Back to results list
Show full item record
Please use this identifier to cite or link to this item:
|Title:||Celebrity endorsement effectiveness for print destination advertising||Authors:||Veen, Robert van der||Degree:||Ph.D.||Issue Date:||2009||Abstract:||The aim of this study is to develop a conceptual framework for understanding the effects of celebrity endorsement on the attitude and visitation intentions of tourists. Ohanian's (1991) model is applied and modified to examine the effects of celebrity endorsement in print destination advertisements in the context of Mainland Chinese tourists visiting Hong Kong. In such a context, this study widens the application of the model beyond a single discipline and cultural setting. The model is modified by introducing a continuous moderator 'Matchup' and two mediators, Attitude towards the Advertisement and Attitude towards the Destination. The key research constructs within the framework and effects are Attractiveness, Trustworthiness and Expertise. Combined, these constructs influence Attitude towards the Advertisement and Attitude towards the Destination, which then impacts on Visitation Intentions. Structural equation modeling is the main statistical technique for testing and estimating the relationships. The structural relationships between the dimensions are hypothesized as follows: (1) Attractiveness, Trustworthiness and Expertise have significant and positive direct effects on Attitude towards the Advertisement and Attitude towards Hong Kong. (2) Attitude towards the Advertisement and Attitude towards Hong Kong both have significant and positive direct effects on Visitation Intentions. (3) Attitude dimensions mediate the relationships between the celebrity endorsement factors and visitation intentions. (4) The component 'Matchup' moderates the relationship between credibility factors and attitude constructs. A pre-test amongst a mixed student and teacher sample (n=113) was conducted to specify the moderator 'Matchup'. In addition, they helped identify the picture of Hong Kong and the selection of the four celebrity endorsers (Andy Lau, David Beckham, Britney Spears and Maggie Cheung). This study administers a questionnaire using a number of published items and a 7-point semantic differential scale format. A pilot study (n=246) tested the questionnaire and identified all proposed constructs with the exception of the Trustworthiness dimension. A revised questionnaire was employed for the main survey (n=1044) with residents of Guangzhou at five popular shopping locations. Face-to-face street intercept interviews combined with a quota sampling technique were conducted for both the pilot study and the main survey. A renowned research company collected the data and ensured the sample closely represented the strata set out by the visitor profile of the Hong Kong Tourism Board. The revised measurement model and structural model were found to fit the data satisfactorily. The path coefficients and their statistical significance in the revised structural model indicated that all hypotheses were supported, apart from one. Although the path from Attractiveness to Attitude towards the Advertisement is in the specified positive direction, it is not statistically significant. This study found that tourist's attitude towards the advertisement and the destination were predictors of visitation intentions, and at the same time played a mediating role between the celebrity endorsement dimensions and visitation intentions. 'Matchup' reported significant moderation effects on the relationship between perceived expertise and attitude towards a destination. Squared multiple correlations indicate that the revised structural model has a strong statistical ability to predict Mainland Chinese respondents' Attitude towards the Advertisement (47%) and Attitude towards Hong Kong (26%) and to a lesser extent their Visitation Intentions (9%). The experimental research design is able to determine the effects of the four treatment groups versus the control group. By comparing the latent mean differences, it was found that the experimental groups significantly score better on both attitude dimensions. However, no significant differences were detected for visitation intentions. The study also compared celebrity endorser groups on familiarity, likability and all major dimensions. From the data, it appears that Andy Lau is the most appropriate celebrity for endorsing Hong Kong. Finally, the study findings provide support for the proposed model of celebrity endorsement in a tourism setting among an Asian sample. As a result, considerable insight into tourists' responses to celebrity endorsement for destination advertising is gained. In addition, the study sheds light on the appropriateness of Trustworthiness as a construct to examine source credibility for a celebrity endorser. Instead, it appears that the trustworthiness of the celebrity endorser depends on the media providing information about the celebrity. Attitude was found to be one of the significant antecedents, which may serve as a necessary condition for the consequent intentions within future celebrity endorsement frameworks. In addition, significant interaction effects of the relationship between Expertise and Attitude towards Hong Kong are identified. Consequently, the proposed inclusion of the mediator and moderator changes the current thinking regarding the impact of celebrity endorsement on consumer evaluations. By assessing the value of celebrity endorser assets and their effectiveness, this study may help to inform destination marketers who may independently use the model or at least in part, to form, explain and perhaps justify promotional activities.||Subjects:||Hong Kong Polytechnic University -- Dissertations.
Endorsements in advertising.
Tourism -- Marketing.
Celebrities -- China -- Hong Kong.
|Pages:||ix, 357 p. : ill. ; 31 cm.|
|Appears in Collections:||Thesis|
View full-text via https://theses.lib.polyu.edu.hk/handle/200/1934
Citations as of Jun 4, 2023
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.