists’ Intentions to Visit

Abstract

The purpose of this study is to empirically assess the mediating effects of the impact of the perceived image of celebrity endorsers on tourists’ intentions to visit, using celebrity-endorsed print advertisements for travel destinations. The results indicate that celebrity endorsers have a significant impact on people’s attitudes and visit intentions, thus verifying the mediating effects of this variable. The study also provides clues to what extent celebrity-endorsed advertisements differ from nonendorsed advertisements and explores such differences in terms of destination match-up between native and nonnative celebrity-endorsed advertisements.

Keywords

celebrity endorsement destination marketing print advertisements structural equation modeling

Introduction

Marketing campaigns featuring celebrity endorsers are thought to have a positive impact on people’s attitudes and lead to positive behavioral intentions. This is certainly one of the reasons why Jackie Chan was appointed by the United Nations World Tourism Organization as Asia’s Tourism Ambassador at the Asia-Pacific Tourism Ministerial Conference in Macau (June 13, 2006). The aim of his appointment is to promote Asia in order to stimulate more tourist arrivals and boost the region’s image. For similar reasons, he serves as Hong Kong’s Tourism Ambassador, a position to which he was officially appointed in 1995. Celebrity endorsement has played a prominent role in Australian tourism marketing as a means of engaging with potential tourists. The Australia—A Different Light campaign was launched featuring various Australian celebrities (Glover 2009). More recently, Tourism Australia teamed up with Oprah Winfrey in December 2010 and hosted a marketing campaign to persuade North American tourists to visit.

The idea behind celebrity-endorsed destination campaigns is simple: to draw attention to the endorsed destination and to complement the desired image values. Celebrities are the preferred choice because they act as role models and are powerful cultural, financial, and media figures and image creators (Andrews and Jackson 2001). Essentially, celebrities are favored because they are more likely to capture people’s attention than an unknown person (Agrawal and Kamakura 1995; Hsu and McDonald 2002; Misra and Beatty 1990). This phenomenon is related to society’s obsession with fame, which has existed since ancient times (Braudy 1997). Appointing famous spokespeople as endorsers dates back at least to the 1800s, and the aim of such appointments has always been to increase the persuasiveness of the message (Knott and St. James 2004). The ultimate aim is to influence people’s behavioral intentions positively, which may then lead them to take action. However, celebrity endorsement does not always directly influence people’s behavioral intentions (Johnson and Harrington 1998; Ohanian 1991; Stafford, Stafford, and Day 2002). We propose that the impact of celebrity endorsement on behavioral intentions is mediated by attitude, because attitude is conceptualized in many previous studies as an important mediator of behavioral intentions (Ajzen and Fishbein 1980; Armitage and Conner 2000; Sheppard, Hartwick, and Warshaw 1988).

Mediation analysis provides a clue to the sequence of possible effects that lead to behavioral intentions. Ohanian’s (1991) model is particularly appropriate for such an analysis because it examines the impact of celebrity endorsers’ perceived image on consumers’ intention to purchase. Similar to buyers of general consumer products, individuals pass through several stages when making a travel decision. Intentions to visit are a commonly researched topic (Lee 2009; Mechinda, Serirat, and Gulid 2009), and several studies have indicated that attitude is an important factor in tourist choice behavior (Mohsin 2005; Pike 2006; Um and Crompton 1991). Some studies have empirically demonstrated that attitudes toward a destination influences travelers’ preferences and intentions (Court and Lupton 1997; Milman and Pizam 1995). This leads to the twofold research question: (1) Do celebrity endorsers have a positive impact on peoples’ attitudes? and (2) If so, do these attitudes lead to positive visit intentions? This study answers these queries using a modified conceptual framework and a set of hypotheses within the context of Chinese tourists potentially visiting Hong Kong. Past conceptualizations of celebrity endorsement for intangible products are limited, with the current study extending the tradition of endorsement studies often carried out in North America among student participants in laboratory settings. Research that collects information on several indices within a field design is better able to identify whether attitude is part of a range of other outcome variables. To increase the generalizability of the model, it is important to replicate the framework among different populations and for different product types, as it is possible that the structural relationships between the proposed factors or impacts do not occur elsewhere.

The majority of the celebrity endorsement studies use group difference methods, such as ANOVA or MANOVA, to examine the endorser effects and advertising effectiveness. However, this study uses structural equation modeling to analyze interrelationships among latent variables and their observable manifest variables (Kline 2005; Reisinger and Mavondo 2006). While MANOVA is acceptable for an emergent variable system, structured means modeling is more appropriate when the interest is in the comparison of latent variable means because it accounts for the unreliability of measures (Thompson and Green 2006). Celebrity endorsement in the tourism literature is perceived as a potential and effective stimulus; however, extant studies are either experimental or conceptual and do not use structural equation modeling (Chang, Wall, and Tsai 2005; Glover 2009; Higham and Hall 2003; Lee, Scott, and Kim 2008; Magnini, Garcia and Honeycutt 2010; Van der Veen 2008).

The nature of this study is confirmatory, and it empirically tests the proposed structural relationships while taking into account the mediating constructs of cognitive processing. Apart from other external effects, the model examines if people form a favorable attitude toward a destination and advertisement for that destination which is partly due to a celebrity’s endorsement. It considers attitude as a possible significant and necessary antecedent of consequent behavioral intentions and suggests that it be included in relevant follow-up models. The remainder of the article is organized as follows. The next section reviews the phenomenon of celebrity endorsement and introduces the conceptual framework used in this study. The third section discusses the measurement model and method used to test the hypotheses. The fourth section presents the empirical results of both the structural equation model analysis and the structured means analysis. The final section discusses the results and the managerial implications.

Literature Review

Consumer research regards celebrity endorsement as a type of extrinsic cue intended to positively affect people’s perceptions of a product or brand (Dean 1999). Considerable emphasis has been placed on various credibility models concerning the effects of certain stimuli on consumer behavior, thus satisfying the idea that emotions or moods trigger buying responses (Gardner 1985). It is assumed that the effect elicited by contextual stimuli becomes associated with the product or brand, in turn leading to positive feelings and increased likelihood of purchase or action (Gorn 1982; Morris and Boone 1998).

Various models have been proposed to illustrate the endorsement process, of which the elaboration likelihood model by Petty and Cacioppo (1986) is the most popular. This model shows how the elaboration of advertising works, taking into account involvement and whether persuasion follows a central, peripheral, or dual route. With the central route, a person actively considers and judges information related to the message. However, if a person lacks the motivation or ability to think about the information, then it is believed that persuasion follows a peripheral route. The intention of persuasion is to make someone do something, especially by reasoning, urging, or inducing, and thus communication can either be persuasive or unpersuasive (Hunt 1993).

Rather than analyzing the effects on persuasion, McCracken (1989) suggests that the effectiveness of endorsement is better explained by the “meanings” that people associate with the celebrity endorser, which are then transferred to the product or brand. McCracken (1989) provides one of the most frequently cited definitions of a celebrity endorser, which is “any person who enjoys public recognition and who uses this recognition to endorse consumer goods or services by appearing with it in an advertisement” (McCracken 1989, p. 310). Some examples of meaning transfers are Tiger Woods with Nike and Cindy Crawford with Omega. Some features of the sports star, or his “activeness,” is supposed to be transferred to the sports brand, and the image of the actress, or her “elegance,” is transferred to the watch brand (De Mooij 1997).

Celebrity-endorsed print advertisements for a destination can thus be considered successful or persuasive when the properties of the celebrity are in line with, or are transferred to, the properties of the endorsed destination. From this perspective, it appears that the major objective is to show the celebrity in such a way that the target group associates the endorsed destination or brand with the celebrity.

The phenomenon of association was clearly described and proposed as the Match-up hypothesis by Kamins (1990) and will be examined using structured means analysis in this study. However, apart from an effective association, it is necessary to identify the factors that accurately measure those features that represent the celebrity endorser and to determine whether they actually have an impact on the desired outcomes. Ohanian (1990, 1991) explored these issues, and her three-factor source credibility model is held to be the most appropriate model for understanding the factors that explain the impact of celebrity-endorsed print advertisements. Ohanian’s model is in accordance with Source Model Theory, which illustrates that endorsements are usually effective because of the credibility and attractiveness of the source (Sternthal and Craig 1973). Van der Veen and Song (2010) replicated Ohanian’s measurement model in a tourism context, and their results indicate that certain changes are required. Based on their results, the current study further enhances the applicability of the model by specifying a structural equation model for hypothesis testing and compares different groups using structured means analysis (see Figure 1).

Figure 1.

Conceptual framework for celebrity destination endorsement.

One of the antecedents in the model is the perceived attractiveness of a celebrity endorser. Physical attractiveness is included because many studies have found that this feature is one of the most visible and accessible characteristics of a person that others use when making a judgment (Caballero and Solomon 1984; Patzer 1983). However, Ohanian’s (1991) empirical study reports a nonsignificant direct relationship between attractiveness and purchase intentions. She argues that although most celebrity endorsers are physically attractive, attractiveness may not be an important factor in respondents’ final decision-making processes. As Johnson and Harrington (1998) indicate, it remains uncertain whether attractive celebrity endorsers are able to create purchase intentions, but it is likely that they positively improve attitudes toward advertising and brands. Baker and Churchill (1977) find that physically attractive communicators can be successful in changing consumer attitudes. Other source credibility studies similarly indicate that the physical attractiveness of a celebrity endorser has a positive impact on people’s attitudes toward an advertisement and a brand (Kahle and Homer 1985; Kamins 1990; Patzer 1983; Petroshius and Crocker 1989). Therefore, we include the physical attractiveness of a celebrity endorser as an antecedent and propose that the total effect may be mediated by attitude.

Hypothesis 1: Attractiveness positively affects attitudes toward an advertisement.

Hypothesis 2: Attractiveness positively affects attitudes toward a destination.

The two other perceived celebrity image constructs examined in Ohanian’s study were expertise and trustworthiness. However, Van der Veen and Song (2010) were unable to identify these two distinctive factors and revised the measurement model accordingly. They proposed a construct that consists of perceived expertise combined with two manifest variables of trustworthiness, which they call “believability.” The believability construct represents the knowledge to support the claims made in the advertisements that the communicator is perceived to possess. At the same time, it represents the extent to which these claims are provided in an honest and trustworthy way. According to Maddux and Rogers (1980), expert endorsements enhance the believability of an advertisement, primarily because of the greater source credibility. This is important because research on the persuasive effect of credible sources versus less credible sources has consistently demonstrated that credible sources produce greater attitude change (Craig and McCann 1978; Erdogan 1999). Experts are more credible because they are assumed to be more efficient external information seekers and better able to discriminate between relevant and irrelevant information (Alba and Marmorstein 1987; Hutchinson and Moore 1984). However, it is irrelevant whether a celebrity endorser is actually an expert on the subject; the issue is how the target audience perceives the expertise of the celebrity endorser (Ohanian 1991). Therefore, we examine perceived expertise: “the extent to which a communicator is perceived to be a source of valid assertions” (Hovland, Janis, and Kelly 1953, p. 21). Although various studies indicate that the endorser’s perceived expertise is important and affects attitudes and intentions to buy an endorsed brand (Daneshvary and Schwer 2000; Erdogan 1999; Lafferty and Goldsmith 1999; Till and Busler 1998), other scholars have found that it does not necessarily affect purchase intentions (Stafford, Stafford, and Day 2002). We thus propose that the direct relationship between the believability of a celebrity endorser and behavioral intentions is mediated by attitude.

Hypothesis 3: Believability positively affects attitudes toward an advertisement.

Hypothesis 4: Believability positively affects attitudes toward a destination.

As stated in the foregoing hypotheses, we assume that the relationships between the characteristics of the celebrity endorser and visit intentions are mediated by attitude. We include attitude because it serves as a reliable indicator of how people act in a given set of circumstances (Mayo and Jarvis 1981). Attitude can be regarded as “a psychological tendency that is expressed by evaluating a particular entity with some degree of favour or disfavour” (Eagly and Chaiken 1993, p. 1). Attitude in the current study refers to a general positive or negative feeling about an object (i.e., an advertisement and a destination), following the approach of Petty and Cacioppo (1981).

The evaluative responses to advertisements persuading consumers are often measured in terms of attitude, such as attitude toward the advertisement and the brand (Brown and Stayman 1992; Shimp 1981). Attitude toward the advertisement and the brand are both defined as learned predispositions to respond in a consistently favorable or unfavorable manner (MacKenzie and Lutz 1989). The positive effect of the advertisement on intentions is intertwined with the positive effect of the advertisement on the attitude toward the destination. Various consumer studies support the mediating effects of attitude toward the advertisement and the brand on purchase intentions (Biehal, Stephens, and Curlo 1992; Homer 1990; Miniard, Bhatla, and Rose 1990). The outcome variable for this study is visit intentions, which is often a desired strategic aim for a destination marketing organization. Intention is the likelihood of a person intending to do something or a self-prediction or expectation that one will act (Ajzen and Fishbein 1980; Bagozzi, Gurhan-Canli, and Priester 2002). The following three hypotheses focus on the relationship between the attitudinal variables and their presumed impact on visit intentions and explore the mediating nature of these constructs in the framework.

Hypothesis 5: Attitude toward an advertisement positively affects the intention to visit a destination.

Hypothesis 6: Attitude toward an advertisement positively affects attitude toward the destination.

Hypothesis 7: Attitude toward a destination positively affects the intention to visit that destination.

Method

A conceptual framework is proposed to explain and predict the impact of celebrity endorsement on people’s attitudes and intentions to visit a destination. Structural equation modeling is used to assess the relationships because it is the preferred method for analyzing relationships among dimensions and their manifest variables (Reisinger and Mavondo 2006).

The print advertisements presented to respondents showed a color photo of a celebrity endorser and a picture of the famous Wan Chai harbor front in Hong Kong and the official slogan in the background. The advertisements are not presented here due to space limitations. The respondents were randomly exposed to one of the four advertisements featuring one celebrity endorser (Andy Lau, Maggie Cheung, David Beckham, or Britney Spears). The variance across treatment groups will be discussed after the hypothesis testing using structured means analysis. We ensured that celebrities were familiar to the respondents and the study focuses on novices to minimize the effect of past behavior in predicting travel intentions. The respondents were asked a single filter question “Have you visited Hong Kong before?” to exclude repeat travelers to Hong Kong. Controlling for past visits is important because research indicates that the best predictor of behavioral intentions and future actual behavior is the frequency of past relevant behavior (Opperman 1999; Sonmez and Graefe 1998; Quellette and Wood 1998). Woodside and Lysonski (1989) demonstrate that previous destination travel experiences are positively related to the inclusion of that destination in a consumer’s consideration set versus other mental categories of travel destinations. However, perceptions can be formed from external sources, and these preconceived notions are largely unavoidable and may present a limitation to this study.

Nearly half of all celebrity endorsement studies use student samples (Amos, Holmes, and Strutton 2008). The use of student samples is a persistent custom in research on advertisement attitudes, and these studies tend to produce larger effects than those involving nonstudent samples (Brown and Stayman 1992). Therefore, the current study employs a field study approach and minimizes the number of student participants to provide greater confidence in conclusions regarding the generalizability of the postulated relationships. This study focused on people over the age of 20 who were financially independent in an effort to represent potential mainland Chinese tourists as faithfully as possible. The destination was Hong Kong, and the target respondents were residents of Guangzhou. The Hong Kong Tourism Board gives marketing priority to Guangzhou because it produces the largest number of overnight visitor arrivals to Hong Kong and is a large target market for short-break visitors to the territory (Huang and Hsu 2005).

The questionnaire used a 7-point semantic differential scale format to measure the manifest variables. The respondents rated the celebrity endorser in terms of attractiveness and believability in response to the questions “Would you say [name of celebrity] is . . .?” using the following scales: attractive–unattractive, classy–not classy, handsome/beautiful–ugly, or elegant–plain; and “How believable do you judge him or her to be?” using the scales: expert–not an expert, experienced–inexperienced, knowledgeable–unknowledgeable, qualified–unqualified, skilled–unskilled, honest–dishonest, and trustworthy–untrustworthy (Ohanian 1990, 1991; Pornpitakpan 2003; Van der Veen and Song 2010). To measure attitude toward the destination, the respondents were asked “How would you describe your overall attitude toward Hong Kong?” rated on a 7-point scale anchored by good–bad, dislike–like, pleasant–unpleasant, positive–negative, and favorable–unfavorable (MacKenzie and Lutz 1989). The 10 items used to measure attitude toward the advertisement included both cognitive and affective components. The components were merged together because of their strong interrelationship, which eventually determines the predisposition for visiting a destination (Gartner 1993).

Five items measured the cognitive component: unpersuasive–persuasive, uninformative–informative, unbelievable–believable, ineffective–effective, and unconvincing–convincing (Burton and Lichtenstein 1988). Five items measured the affective component, with endpoints labeled bad–good, unappealing–appealing, unattractive–attractive, unpleasant–pleasant, and unlikable–likable (Janiszewski 1988). Visit intention was measured by the question “How likely is it that you will visit Hong Kong in the next 12 months?” with the following scales: probable–improbable, likely–unlikely, and possible–impossible (Lam and Hsu 2006; Yi 1990).

To ensure a high level of clarity, researchers suggest translating questionnaires using the double-translation method (De Mooij 1997; McGorry 2000). Following this approach, the questionnaire was first translated into the target language (Mandarin Chinese) before being translated back into the original language (English). Six interpreters fluent in both languages, and holding postgraduate degrees, translated the questionnaire using this method.

Results

The survey was conducted on Saturday, November 25, 2007, at five shopping locations in Guangzhou from early morning to late evening. Many residents visit these locations over the weekend, and interviewers approached shoppers. A local research company was contracted to distribute the questionnaires. Out of the 1,473 people who were approached, 336 rejected (23%) and 93 did not meet the selection criteria (6.3%), yielding 1,044 questionnaires that were collected and used for the study. The quota sampling method, although not as rigorous as probability sampling, was chosen as the most cost-effective means of obtaining a representative sample of potential visitors to Hong Kong. The quotas were set according to demographic characteristics of the visitor profile published by the Hong Kong Tourism Board (Table 1). Descriptive statistics are presented in Table 2. In terms of data distribution, Kline (2005) recommends cutoff values of 3 for skewness and 8 for kurtosis. The univariate skewness ranged from 0.728 to 2.356 and the univariate kurtosis ranged from −0.481 to 5.764. Mardia’s coefficient was 1.74; thus, there was no extreme violation of the normal distribution and the data were appropriate for analysis.

Table 1.

Demographic Characteristics of the Sample.

Item	Valid n	Study (%)	HKTB 2006 (%)
Gender
Male	361	51.1	43
Female	345	48.9	57
Marital status
Married	391	55.4	70
Other	315	44.6	30
Age
20–25	316	44.8	21
26–35	243	34.4	37
36–45	121	17.1	24
46–55	22	3.1	11
56–65	4	0.6	5
>66	–	–	2
Occupation
Working	533	75.5	75
Housewife	36	5.1	10
Student	89	12.2	7
Retired	2	0.3	6
Other	49	6.9	2
Education
No education	13	1.8	1
Primary	3	0.4	3
High school	135	19.1	33
College	530	75.1	60
Postgraduate	25	3.5	5
Monthly income (RMB)
<1000	67	9.5
1001–3000	338	47.9
3001–6000	180	25.5
6001–9000	29	4.1
>9001	10	1.4
No regular income	82

Note: HKTB = Hong Kong Tourism Board.

Table 2.

Descriptive Statistics.

	Mean	Standard Deviation	COV%
Attractiveness
Elegant	2.26	1.51	66.77
Classy	2.06	1.37	66.55
Attractive	1.92	1.26	65.63
Handsome/beautiful	1.87	1.15	61.71
Believability
Knowledgeable	2.72	1.92	70.48
Honest	2.46	1.59	64.43
Trustworthy	2.45	1.59	64.86
Expert	2.42	1.60	65.95
Qualified	2.38	1.65	69.37
Skilled	2.16	1.41	65.28
Experienced	2.14	1.47	68.69
Attitude toward Hong Kong
Pleasant	2.03	1.34	65.91
Positive	1.92	1.19	61.77
Like	1.81	1.16	63.92
Good	1.75	1.05	59.77
Favorable	1.74	1.18	67.82
Attitude toward the advertisement
Informative	3.10	1.86	59.94
Persuasive	2.86	1.69	59.13
Convincing	2.75	1.62	58.84
Effective	2.73	1.61	59.05
Appealing	2.56	1.58	61.88
Like	2.56	1.58	61.76
Believable	2.55	1.66	65.10
Attractive	2.55	1.57	61.69
Pleasant	2.42	1.47	60.66
Good	2.40	1.43	59.38
Visit intentions
Probable	2.72	2.11	77.61
Likely	2.70	2.05	75.96
Possible	2.21	1.71	77.47

Note: The scale ranges from 1 (positive) to 7 (negative).

The factor structure was then subjected to confirmatory factor analysis, using the LISREL 8.80 software package (Scientific Software International, Inc., Lincolnwood, IL). In terms of model fit, Hair et al. (2006) suggest reporting the Chi-square value and the associated degrees of freedom, the comparative fit index (CFI), and the root mean square error of approximation (RMSEA), since together they provide sufficient information to evaluate the model fit. The normed fit index (NFI) of Bentler and Bonett (1980) has shown good overall performance in simulation studies and is reported here (Chou and Bentler 1995; Kaplan 2000). The minimum threshold of acceptability for both the NFI and the CFI is 0.90 (Chau 1997). Hu and Bentler (1999) recommend that an RMSEA value of 0.06 or less indicates a good fit between the hypothesized model and the observed data.

After removing the control group from the data set and using the listwise deletion option, a sample size of 706 was retained and used for confirmatory factor analysis. The fit indices for the measurement model are χ² = 1273.47, df = 367, p < 0.01, CFI = 0.97, NFI = 0.96, and RMSEA = 0.059. The majority of the alternative fit indices illustrate an acceptable fit and all factor loadings are significant (see Table 3). Further modifications to improve model fit were not conducted, because data-driven model modifications tend to capitalize on chance and features of the sample, making it less likely that the model will be generalizable to other samples (Golob 2003; MacCallum, Rosnowski, and Necowitz 1992; Silvia and MacCallum 1988).

Table 3.

Results Confirmatory Factor Analysis (n = 706).

Factor/Item	Standard Factor Loading	t-Values	Composite Reliability	Average Variance Extracted
Attitude toward the advertisement			0.95	0.64
Convincing	0.86	28.56
Like	0.86	28.37
Good	0.84	27.59
Attractive	0.84	27.48
Appealing	0.84	27.35
Effective	0.79	24.69
Pleasant	0.78	24.61
Informative	0.74	22.50
Persuasive	0.73	22.14
Believable	0.73	22.06
Visit intentions			0.90	0.75
Likely	0.92	30.19
Probable	0.89	28.65
Possible	0.78	23.94
Believability			0.90	0.56
Qualified	0.83	26.17
Honest	0.77	23.45
Experienced	0.76	22.98
Trustworthy	0.74	22.30
Skilled	0.73	21.85
Knowledgeable	0.72	21.43
Expert	0.68	19.99
Attitude toward Hong Kong			0.88	0.59
Favorable	0.79	23.94
Good	0.79	23.92
Like	0.78	23.57
Positive	0.76	22.94
Pleasant	0.73	21.43
Attractiveness			0.81	0.52
Attractive	0.80	23.77
Handsome/Beautiful	0.78	22.71
Classy	0.70	19.69
Elegant	0.57	15.14

Note: χ2 = 1273.47, df = 367, p < 0.01, comparative fit index = 0.98, normed fit index = 0.97, root mean square error of approximation = 0.059.

The composite reliability values are all 0.70 or higher, which suggests strong reliability (Bagozzi and Yi 1988). The average variance extracted reached the minimum threshold of 50% and was higher than the square of the correlation estimates. The correlation matrix of the latent variables indicated that the derived constructs were consistent with the theoretically derived dimensions and were sufficiently distinct (see Table 4). The program default settings in estimating the model were used and the moment matrix is available on request.

Table 4.

Correlation Matrix of the Latent Variables.

	Attractiveness	Believability	Attitude toward Hong Kong	Attitude toward the Advertisement
Believability	0.64 (0.41)
Attitude toward Hong Kong	0.48 (0.23)	0.43 (0.18)
Attitude toward the advertisement	0.45 (0.20)	0.68 (0.46)	0.42 (0.18)
Visit intentions	0.18 (0.03)	0.17 (0.03)	0.26 (0.07)	0.25 (0.06)

Note: All correlations are significant at the p <0.01 level, with squared correlations in parentheses.

The results for the structural model test with the same data set (n = 706) indicate an acceptable fit (χ² = 1274.08, df = 369, p < 0.01, χ²/df = 3.45, CFI = 0.98, NFI = 0.97, RMSEA = 0.059, 90% confidence interval for RMSEA = 0.055 – 0.062). Allocating a large sample size to a structural test can inflate the chi-square values for statistical significance, erroneously implying a poor data-to-model fit (Schumacker and Lomax 2004). Chi-square estimates tend to be sensitive to small model discrepancies when sample sizes are larger than 200 or when the model contains a large number of variables (Bagozzi and Yi 1988; Bollen and Long 1993). Cudeck and Henley (1991) argue that virtually all proposed models are rejected with large samples, although large samples are desirable to minimize sampling bias. To remedy this issue, Hair et al. (2006) suggest setting the sample size to 200 in the LISREL command line. This procedure yielded a nonsignificant chi-square (χ² = 359.63, df = 369, p = 0.63, χ²/df = 0.97, RMSEA = 0.00), which indicates a good model fit.

The structural equation model results are illustrated in Figure 2, and hypotheses 1 and 4 were not supported. The structural path estimate between attractiveness and attitude toward the advertisement had a maximum likelihood estimate of 0.03 and a t-value of 0.62. The results also showed a nonsignificant relationship between believability and attitude toward Hong Kong (γ = 0.05/t-value = 0.73). The remaining hypotheses were supported. The results indicated that believability had a statistically significant and positive impact on the respondents’ attitudes toward the advertisement (γ = 0.66/t-value = 13.48). Attitude toward Hong Kong was positively affected by attractiveness (γ = 0.35/t-value = 6.17). Likewise, attitude toward the advertisement had a positive and statistically significant impact on attitude toward Hong Kong (β = 0.23/t-value = 4.32). The two attitude factors, attitude toward the advertisement (β = 0.17/t-value = 3.78) and attitude toward Hong Kong (β = 0.19/t-value = 4.24), both evidenced a statistically significant impact on visit intentions.

Figure 2.

Structural model for celebrity destination endorsement.

The structural equation model results indicate that mainland Chinese respondents who had a positive attitude toward both the advertisement and Hong Kong intend to visit Hong Kong in the next 12 months. Given that five out of the seven estimates were consistent with the hypotheses, the results generally supported the theoretical model, although the path between attractiveness and attitude toward the advertisement and believability and attitude toward Hong Kong were not supported. The squared multiple correlations were acceptable for attitude toward Hong Kong (28%) and attitude toward the advertisement (47%), but only moderate for visit intentions (9%). In terms of mediation results, neither attractiveness (γ = 0.05/t-value = 0.78) nor believability (γ = 0.07 /t-value = −1.05) had a statistically significant direct effect on visit intentions. The majority of the paths and their directions showed significant indirect effects, which was a reasonable indication of mediation effects (Baron and Kenny 1986). Hair et al. (2006) suggest that when a structural model provides a reasonable fit the presence of mediating constructs is supported.

The purpose of the contrast study was to provide clues on the extent to which celebrity-endorsed advertisements differed from nonendorsed advertisements and explore the differences in terms of destination match-up between native and nonnative celebrity-endorsed advertisements. The multigroup modeling function in LISREL was used to test for variance across the groups. The missing values for each group were imputed using the expectation maximization algorithm. The four groups, Andy Lau (n = 206), David Beckham (n = 204), Britney Spears (n = 205), and Maggie Cheung (n = 211), were respectively compared to the control group (n = 218) on the three exogenous dimensions: attitude toward the advertisement, attitude toward Hong Kong, and visit intentions. The respondents in the control group were shown the same advertisement without a celebrity endorser. The global fit indices indicate a reasonable fit (χ² = 1829.03, df = 780, p < 0.01; CFI = 0.96; NFI = 0.93; RMSEA = 0.08). Taking into account the multigroup model, the alternative fit indices are adequate, while the RMSEA value of 0.8 can be considered mediocre (MacCallum, Browne, and Sugawara 1996).

For the purposes of our study, group effects are of interest because they indicate whether celebrities have any effect on the three dimensions of interest. When compared to the control group, celebrity-endorsed advertisements performed generally better and effects were in the proposed direction (see Table 5). The majority of the latent mean scores were minus, which indicates a score closer to one which is positive. Even though the mean vectors of independent variable scores were in the hypothesized direction, none of the values for visit intentions were significantly different from the control group. The latent means for attitude toward the advertisement for the Andy Lau (–0.61) and Maggie Cheung (–0.48) groups were significantly lower than the control group, indicating that they performed better. However, the group for David Beckham (0.16) and Britney Spears (0.81) achieved higher mean scores, with the latter group significantly higher (i.e., performed worse than the control group). With regard to attitude toward Hong Kong, all treatment groups performed significantly better than the control group, except the Britney Spears (0.03) group.

Table 5.

Latent Mean Differences across Groups.

Attitude toward the Advertisement				Attitude toward Hong Kong			Visit Intentions
	M	SE		M	SE		M	SE
Britney	0.81	0.13	*	0.03	0.09	NS	−0.07	.19	NS
David	0.16	0.14	NS	−0.32	0.09	*	−0.28	.20	NS
Maggie	−0.48	0.11	*	−0.29	0.08	*	−0.20	.19	NS
Andy	−0.61	0.12	*	−0.46	0.08	*	−0.31	.20	NS

Note: Control group is constant. M = Mean Difference; SE = Standard Error; NS = Nonsignificant.

p < 0.01.

For both attitude dimensions, the native celebrity groups indicated significantly lower latent mean scores than the control group. Such an occurrence is extremely unlikely due to sampling error alone (Bray and Maxwell 1985). These results correspond with previous studies in that celebrity-endorsed advertisements generate positive responses (Atkin and Block 1983; Tom et al. 1992; Tripp, Jensen, and Carlson 1994). The Britney Spears group evidenced higher mean scores than the control group for both attitude constructs. It is likely that the poor performance was due to negative tabloid news coverage surrounding her personal life and career during the course of the study. Although her presence in the advertisement negatively affected its performance, this result clearly demonstrates that her apparent endorsement had a certain impact.

To measure the association between a celebrity endorser and the advertised destination, a match-up dimension was included in the structured means analysis with the following indicators: related, representative, combined, relative, effective, and consistent. The results are presented in Tables 6 and 7.

Table 6.

Latent Mean Differences for Attractiveness, Believability, and Match-up.

Attractiveness			Believability		Match-up
	M	SE	M	SE	M	SE
Andy (constant)
Britney	1.03	0.12*	2.18	0.11*	2.88	0.13*
David	0.10	0.10NS	1.19	0.11*	1.99	0.14*
Maggie	0.16	0.10NS	0.35	0.07*	0.37	0.08*
Britney (constant)
Andy	−1.03	0.12*	−2.18	0.11*	−2.88	0.13*
David	−0.93	0.12*	−0.99	0.13*	−0.88	0.17*
Maggie	−0.87	0.11*	−1.82	0.11*	−2.51	0.13*
David (Constant)
Andy	−0.10	0.10NS	−1.19	0.11*	−1.99	0.14*
Britney	0.93	0.12*	0.99	0.13*	0.88	0.17*
Maggie	0.06	0.10NS	−0.84	0.11*	−1.62	0.15*
Maggie (constant)
Andy	−0.16	0.10NS	−0.35	0.07*	−0.37	0.08*
Britney	0.87	0.11*	1.82	0.11*	2.51	0.13*
David	−0.06	0.11NS	0.84	0.11*	1.62	0.15*

Note: M = Mean Difference; SE = Standard Error; NS = Nonsignificant.

p < 0.01.

Table 7.

Latent Mean Differences for Attitude toward the Advertisement, Attitude toward Hong Kong, and Visit Intentions.

	Attitude toward the Advertisement		Attitude toward Hong Kong		Visit Intentions
	M	SE	M	SE	M	SE
Andy (constant)
Britney	1.43	0.13*	0.49	0.09*	0.25	0.13NS
David	0.78	0.14*	0.13	0.08NS	0.02	0.14NS
Maggie	0.13	0.10NS	0.17	0.08**	0.12	0.08NS
Britney (constant)
Andy	−1.43	0.13*	−0.49	0.19*	−0.25	0.19NS
David	−0.65	0.15*	−0.35	0.19*	−0.23	0.19NS
Maggie	−1.30	0.12*	−0.32	0.18*	−0.13	0.18NS
David (constant)
Andy	−0.78	0.14*	−0.13	0.08NS	−0.02	0.20NS
Britney	0.65	0.15*	0.35	0.09*	0.23	0.19NS
Maggie	−0.64	0.13*	0.03	0.09NS	0.10	0.19NS
Maggie (constant)
Andy	−0.13	0.10NS	−0.17	0.08**	−0.12	0.19NS
Britney	1.30	0.12*	0.32	0.09*	0.13	0.18NS
David	0.64	0.13*	0.03	0.09NS	−0.10	0.19NS

Note: M = Mean Difference; SE = Standard Error; NS = Nonsignificant.

p < 0.01; **p < 0.05.

The Britney Spears group reported a significantly higher latent mean score on attractiveness than the other endorsers. Therefore, she was perceived to be significantly less attractive than the other three celebrities. However, there were no statistically significant differences in perceived attractiveness among the other three celebrities. Both David Beckham and Britney Spears groups reported significantly higher mean scores on believability than the Andy Lau (1.19/2.18) or Maggie Cheung (0.84/1.82) groups. This indicates that the two native celebrity endorsers (Andy Lau and Maggie Cheung), when compared to the two nonnatives, were perceived to be more believable because of an increased expertise regarding Hong Kong. The same result was present in the match-up dimension. The two native celebrity endorsers were perceived to be more associated with Hong Kong than the two nonnatives. In a similar vein, the results showed that the native celebrity endorsers significantly outperformed the two nonnative celebrity endorsers at influencing the attitude dimensions. As previously indicated, the celebrity endorser did not influence visit intentions any more positively than the control group, and no statistically significant differences were found between the treatment groups. For these data, Britney Spears was the least appropriate celebrity endorser for Hong Kong, as the participants assigned to that group reported significantly higher mean values on the majority of dimensions, whereas Andy Lau was the most appropriate celebrity endorser in terms of his perceived believability and match-up to Hong Kong. Taken together, the results illustrate that when targeting Chinese tourists, native celebrity endorsers are more appropriate spokespeople for Hong Kong than nonnatives.

Discussion

The results from the current study confirm that prior effects regarding a travel destination are applicable and useful for prediction extending beyond a student population and laboratory setting to a Chinese adult consumer population. Broadening the context in which a conceptual framework is examined increases the confidence in proposed impacts and offers the opportunity for the assessment of other mediator or moderator variables. However, not all hypotheses were supported. The path from attractiveness to attitude toward the advertisement was not statistically significant. We examined certain studies that were similarly unable to detect any significant effect of the endorser’s attractiveness to try to explain this result (Caballero and Solomon 1984; Lord and Putrevu 2009). It seems that products employed in such studies were not directly related to the user’s attractiveness, which may explain their findings. Kamins (1990) argues that attractive celebrities are more effective endorsers of products that are used to enhance one’s own attractiveness. Thus, celebrity attractiveness has a minimal impact on the evaluation of products and advertisements when the product is not related to the user’s physical attractiveness (Baker and Churchill 1977; Kahle and Homer 1985). This may explain why the current study found a nonsignificant relationship between the two constructs: a destination or celebrity endorser would not directly enhance a tourist’s physical attractiveness. The perceived attractiveness of the celebrity has a significant impact on attitude toward Hong Kong, which concurs with the findings of previous studies (Baker and Churchill 1977). If celebrities are able to generate positive feelings in consumers, then it is likely that this will lead to a favorable response toward the brand being endorsed (Atkin and Block 1983; Till, Priluck, and Stanley 2008). The proposed relationship between believability to attitude toward Hong Kong was rejected. It appears that the celebrity-endorsed destination advertisement did not have a significant impact on respondents’ existing attitudes toward Hong Kong.

The current study empirically confirms the positive impact believability has on attitude toward the advertisement, and this result is consistent with previous studies (Han and Ki 2010; Magnini, Honeycutt, and Cross 2008). From the mediation results and model fit, it can be assumed that attitude is one of the transporters of information for the subjective evaluation of behavioral intentions, and may thus be a necessary condition in celebrity endorsement frameworks. However, there appears to be a gap between the relative explanatory power of attitude dimensions and visit intentions. The contrast results show similar findings in that celebrity endorsement may not necessarily be better than its absence in influencing people’s visit intentions. This finding concurs with Sanbonmatsu and Kardes (1988), who found that respondents’ willingness to purchase did not differ between those exposed to an advertisement with a celebrity and those exposed to an advertisement featuring no celebrity. This outcome is probably because behavioral intentions require consumers to exhibit a higher level of commitment than attitude formation (Peterson and Jolibert 1995). Tourists are relatively unsure of their intentions to visit a destination when asked some months beforehand for various reasons (Sirakaya, McLellan, and Uysal 1996). The proposed model helps destination marketers predict tourists’ attitudes toward the advertisement and destination—and to a lesser extent their visit intentions—for a celebrity-endorsed print destination advertisement.

In terms of managerial implications, various members of a distinct market segment could be identified and compared according to similarities of responses to the endorsement claim. The results show that native celebrity endorsers, as compared to nonnatives, perform better in generating a positive attitude response. This is mainly due to their positive ratings of perceived believability and match-up with the destination. This is certainly not unexpected, corresponding with pervious research indicating that satisfactory advertising effectiveness exists when congruence between the characteristics of the endorser and the endorsed brand is present (e.g., Clark, Cornwell, and Pruitt 2009; Ding, Molchanov, and Stork 2011; Magnini, Garcia and Honeycutt 2010; Misra and Beatty 1990). For example, Choi, Lee, and Kim (2005) show that domestic celebrities, when compared to celebrities with international recognition, are more effective in delivering messages consistent with the prevalent cultural values in a given country. This is perhaps because people are likely to identify themselves with the domestic celebrities. In other words, the native celebrities (Andy Lau and Maggie Cheung) illustrate personal relevance toward the destination endorsed. Tesser and Martin (1996) state that in order to get the best effect a certain degree of “similarity” between the image of the product and the image of the celebrity should be considered. The more relevant the association is, the more positive the person’s response will be.

Conclusion

Providing insights into alternative ways to promoting destinations is important because unlike consumer goods, tourism marketers have limited funding (Pike, Murdy, and Lings 2011) and a restricted ability (Litvin and Mouri 2009) to modify the destination image. Because of these limitations, it is important that destinations highlight a distinctive feature—a celebrity—that the market can readily identify as belonging to and representing the destination (Litvin and Mouri 2009). For example, the relationship between the native-born celebrity and destination cannot be duplicated, creating a unique selling proposition that other destinations cannot offer (Kwon and Vogt 2010). The potential of celebrity-endorsed print destination advertising is that it can serve as a differentiating tool for destinations (Pike et al. 2011) or the celebrity can be a symbol to create awareness of the destination (Campelo, Aitken, and Gnoth 2011). The results concur with Kwon and Vogt (2010) in that the positive affective reactions to the celebrity-endorsed destination advertisements are more likely to evaluate the destination favorably. Results further indicate that the proposed method constitutes a comprehensive approach for selecting a celebrity for endorsement purposes and helps in terms of evaluating if and how the marketing works to affect tourist behavior (Pratt et al. 2010).

The proposed evaluative dimensions combined with the testing method allow for comparisons between celebrity endorsers to assist destination marketers in selecting an appropriate celebrity endorser before a campaign is launched. This way, one can determine which celebrity endorser performs better and in which areas. Destination marketers could integrate these constructs into their market analysis to predict whether their potential visitors have positive or negative attitudes towards celebrity-endorsed print destination advertisements. The resulting information could then provide concrete and practical insights that may help in selecting an appropriate celebrity endorser to target specific markets.

Our study featured a celebrity-endorsed print advertisement that served mainly academic purposes. Given that we only used one experimental stimulus, we must be cautious in drawing inferences about the impact on people’s processing set. Other forms of media (e.g., videos would allow for more interaction) and the amount of time spent or involvement may have a different impact on people’s responses to a campaign. However, supporting our assumptions, the treatment did influence the way participants responded to the advertisement.

A longitudinal evaluation of a full celebrity endorsement campaign would provide a more comprehensive view of the impact. The possible influence of external factors on the findings cannot be ruled out in the current study. Finally, the quotas did not match perfectly, and the findings of the current study cannot be generalized to respondents with different demographic characteristics or to the population in general. Nevertheless, our results confirm that the effects regarding travel destinations are applicable and useful for prediction beyond a student population and laboratory setting to the Chinese adult consumer population. Broadening the context in which a framework is examined increases the confidence in the proposed impacts and offers the opportunity for assessment of other mediator or moderator variables.

Our focus on celebrity endorsers and the outcome variables may be obscured by other psychological variables. The predictive power of the celebrity endorser factors might have been less if other variables had been included in the study. Marketers should thus be cautious in being overly dependent on one stimulus to influence the decision making of tourists. It is misguided to think that appointing a high-profile and popular celebrity endorser is the best way to reach a certain target market. Without doubt, celebrity endorsements can be effective, but it is necessary to ensure that the right message is being conveyed. The framework presented here is certainly not conclusive, and subsequent research would benefit from testing additional relationships in terms of level of involvement or impact on brand equity and/or brand recall. Hence, further research is required to examine the effects of tourist attitudes on the decision-making process, along with other behavioral stimuli and determinants, to evaluate celebrity-endorsed destination campaigns

Footnotes

Acknowledgements

The authors would like to acknowledge the financial support of the Hong Kong Polytechnic University (Grant No. 1-ZVA5).

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Financial support was provided by the Hong Kong Polytechnic University. Grant number: 1-ZVA5.

Author Biographies

Robert van der Veen, PhD, is a research fellow in the Australian Centre for Asian Business in the International Graduate School of Business at the University of South Australia. His research interests include destination marketing, celebrity endorsement and tourist satisfaction.

Haiyan Song, PhD, is chair professor of tourism in the School of Hotel and Tourism Management at The Hong Kong Polytechnic University. His research interests include tourism forecasting, tourist satisfaction and tourism supply chain management.

References

Agrawal

Kamakura

W. A.

(1995). “The Economic Worth of Celebrity Endorsers: An Event Study Analysis.” Journal of Marketing, 59 (3): 56-63.

Ajzen

Fishbein

(1980). Understanding Attitudes and Predicting Social Behavior. Englewood Cliffs, NJ: Prentice Hall.

Alba

J. W.

Marmorstein

(1987). “The Effects of Frequency Knowledge on Consumer Decision Making.” Journal of Consumer Research, 14 (1): 14-25.

Andrews

D. L.

Jackson

S. J.

(2001). Sports Stars: The Cultural Politics of Sporting Celebrity. New York: Routledge.

Amos

Holmes

Strutton

(2008). “Exploring the Relationship between Celebrity Endorser Effects and Advertising Effectiveness: A Quantitative Synthesis of Effect Size.” International Journal of Advertising, 27 (2): 209-34.

Armitage

C. J.

Conner

(2000). “Efficacy of the Theory of Planned Behavior: A Meta-analytic Review.” British Journal of Social Psychology, 40 (4): 471-99.

Atkin

Block

(1983). “Effectiveness of Celebrity Endorsers.” Journal of Advertising Research, 23 (1): 57-61.

Bagozzi

R. P.

Gurhan-Canli

Priester

(2002). The Social Psychology of Consumer Behaviour. Milton Keynes: Open University Press.

Bagozzi

R. P.

(1988). “On the Evaluation of Structural Equation Models.” Journal of the Academy of Marketing Science, 16 (1): 74-94.

10.

Baker

M. J.

Churchill

G. A.

Jr (1977). “The Impact of Physically Attractive Models on Advertising Evaluations.” Journal of Marketing Research, 14 (4): 538-55.

11.

Baron

R. M.

Kenny

D. A.

(1986). “The Moderator-Mediator Variable Distinction in Social Psychological Research: Conceptual, Strategic, and Statistical Considerations.” Journal of Personality and Social Psychology, 51 (6): 1173-82.

12.

Bentler

P. M.

Bonett

D. G.

(1980). “Significance Tests and Goodness of Fit in the Analysis of Covariance Structures.” Psychological Bulletin, 88 (3): 588-606.

13.

Biehal

Stephens

Curlo

(1992). “Attitude toward the Ad and Brand Choice.” Journal of Advertising, 21 (3): 19-36.

14.

Bollen

Long

(1993). Testing Structural Equation Models. New York: Sage.

15.

Braudy

(1997). The Frenzy of Renown. New York: Vintage Books.

16.

Bray

J. H.

Maxwell

S. E.

(1985). Multivariate Analysis of Variance. Beverly Hills, CA: Sage.

17.

Brown

S. P.

Stayman

D. M.

(1992). “Antecedents and Consequences of Attitude toward the Ad: A Meta-Analysis.” Journal of Consumer Research, 19 (1): 34-51.

18.

Burton

Lichtenstein

D. R.

(1988). “The Effects of Ad Claims and Ad Context on Attitude toward the Advertisement.” Journal of Advertising, 17 (1): 3-11.

19.

Caballero

Solomon

P. J.

(1984). “Effects of Model Attractiveness on Sales Response.” Journal of Advertising, 13 (1): 17-23.

20.

Campelo

Aitken

Gnoth

(2011). “Visual Rhetoric and Ethics in Marketing of a Destination.” Journal of Travel Research, 50 (1): 3-14.

21.

Chang

Wall

Tsai

C. T.

(2005). “Endorsement Advertising in Aboriginal Tourism: An Experiment in Taiwan.” International Journal of Tourism Research, 7 (6): 347-56.

22.

Chau

(1997). “Re-Examining a Model for Evaluating Information Center Success Using a Structural Equation Modeling Approach.” Decision Sciences, 28 (2): 309-34.

23.

Choi

S. M.

Lee

W. N.

Kim

(2005). “Lessons from the Rich and Famous: A Cross-Cultural Comparison of Celebrity Endorsement in Advertising.” Journal of Advertising, 34 (2): 85-98.

24.

Chou

C. P.

Bentler

P. M.

(1995). “Estimates and Tests in Structural Equation Modeling.” In Structural Equation Modeling: Concepts, Issues, and Applications, edited by Hoyle

R. H.

Thousand Oaks, CA: Sage, pp. 37-55.

25.

Clark

J. M.

Cornwell

T. B.

Pruitt

S. W.

(2009). “The Impact of Title Event Sponsorship Announcements on Shareholder Wealth.” Marketing Letters, 20 (2): 169-82.

26.

Court

Lupton

R. A.

(1997). “Customer Portfolio Development: Modeling Destination Adopters, Inactives, and Rejecters.” Journal of Travel Research, 36 (1): 35-43.

27.

Craig

C. S.

McCann

J. M.

(1978). “Assessing Communication Effects of Energy Conservation.” Journal of Consumer Research, 5 (2): 82-88.

28.

Cudeck

Henley

S. J.

(1991). “Model Selection in Covariance Structures Analysis and the “Problem” of Sample Size: A Clarification.” Psychological Bulletin, 109 (3):520-23.

29.

Daneshvary

Schwer

(2000). “The Association Endorsement and Consumers’ Intention to Purchase.” Journal of Consumer Marketing, 17 (3): 203-11.

30.

Dean

D. H.

(1999). “Brand Endorsement, Popularity, and Event Sponsorship as Advertising Cues Affecting Consumer Pre-Purchase Attitudes.” Journal of Advertising, 28 (3): 2-12.

31.

De Mooij

(1997). Global Marketing and Advertising. Thousand Oaks, CA: Sage.

32.

Ding

Molchanov

Stork

(2011). “The Value of Celebrity Endorsements: A Stock Market Perspective.” Marketing Letters, 22 (2): 147-63.

33.

Eagly

Chaiken

(1993). The Psychology of Attitudes. Fort Worth, TX: Harcourt Brace Jovanovich.

34.

Erdogan

B. Z.

(1999). “Celebrity Endorsement: A Literature Review.” Journal of Marketing Management, 15 (4): 291-314.

35.

Gardner

M. P.

(1985). “Mood States and Consumer Behavior: A Critical Review.” Journal of Consumer Research, 12 (3): 281-300.

36.

Gartner

W. C.

(1993). “Image Formation Process.” Journal of Travel and Tourism Marketing, 2 (2/3): 191-215.

37.

Glover

(2009). “Celebrity Endorsement in Tourism Advertising: Effects on Destination Image.” Journal of Hospitality and Tourism Management, 16 (1): 16-23.

38.

Golob

T. F.

(2003). “Structural Equation Modeling for Travel Behavior Research.” Transportation Research Part B: Methodological, 37 (1): 1-25.

39.

Gorn

G. J.

(1982). “The Effects of Music in Advertising on Choice Behavior: A Classical Conditioning Approach.” Journal of Marketing, 46 (1): 94-101.

40.

Hair

J. F.

Jr Anderson

R. E.

Tatham

R. L.

Black

W. C.

(2006). Multivariate Data Analysis, 6th edition. Upper Saddle River, NJ: Prentice Hall.

41.

Han

(2010). “Developing a Measure of Celebrity Reputation.” Public Relations Review, 36 (2): 199-201.

42.

Higham

J. E. S.

Hall

C. M.

(2003). “Sport Tourism in Australia and New Zealand: Responding to a Dynamic Interface.” Journal of Sport Tourism, 8 (3): 131-43.

43.

Homer

P. M.

(1990). “The Mediating Role of Attitude toward the Ad: Some Additional Evidence.” Journal of Marketing Research, 27 (1): 78-86.

44.

Hovland

C. I.

Janis

Kelly

(1953). Communication and Persuasion. New Haven, CT: Yale University Press.

45.

Hsu

McDonald

(2002). “An Examination on Multiple Celebrity Endorsers in Advertising.” Journal of Product and Brand Management, 11 (1): 19-29.

46.

L. T.

Bentler

P. M.

(1999). “Cut-off Criteria for Fit Indices in Covariance Structure Analysis: Conventional Criteria versus New Alternatives.” Structural Equation Modeling, 6 (1): 1-55.

47.

Huang

Hsu

C. H. C.

(2005). “Mainland Chinese Residents’ Perceptions and Motivations of Visiting Hong Kong: Evidence from Focus Group Interviews.” Asia Pacific Journal of Tourism Research, 10 (2): 191-206.

48.

Hunt

S. D.

(1993). “Objectivity in Marketing Theory and Research.” Journal of Marketing, 57 (2): 76-91.

49.

Hutchinson

J. W.

Moore

D. L.

(1984). “Issues Surrounding the Examination of Delay Effects in Advertising.” Advances in Consumer Research, 11 (1): 650-55.

50.

Janiszewski

(1988). “Preconscious Processing Effect: The Independence of Attitude Formation and Conscious Thought.” Journal of Consumer Research, 15 (2): 199-209.

51.

Johnson

R. S.

Harrington

(1998). “The Jordan Effect.” Fortune, 137 (12): 124-32.

52.

Kahle

L. R.

Homer

P. M.

(1985). “Physical Attractiveness of the Celebrity Endorser: A Social Adaptation Perspective.” Journal of Consumer Research, 11 (4): 954-61.

53.

Kamins

M. A.

(1990). “An Investigation into the ‘Match-Up’ Hypothesis in Celebrity Advertising: When Beauty May Be Only Skin Deep.” Journal of Advertising, 19 (1): 4-13.

54.

Kaplan

(2000). Structural Equation Modeling: Foundations and Extensions. London: Sage.

55.

Kline

R. B.

(2005). Principles and Practice of Structural Equation Modeling. 2nd edition. New York: Guilford.

56.

Knott

C. L.

St. James

(2004). “An Alternate Approach to Developing a Total Celebrity Endorser Rating Model Using the Analytic Hierarchy Process.” International Transactions in Operational Research, 11 (1): 87-95.

57.

Kwon

Vogt

C. A.

(2010). “Identifying the Role of Cognitive, Affective, and Behavioural Components in Understanding Residents’ Attitudes toward Place Marketing.” Journal of Travel Research, 49 (4): 423-35.

58.

Lafferty

B. A.

Goldsmith

R. E.

(1999). “Corporate Credibility’s Role in Consumers’ Attitudes and Purchase Intentions When a High versus a Low Credibility Endorser Is Used in the Ad.” Journal of Business Research, 44 (2): 109-16.

59.

Lam

Hsu

C. H. C.

(2006). “Predicting Behavioral Intention of Choosing a Travel Destination.” Tourism Management, 27 (4): 589-99.

60.

Lee

T. H.

(2009). “A Structural Model to Examine How Destination Image, Attitude, and Motivation Affect the Future Behavior of Tourists.” Leisure Sciences, 31: 215-36.

61.

Lee

Scott

Kim

(2008). “Celebrity Fan Involvement and Destination Perceptions.” Annals of Tourism Research, 35 (3): 809-32.

62.

Litvin

S. W.

Mouri

(2009). “A Comparative Study of the Use of Iconic Versus Generic Advertising Images for Destinations Marketing.” Journal of Travel Research, 48 (2): 152-61.

63.

Lord

Putrevu

(2009). “Informational and Transformational Responses to Celebrity Endorsement.” Journal of Current Issues and Research in Advertising, 31 (1): 1-13.

64.

Maccallum

R. C.

Browne

M. W.

Sugawara

H. M.

(1996). “Power Analysis and Determination of Sample Size for Covariance Structure Modeling.” Psychological Methods, 1 (2): 130-49.

65.

Maccallum

R. C.

Rosnowski

Necowitz

L. B.

(1992). “Model Modifications in Covariance Structure Analysis: The Problem of Capitalization on Chance. Psychological Bulletin, 111 (3): 490-504.

66.

MacKenzie

S. B.

Lutz

R. J.

(1989). “An Empirical Examination of the Structural Antecedents of Attitude toward the Ad in an Advertising Pretesting Context.” Journal of Marketing, 53 (2): 48-65.

67.

Maddux

J. E.

Rogers

R. W.

(1980). “Effects of Source Expertness, Physical Attractiveness, and Supporting Arguments on Persuasion: A Case of Brains over Beauty.” Journal of Personality and Social Psychology, 39 (2): 235-44.

68.

Magnini

Garcia

Honeycutt

(2010). “Identifying the Attributes of an Effective Restaurant Chain Endorser.” Cornell Hospitality Quarterly, 51 (2): 238-50.

69.

Magnini

Honeycutt

Cross

(2008). “Understanding the Use of Celebrity Endorsers for Hospitality Firms.” Journal of Vacation Marketing, 14 (1): 57-69.

70.

Mayo

E. J.

Jr Jarvis

L. P.

(1981). The Psychology of Leisure Travel: Effective Marketing and Selling of Travel Services. Boston: CBI.

71.

McCracken

(1989). “Who Is the Celebrity Endorser? Cultural Foundations of the Endorsement Process.” Journal of Consumer Research, 16 (3): 310-20.

72.

McGorry

S. Y.

(2000). “Measurement in a Cross-Cultural Environment: Survey Translation Issues.” Qualitative Market Research, 3 (2): 74-81.

73.

Mechinda

Serirat

Gulid

(2009). “An Examination of Tourists’ Attitudinal and Behavioral Loyalty: Comparison between Domestic and International Tourists.” Journal of Vacation Marketing, 15 (2): 129-48.

74.

Milman

Pizam

(1995). “The Role of Awareness and Familiarity with a Destination: The Central Florida Case.” Journal of Travel Research, 33 (3): 21-27.

75.

Miniard

P. W.

Bhatla

Rose

R. L.

(1990). “On the formation and Relationship of Ad and Brand Attitudes: An Experimental and Causal Analysis.” Journal of Marketing Research, 27 (3): 290-303.

76.

Misra

Beatty

S. E.

(1990). “Celebrity Spokesperson and Brand Congruence: An Assessment of Recall and Affect.” Journal of Business Research, 21 (2): 159-73.

77.

Mohsin

(2005). “Tourist Attitudes and Destination Marketing: The Case of Australia’s Northern Territory and Malaysia.” Tourism Management, 26 (5): 723-33.

78.

Morris

J. D.

Boone

M. A.

(1998). “The Effects of Music on Emotional Response, Brand Attitude, and Purchase intent in an Emotional Advertising Condition.” Advances in Consumer Research, 25: 518-26.

79.

Ohanian

(1990). “Construction and Validation of a Scale to Measure Celebrity Endorsers’ Perceived Expertise, Trustworthiness, and Attractiveness.” Journal of Advertising, 19 (3): 39-52.

80.

Ohanian

(1991). “The Impact of Celebrity Spokespersons’ Perceived Image on Consumers’ Intention to Purchase.” Journal of Advertising Research, 31 (1): 46-54.

81.

Opperman

(1999). “Predicting Destination Choice: A Discussion of Destination Loyalty.” Journal of Vacation Marketing, 5 (1): 51-65.

82.

Patzer

G. L.

(1983). “Source Credibility as a Function of Communicator Physical Attractiveness.” Journal of Business Research, 11 (2): 229-41.

83.

Peterson

R. A.

Jolibert

A. J. P.

(1995). “A Meta-Analysis of Country-of-Origin Effects.” Journal of International Business Studies, 26 (4): 883-900.

84.

Petroshius

S. M.

Crocker

K. E.

(1989). “An Empirical Analysis of Spokesperson Characteristics on Advertisement and Product Evaluations.” Journal of the Academy of Marketing Science, 17 (3): 217-25.

85.

Petty

R. E.

Cacioppo

J. T.

(1981). Attitudes and Persuasion: Classic and Contemporary Approaches. Dubuque, IA: Brown.

86.

Petty

R. E.

Cacioppo

J. T.

(1986). “The Elaboration Likelihood Model of Persuasion.” In Advances in Experimental Social Psychology, Vol. 19, edited by Berkovitz

New York: Academic Press, pp. 123-205.

87.

Pike

(2006). “Destination Decision Sets: A Longitudinal Comparison of Stated Destination Preferences and Actual Travel.” Journal of Vacation Marketing, 12 (4): 319-28.

88.

Pike

S. D.

Murdy

Lings

(2011). “Visitor Relationship Orientation of Destination Marketing Organisations.” Journal of Travel Research, 50 (4): 443-53.

89.

Pornpitakpan

(2003). “Validation of the Celebrity Endorsers’ Credibility Scale: Evidence from Asians.” Journal of Marketing Management, 19 (1/2): 179-95.

90.

Pratt

McCabe

Cortes-Jiminez

Blake

(2010). “Measuring the Effectiveness of Destination Marketing Campaigns: Comparative Analysis of Conversion Studies.” Journal of Travel Research, 49 (2): 179-90.

91.

Quellette

J. A.

Wood

(1998). “Habit and Intention in Everyday Life: The Multiple Processes by Which Past Behavior Predicts Future Behavior.” Psychological Bulletin, 124 (1): 54-74.

92.

Reisinger

Mavondo

F. T.

(2006). “Structural Equation Modeling: Critical Issues and New Developments.” Journal of Travel and Tourism Marketing, 21 (4): 41-71.

93.

Sheppard

B. H.

Hartwick

Warshaw

P. R.

(1988). “The Theory of Reasoned Action: A Meta-Analysis of Past Research with Recommendations for Modifications and Future Research.” Journal of Consumer Research, 15 (3): 325-43.

94.

Schumacker

Lomax

(2004). A Beginner’s Guide to Structural Equation Modeling. Mahwah, NJ: Lawrence Erlbaum.

95.

Shimp

T. A.

(1981). “Attitude toward the Ad as a Mediator of Consumer Brand Choice.” Journal of Advertising, 10 (2): 9-15.

96.

Silvia

E. S. M.

Maccallum

R. C.

(1988). “Some Factors Affecting the Success of Specification Searches in Covariance Structural Modeling.” Multivariate Behavioral Research, 23 (3): 297-326.

97.

Sirakaya

McLellan

R. W.

Uysal

(1996). “Modeling Vacation Destination Decisions: A Behavioral Approach.” Journal of Travel and Tourism Marketing, 5 (1/2): 57-75.

98.

Sanbonmatsu

D. M.

Kardes

F. R.

(1988). “The Effects of Physiological Arousal on Information Processing and Persuasion.” Journal of Consumer Research, 15 (3): 379-86.

99.

Sonmez

S. F.

Graefe

A. R.

(1998). “Determining Future Travel Behavior from Past Travel Experience and Perceptions of Risk and Safety.” Journal of Travel Research, 37 (2): 171-77.

100.

Stafford

M. R.

Stafford

T. F.

Day

E. A.

(2002). “Contingency Approach: The Effects of Spokesperson Type and Service Type on Service Advertising Perceptions.” Journal of Advertising, 31 (2): 17-37.

101.

Sternthal

Craig

C. S.

(1973). “Humor in Advertising.” Journal of Marketing, 37 (10): 12-18.

102.

Tesser

Martin

(1996). The Psychology of Evaluation. New York: Guilford.

103.

Thompson

M. S.

Green

S. B.

(2006). “Evaluating Between-Group Differences in Latent Variable Means.” In A Second Course in Structural Equation Modeling, edited by Hancock

G. R.

Mueller

R. O.

Greenwich, CT: Information Age.

104.

Till

B. D.

Busler

(1998). “Matching Products with Endorsers: Attractiveness versus Expertise.” Journal of Consumer Marketing, 15 (6): 576-86.

105.

Till

B. D.

Priluck

Stanley

S. M.

(2008). “Classical Conditioning and Celebrity Endorsers: An Examination of Belongingness and Resistance to Extinction.” Psychology and Marketing, 25 (2): 183-201.

106.

Tom

Clark

Elmer

Grech

Masetti

Jr. Sandhar

(1992). “The Use of Created versus Celebrity Spokespersons in Advertisements.” Journal of Consumer Marketing, 9 (4): 45-51.

107.

Tripp

Jensen

T. D.

Carlson

(1994). “The Effects of Multiple Product Endorsements by Celebrities on Consumers’ Attitudes and Intentions.” Journal of Consumer Research, 20 (4): 535-47.

108.

Crompton

J. L.

(1991). “Development of Pleasure Travel Attitude Dimensions.” Annals of Tourism Research, 18 (3): 374-78.

109.

Van der Veen

(2008). “Analysis of the Implementation of Celebrity Endorsement as a Destination Marketing Instrument.” Journal of Travel and Tourism Marketing, 24 (3): 213-22.

110.

Van der Veen

Song

(2010). “Exploratory Study of the Measurement Scales for the Perceived Image and Advertising Effectiveness of Celebrity Endorsers in a Tourism Context.” Journal of Travel and Tourism Marketing, 27 (5):460-73.

111.

Woodside

A. G.

Lysonski

(1989). “A General Model of Traveler Destination Choice.” Journal of Travel Research, 27 (4): 8-14.

112.

(1990). “Cognitive and Affective Priming Effects of the Context for Print Advertisements.” Journal of Advertising, 19 (2): 40-48.