MOVIES’ IMPACT ON PLACE IMAGES AND VISITATION INTEREST:
A PRODUCT PLACEMENT PERSPECTIVE
By
Fang Yang

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of requirements
for the degree of
DOCTOR OF PHILOSOPHY
Communication Arts and Sciences-Media and Information Studies
2011

ABSTRACT
MOVIES’ IMPACT ON PLACE IMAGES AND VISITATION INTEREST:
A PRODUCT PLACEMENT PERSPECTIVE
By
Fang Yang
In the field of destination branding, the biggest challenge for marketers is how to
build close emotional ties between the potential visitors and the locations. Among all the
possible media channels to promote destinations, movies are believed to be important
motivators for mass tourism. The emotionally based images from movies can provide
some essential differentiation of places, and help them to compete in a crowded
marketplace. This study explored how entertainment movies, as autonomous image
formation agents, influence viewers’ perceptions of the places portrayed and consequent
visitation interest.
First, this study looked at the embedded places as products placed in the movies. On the
basis of the Adapted Meaning Transfer Model, it explored whether movie genre will have any
influence on people’s place perceptions and visitation interest. On one hand, the study found
that the violent crime movie had a significant negative impact on the viewers’ affective and
cognitive place images immediately after the movie exposure. On the other hand, the results
demonstrated that, contrary to the expectations, the romantic drama did not generate any
significant positive impact on the viewers immediately after the movie exposure.
Secondly, based on the Transportation Theory, this study explored movie
transportation’s role in the relationship between movie watching and tourism. Movie
transportation is defined as the state of immersion into a movie. The results indicated that

movie transportation had a significant impact on the viewers’ affective place images, cognitive
places images, and visitation interest. Particularly, regardless of movie genre, the more the
movie viewers were transported, the more favorable impressions they had for the featured
tourism sites and consequently the more interested they were in traveling to the target place.
Moreover, this study explored which was more powerful: the movie genre’s effect or
the movie transportation’s impact. The findings demonstrated that movie transportation did, to
a large degree, weaken movie genre’s influence. Particularly, for the highly transported
audience, there were no significant differences between the movie groups in terms of their
perceptions of place pleasantness, tourism attraction, and community quality. However,
significant differences were found among the audience that was not well transported. This
suggests that movie transportation is an essential mechanism by which movies can influence
place perceptions and visitation interest. The effect of movie transportation is more powerful
than the effect of movie genre. Dark movies still have the potential to enhance place images if
they can transport the audience well.
In addition, movie transportation’s influence was evaluated when the variance of initial
place familiarity was considered. The results indicated that initial place familiarity generally
had a significant positive impact on the audiences’ cognitive place images and visitation
interest, regardless of movie genre. Moreover, it also found that movie transportation is a
significant moderator of initial place familiarity’s influence on affective place images.
Last but not least, this study made an attempt to explore movies’ long-term impact. The
results showed that movie genres’ main effect, movie transportation’s main effect, and movie
transportation’s moderation effect were all meaningful on a long-term basis.

This dissertation is dedicated to my husband, Zhangwei, and my daughter, Shirley, for their
love, support, and encouragement.

iv

ACKNOWLEDGEMENTS

I would like to acknowledge and thank the following people, without whom I could not have
completed the Ph.D. program:
Dr . Bruce Vanden Bergh, committee chair, and my mentor for four years. He always
strongly stood behind me for financial and mental support. His affection toward teaching and
research always motivated and inspired me.
Dr. Hairong Li, a member of my dissertation committee. He provided exceptional
direction and critical review of my research. Thanks to his vision, I was lucky to have
movie-induced tourism and place branding as my dissertation topic.
Dr. Richard Cole, a member of my dissertation committee. He not only provided
encouragement and direction, but also expanded my understanding of movies’ impact from a
relatively objective point of view, while most of the studies in the literature took a positive
perspective.
Dr. Tom Page, a member of my dissertation committee. He has for years unflaggingly
withstood all my persistent questions and inquiries. My research skills grew thanks to his
lectures, advice, and feedback.
Moreover, I also owe thanks to my professors, Dr. Mira Lee, Dr. Janice Bukovac, Dr.
Franklin Boster, and Dr. Tim Levine, for their assistance with research design, data collection,
and statistical analysis.
Last, but not least, I would like to thank my fellow graduate students, Hyun Ju Jeong,
Jing Zhao, Dan Lake, Caitlin McLaughlin, Suzanna Hill, Jie Li, Christy Lee, and Mike
Friedman who shared their knowledge, support, and friendship.
v

TABLE OF CONTENTS

LIST OF TABLES

viii

LIST OF FIGURES

xi

CHAPTER 1: INTRODUCTION

1

CHAPTER 2: LITERATURE REVIEW
Place Branding
Movie-Induced Tourism
Destination as Product Placed in Movies
Conceptual Frame Work
Relevant Theories
The Transportation Theory
The Adapted Meaning Transfer Model
Relevant Constructs
Movie Transportation
Movie Genre
Place Image and Visitation Interest
Initial Place Familiarity
Hypotheses and Research Questions

5
5
9
11
12
15
15
18
19
19
20
23
26
27

CAHPTER 3: METHOD
Design
Stimuli
Procedure
Measures

33
33
34
37
39

CHAPTER 4: RESULT SUMMARY
Hypothesis 1
Hypothesis 2
Hypothesis 3
Hypothesis 4
Research Question 1
Hypothesis 5
Hypothesis 6
Hypothesis 7
Hypothesis 8
Research Question 2

48
48
56
58
68
69
83
92
93
100
102

CHAPTER 5: DISCUSSION
Findings

135
135
vi

Practical Implications
Limitations and Future Studies

142
144

APPENDIX 1: Pretest Questionnaire 1

148

APPENDIX 2: Posttest Questionnaire 1 for the Experimental Groups
Kill Bill
Lost in Translation

151
151
159

APPENDIX 3: Posttest Questionnaire 1 for the Control Group

167

APPENDIX 4: Posttest Questionnaire 2

173

REFERENCES

178

vii

LIST OF TABLES

Table 1: Respondent Characteristics

33

Table 2: Most Frequently Mentioned Movies and Locations

35

Table 3: Infrequently Mentioned Movies and Locations

35

Table 4: Descriptive Statistics of Transportation Scale Items

40

Table 5: Descriptive Statistics of Initial Place Familiarity Items

41

Table 6: Factor Analysis of Cognitive Place Image Items

43

Table 7: Measures of Affective Place Image

45

Table 8: Descriptive Statistics of Earthquake’s Impact Items

47

Table 9: H1 Test Result for Affective Image-Pleasant/Unpleasant

49

Table 10: H1 Test Result for Affective Image-Sleepy/Arousing

51

Table 11: H1 Test Result for Cognitive Image-Tourism Attraction

53

Table 12: H1 Test Result for Cognitive Image-Community Quality

55

Table 13: H2 Test Result for Visitation Interest

57

Table 14: H3 Test Result for Affective Image-Pleasant/Unpleasant

59

Table 15: H3 Test Result for Affective Image-Sleepy/Arousing

62

Table 16: H3 Test Result for Cognitive Image-Tourism Attraction

65

Table 17: H3 Test Result for Cognitive Image-Community Quality

67

Table 18: H4 Test Result for Visitation Interest

69

Table 19: RQ 1 Result for Affective Place Image-Pleasant/Unpleasant

70

Table 20: RQ 1 Result for Affective Place Image-Sleepy/Arousing

73

viii

Table 21: RQ 1 Result for Cognitive Place Image-Tourism Attraction

76

Table 22: RQ 1 Result for Cognitive Place Image-Community Quality

78

Table 23: RQ 1 Result for Visitation Interest

81

Table 24: H5 Test Result for Affective Image-Pleasant/Unpleasant

84

Table 25: H5 Test Result for Affective Image-Sleepy/Arousing

87

Table 26: H5 Test Result for Cognitive Image-Tourism Attraction

90

Table 27: H5 Test Result for Cognitive Image-Community Quality

91

Table 28: H6 Test Result for Visitation Interest

92

Table 29: H7 Test Result for Affective Image-Pleasant/Unpleasant

94

Table 30: H7 Test Result for Affective Image-Sleepy/Arousing

96

Table 31: H7 Test Result for Cognitive Image-Tourism Attraction

98

Table 32: H7 Test Result for Cognitive Image-Community Quality

99

Table 33: H8 Test Result for Visitation Interest

101

Table 34: Genre’s Long-Term Effect on Affective Image-Pleasant/Unpleasant

103

Table 35: Genre’s Long-Term Effect on Affective Image-Sleepy/Arousing

105

Table 36: Genre’s Long-Term Effect on Cognitive Image-Tourism Attraction

106

Table 37: Genre’s Long-Term Effect on Cognitive Image-Community Quality

108

Table 38: Genre’s Long-Term Effect on Visitation Interest

110

Table 39: Transportation’s Long-Term Effect on Affective Image
-Pleasant/Unpleasant

112

Table 40: Transportation’s Long-Term Effect on Affective Image
-Sleepy/Arousing

115

ix

Table 41: Transportation’s Long-Term Effect on Cognitive Image
-Tourism Attraction

118

Table 42: Transportation’s Long-Term Effect on Cognitive Image
-Community Quality

121

Table 43: Transportation’s Long-Term Effect on Visitation Interest

123

Table 44: Transportation’s Long-Term Moderation Effect on Affective
Image-Pleasant/Unpleasant

126

Table 45: Transportation’s Long-Term Moderation Effect on Cognitive
Image-Tourism Attraction

129

Table 46: Transportation’s Long-Term Moderation Effect on Cognitive
Image-Community Quality

132

x

LIST OF FIGURES

Figure 1: Interaction between Movie Transportation and Initial Place Familiarity
on Affective Place Image-Pleasant/Unpleasant

95

Figure 2: Interaction between Movie Transportation and Initial Place Familiarity
on Affective Place Image-Sleepy/Arousing

97

xi

Introduction
In today’s globalized world, cities, regions, and countries are faced with fierce
competition for resource, foreign investment, and visitors (Kotler et al., 1999). In response
to the demands of the competition, place administrators are interested in applying
marketing theory and practice to place marketing. For example, Mark Leonard, the author
of the book Britain, stated that people will tend to pay more attention to a place that has a
strong image. People will even be willing to pay more for goods and services from a place
that has established a powerful and attractive identity (Crane, 1998). Tony Blair, the former
British Prime Minister, took the idea of “branding” Britain quite seriously and argued that
it is time to reposition Britain as “one of the world’s pioneers rather than one of its
museums” (Crane, 1998).
The challenge that place marketers are faced with is how to make an effective
impression in the competitive world. They have been looking for efficient ways to convince
consumers that their products and services are of leading-edge quality. They also have
attempted to persuade tourists that their places are great destinations to visit. According to
Ashworth and Kavaratzis (2007), place branding should go beyond logos and slogans. In
essence, relevant “stories” need to be built into the places. In the field of destination
branding, the biggest challenge is building close emotional ties between the potential
visitors and the locations, as it is how the places make the customers feel that will
ultimately determine their reputations and values.
Among all the possible media channels to promote destinations, movies are
believed to be important motivators for mass tourism. Schofield (1996) suggested that
1

contemporary tourists’ autonomous images of places are shaped through the vicarious
consumption of movies and television programs without the perceptual bias of
promotional materials. Sue Beeton (2005) stated that movies can contribute to viewers’
fantasies and dreams, which in turn will influence their perceptions of places. This means
that emotionally based images from movies can provide some essential differentiation of
places, and help them to compete in a crowded marketplace.
The movie industry is a huge business all over the world and contributes to the
growth of movie-induced tourism worldwide. According to the Motion Picture Association
of America (MPAA), the global box-office receipts for all films released last year reached
a high of $31.8 billion (Verrier, 2010). Based on the theatrical market statistics report
complied annually by MPAA, ticket sales in the U.S. and Canada reached $10.6 billion
last year. Despite a weak economy, the largest growth occurred in Latin America and the
Asia Pacific region, which grew 25% and 21%, respectively, and accounted for $10.8
billion in box-office revenue (Verrier, 2010).
Thanks to the booming movie market, movie-induced tourism is a rapidly growing
sector of the tourism industry with increasing economic importance. According to Tetley
(1997), filming not only can provide short-term employment and publicity for the chosen
location, but also can create long-term tourism opportunities. For example, The Lord of the
Rings trilogy was filmed in New Zealand and the movie has continuously helped with
visitor arrival in the South Pacific island country. As reported by Marco in The
Washington Times (2003), New Zealand had more than 2 million visitors from November
2001 to November 2002, which doubled international arrivals in the past decade.

2

Moreover, Cardy stated in The Dominion Post (2006) that tourists were still going to New
Zealand because of the movie, even more than 2 years after the last movie in the trilogy
was released. It was found that many visitors would take friends and family to revisit the
featured places several times.
In the literature on movie-induced tourism, a number of studies have found that
movies appear to alter visitation to tourism areas in terms of tourist numbers (Riley & Van
Doren, 1992; Riley, Baker, & Van Doren, 1998; Tooke & Baker, 1996), but there are few
empirical investigations of the ways that movies affect place images and visitation interest.
In other words, although evidence has been found that movies can have a significant
impact on visitation trends at the macro level, it is still unclear how movies influence
viewers’ perceptions of places and visitation interest at the micro level. Kim and
Richardson (2003) conducted a study to explore whether empathy with characters works as
a linkage between movie viewing and visitation interest, but they failed to find the expected
relationship. In this regard, Hudson and Ritchie (2006) claimed that more research is
required into psychological and behavioral aspects of movie-induced tourism.
The objective of this study is to start to fill in the gap in the literature by exploring
how entertainment movies, as autonomous image formation agents, influence viewers’
perceptions of the places portrayed and consequent visitation interest. The study will focus
on elements that have not yet been addressed or have received little attention in the literature.
First, the study will explore whether movie genre will have any influence on people’s place
perceptions and visitation interest. The major questions are: Do feel-good movies improve
the embedded place images and increase people’s visitation interest? Do dark movies tend

3

to attract visitors or drive away visitors?
Secondly, this study will explore movie transportation’s role (i.e., movie
immersion’s role) in the relationship between movie watching and tourism. Based on
Transportation Theory, I propose that movie transportation is a key linkage between movie
viewing, perception changes of depicted places, and visitation interest. Particularly, the
more the viewers can be transported by movies, the more positive place images they will
have, and the more interested they will be in visiting the embedded locations.
Moreover, this study will explore if the movie genre’s effect is more powerful or if
the movie transportation’s impact is more powerful. The major questions are: Will the
viewers who are highly transported by dark movies tend to generate more favorable place
images or more unfavorable place images? Will the viewers who are highly transported by
dark movies tend to be more interested or less interested in visiting the embedded place?
In addition, movie transportation’s influence will also be evaluated when the
variance of initial place familiarity is considered. The questions are: Does initial place
familiarity have a positive linear relationship with place images and visitation interest? Will
movie transportation moderate initial place familiarity’s influence on people’s place
perceptions and visitation interest?
Finally, as movies’ effects over time have not yet been addressed in the literature,
this study will explore whether movies’ impact on tourism is merely an instant effect, or
whether it is something that can last on a long-term basis.

4

Literature review
Place Branding
From the perspective of marketers, places are seen as mega-products (Florek, Insch, &
Gnoth, 2006). According to Zenker and Braun (2010), a place brand is “a network of
associations in the consumers’ mind based on the visual, verbal, and behavioral expression of a
place, which is embodied through the aims, communication, values, and the general culture of
the place’s stakeholders and the overall place design” (p.3). It is believed that places can satisfy
functional, symbolic, and emotional needs just like general product brands (Rainisto, 2003). In
this sense, place branding is regarded as a good starting point for place marketing and it is a
solid framework by which to manage the place’s image (Kotler, Asplund, Rein, & Heider,
1999). The goal of place branding is to maximize the efficient social and economic functioning
of an area and to promote a place’s values and image so that potential users are fully aware of
its distinctive advantages.
In recent years, the branding of places has gained popularity, which is illustrated by
the development of city brand rankings such as the Anholt-GMI City Brand Index (Anholt,
2006). Places increasingly compete with each other in an effort to attract tourists, investors,
companies, and new citizens (Kavaratzis, 2005). As a result, they invest a considerable
amount of taxpayers’ money on their marketing activities. For example, Michigan had a total
tourism marketing budget of $28 million in 2009 (Hampson, 2010), Australia invested $20
million for a three-year nation brand campaign since 2010 (Lee, 2009), and Berlin maintained
a marketing budget of five million euros per year (Jacobsen, 2009).
Place branding is a complex subject and according to Kavaratzis (2005), there are at

5

least three major distinct trends of discussion, including nation branding, city branding, and
destination branding. The first school of studies is nation branding (e.g., Anholt, 2002; Ham,
2001), which deals with the positive branding of the nation to help develop tourism and
attract foreign investment. Simon Anholt coined the phrase nation brand in the 1990s and
when he associated “brand” with places, the metaphor was used to describe how places
compete with each other in the global marketplace for products, services, events, ideas,
visitors, talent, and investment.
According to Anholt (2010), nation branding is more about public policy than
marketing communication. He stated that a policy approach for nation branding will enable
nations to improve the efficiency and effectiveness with which they achieve deserved images.
As he said, there are five key ideas for nation branding. First, places must engage with the
outside world in a clear, coordinated and communicative way if they are to influence public
opinion. Second, the notion of brand image is critical, and reputation is an essential factor
that underpins every transaction between the place brand and the consumers. Thirdly, the
notion of brand equity is important, and place reputation is a huge valuable asset that needs to
be managed. In addition, the notion of brand purpose needs to be clearly defined and the idea
of uniting groups of people around a common strategic vision can create a powerful dynamics
for progress. Finally, if public opinion will be influenced, it is important that sustained and
coherent innovations need to take place in all sectors of the national activities.
Based on the idea of nation brand, place brand as a public diplomacy and marketing
concept has been gradually applied to counties, cities, and regions. The second
steady-growing study trend in place branding is city branding, which explores the

6

possibilities of branding as an approach to integrate and guide place management. According
to Anholt (2007), a city’s international status and standing can be evaluated based on six
image dimensions, including the presence, the place, the potential, the pulse, the people, and
the prerequisites. The dimension of the presence refers to a city’s popularity among the
people and its significance in the contribution to the world in terms of culture, science, and so
on. The dimension of the place is about the people’s perceptions of the physical aspect of a
city, such as outdoor pleasantness, natural beauty, and climate. The dimension of the potential
refers to a city’s capacity to provide economic and educational opportunities for visitors,
businesses, and immigrants. The dimension of the pulse is about how exciting the city is and
how easy it is to find interesting things to do. The dimension of the people refers to the
friendliness of the inhabitants and diversity of the community. Finally, the dimension of the
prerequisites refers to the basic qualities of the city, including affordable accommodations
and public amenities.
As for city branding, Baker (2007) stated that when practices of branding and
marketing are introduced to a place, the very close emotional ties between the people (both
the residence and the visitors) and the places are something unique that the marketers need to
consider. He said that place is an experiential product and it is how the place makes the
customers feel that will ultimately determine its reputation and value. The goal for place
branding is to make an experience of a place as memorable, different, and exceptional as it
can possibly be. In a sense, when applied to branding cities, it is a good idea to rely on an
intangible umbrella brand rather than a tangible umbrella brand. For example, the idea of
“good living” will open opportunities for fine dining, quality entertainment, boutique stories,

7

spas, resorts, and galleries.

If a city merely focuses on the tangible attributes, the

competitors can easily copy, weaken, or match the claims of superiority.
In addition, Ashworth and Kavaratzis (2007) discussed the possibility of applying the
concept of corporate branding and specific methodologies developed in this field to city
branding. According to them, people will understand cities in the same way as brands. It is in
the people’s minds that the city takes form through the processing of perceptions and images.
The process is the same as followed in the formation of images of other entities like products
or corporations, which have long been managed as brands. As they stated, a city brand
resembles a corporate umbrella brand in many ways. For example, like corporate brands, a
place brand’s image needs both the tangible “service” characteristics and the intangible brand
personality. However, from the perspective of brand architecture, they stated that it is not a
good idea to create an umbrella nation brand, under which city brands will be managed. This
is because an umbrella nation brand will be too heterogeneous (i.e. non-brand), too bland (i.e.
appealing to no one in particular), and too skewed (i.e. focusing on certain activities at the
expense of others).
Finally, the most developed trend in the place branding literature is destination
branding. The studies in this area focus on the investigation of the role of branding in the
marketing of tourism destinations (e.g. Morgan, Pritchard, & Pride, 2002). Among the
destination branding studies, the exploration of movie-induced tourism has generated great
interest in recent years. The basic purpose for movie-induced tourism studies is to explore the
image enhancement opportunities that exist through the medium of movies. According to
Urry (1990, p.3), the tourist gaze is “ …constructed and sustained through a variety of

8

non-tourist practices, such as film, TV, literature, magazines, records, and video”. What Urry
advocated was that the image consumers have of a place in today’s world is strongly formed
and influenced by such media forms as movie and television. In the following section, I will
provide a detailed introduction of the movie-induced tourism studies.
Movie-induced tourism
Movie-induced tourism has been defined by Hudson and Ritchie (2006) as “tourist
visits to a destination or attraction as a result of the destination being featured on television,
video, or the cinema screen” (p.317). Movies have received special attention from destination
marketers due to the belief that they can “generate and sustain interest in a destination in a
way which destination marketers cannot afford to do” (Tooke and Baker 1996, p. 87). Since
movies are perceived to be more reliable and trustworthy than biased promotions and
advertisements, they have a better chance at influencing destination images.
A growing body of knowledge shows that movies and television programs can induce
a meaningful increase in the number of visitors to areas which were at the center of movies or
television programs. For example, Riley, Baker and Van Doren (1998) investigated visitation
to 12 U.S. cities depicted in movies, by compiling visitation data 10 years before and 5 years
after movie releases for each location. The results of their study showed that movies
increased visitation to study locations for at least 4 years after their releases. Similarly, Tooke
and Baker (1996) investigated the effect of British television films on the popularity of movie
destinations, manifested by visitor numbers. They analyzed the literature in academic
research, journals, and newspapers regarding four U.K. destinations depicted in four dramas
and claimed that movies can cause an increase in visitor numbers at the movies’ locations.

9

Although popular movies appear to impact visitation to tourism areas, it is still
unclear how movies influence viewers’ perceptions of places and visitation interest and why
the positive impact happens. As Beeton (2005) stated, the promotional capability of movies is
not equal; for example, some movies may have little impact, while others may be both
influential and memorable. In the literature, there are many intuitive explanations for what
movie factors can impact place perceptions and visitation interest. For example, Riley, Baker,
and Van Doren (1998) stated that a movie needs to have an “icon” component to generate
visitor interest. An “icon” implies a movie’s symbolic content, a single event, a favorite
performer, a location’s physical features, or a theme which can represent all that is popular
and compelling about the movie. Riley and Van Doren (1992) suggested that the audience’s
empathetic involvement with movie characters and vicarious experience of a place might be
the linkage between movie viewing and place impression. Hudson and Ritchie (2006)
suggested that a variety of movie-specific factors can influence the visitor’s interest in
destinations, including identifiable and accessible locations, relevance of the story to the
locations, untainted environment, and so on.
A very limited number of empirical studies on how movies influence viewers’
perceptions of places have been conducted so far. For example, Kim and Richardson (2003)
did a study to test the mediating role of character attachment between movie viewing and
place impression, but they failed to find the expected relationship. In addition, Tasci (2009)
explored whether the change in audience’s concept of social distance can relate movie
watching to visitation interest. The results show that movie exposure does change the social
distance concept among the audience along with place impression, but a direct relationship

10

cannot be established.
To better understand the relationship between movies and tourism, the current study
explores the underlying psychological mechanism within individuals. The findings from this
study will offer theoretical insight to understand at the micro level as to how and why movies
could influence the perception of place images and visitation interest.
Destination as product placed in movies
According to Morgan and Prichard (1998), placing a destination in a movie is the
ultimate in tourism product placement. Product placement is an emerging phenomenon, and
has been defined as the planned entries of products into movies or television shows that may
influence viewers’ product beliefs and/or behaviors favorably (Balasubaramanian, 1994).
According to Karrh, McKee, and Pardum (2003), marketers have recently discovered that
communications via product placement can be more sophisticated, more targeted, and more
widely seen than traditional advertising methods. Of the dozens of studies in product
placement in the literature, so far none of them have looked at placement of destinations in
movies and its influence on tourism. However, some of the findings will provide helpful
insights for movie-induced tourism research.
First of all, in most of the studies, respondents have a positive view toward product
placement. A number of research projects have provided strong support for the positive
impact of placement on memory (Babin & Carder, 1996; Gupta & Lord, 1998; Ong& Meri,
1994; Vollmers & Mizerski, 1994). For example, Gupta and Lord (1998) examined the nature
of brand appearances in movies and used a two-dimensional approach to categorize different
types of product placement, including the mode of presentation and level of prominence.

11

They found that people have stronger memories for brands and claims that are placed than
those that are advertised.
Moreover, the product placement literature indicates that product placement can have
a positive impact on the audience’s brand attitude. Russell and Puto (1999) proposed that the
audience’s connectedness to the television show will be positively related to the placed brand
attitude. The concept of connectedness refers to the higher relationship that a viewer develops
with the characters and contextual settings of a program in the parasocial television
environment. Following this logic, Russell and Stern (2006) conducted an empirical study in
sitcom context and found that consumers will align their product attitudes with the characters’
product attitudes and this process is driven by the consumers’ attachment to the characters.
Just as product placement will influence the viewer’s attitude toward a brand, so too
will the movie have an impact on the destination image if the location plays a part in the
movie. Since Russell and Puto (1999) stated that the concept of transformational function of
placement in entertainment programs is important, then movie-evoked experiential feelings
about places should work in a similar way to influence the movie viewer’s perception of
destination images. The current study examines how movies impact place images and
visitation interest from the perspective of product placement.
Conceptual Framework
Exploring how movies impact place images is somewhat similar to trying to explain
how advertisements influence brand images. When a place is embedded in a movie, the
movie can be seen as a transformational ad or a drama ad about the place. As a result,
understanding how advertisements work to improve brand images can provide helpful

12

insights to understand how movies influence place images.
Puto and Wells (1984) proposed that transformational ads are highly affect based.
Transformational ads can make the experience of using the product more exciting and
enjoyable than that obtained solely from an objective description of the advertised brand.
They can connect the experience of the ad so tightly with the experience of using the brand
that consumers cannot remember the brand without recalling the experience generated by ad.
Similarly, Deighton, Romer and McQueen (1989) contended that drama is one of the
important paths for advertising to persuade. In drama advertising, the claim is framed as
subjective, appeals to personal experience, and is not open to objective testing. Effective
drama advertising is found to influence belief by evoking more expression of feelings and
verisimilitude, less counterargument, and less direct elicitation of belief. When drama
advertising is successful, the audience becomes “lost” in the story and experiences the
concerns and feelings of the characters.
In addition, Boller and Olson (1991) have called empathic projection onto advertising
characters the heart of the dramatic advertisement persuasion process. They stated that any
theoretical account of drama ad processing must start by describing how viewers “process”
the ad characters. They also stated that viewers generally process drama ads by building
empathic relationships with the ad characters. Through their empathy with ad characters,
viewers can vicariously experience the personal relevance of the advertised brand.
Although movie-place relationship is analogous to ad-brand relationship, there are
some fundamental differences. First, the primary message in a movie is usually the story and
place relevant information is of secondary importance. In most cases, only when place has

13

something to do with a story can it be followed with interest. Otherwise, place would only be
regarded as a general setting where the story happens. Second, movie persuades in a more
complex way than drama ads do. As Tooke and Baker (1996) mentioned, movies usually
involve three basic influential features, including characters, logic/reasoning, and emotions.
Similarly, Iawashita (2010) proposed that places can be transformed by movies by storyline
themes, exciting sequences, actual physical landscapes, major characters, or any combination
of these. This means that the underlying mechanism that works for drama ads might not be
enough to explain how movies work for place image enhancement. In other words, merely
focusing on empathy with characters might not be adequate to locate the effective linkage
between movie viewing and place image perceptions. The underlying mechanism may
simultaneously involve a number of factors, including empathy with characters, connections
with story plot, identification with specific human relationships, attention to particular
physical settings, and so on. This might be the reason why Kim and Richardson (2003) did
not find empathy with characters a significant mediator between movie viewing and place
perception changes.
In order to have a theoretical understanding of the mechanism under which movies
influence place images and visitation interest, the current study explores whether meaning
transfer and movie transportation are the missing links between movie viewing and place
perception. This study is theoretically based on the Transportation Theory and the Adapted
Meaning Transfer Model. A conceptual framework that describes the relationships between
movie watching and selected relevant constructs is proposed and tested. These constructs are
movie transportation, movie genre, place image (affective place image and cognitive place

14

image), visitation interest, and initial place familiarity.
Relevant Theories
The Transportation Theory
Transportation into a narrative world is a state of immersion into a story (Green &
Brock, 2000). It was proposed as a mechanism whereby narratives can affect beliefs.
Transportation entails imagery, affect, and attention focus. It has three main components,
which are cognitive engagement, emotional engagement, and mental imagery.
Green and Brock (2000) demonstrated how persuasive influence in literature is a
function of transportation and relies on a narrative world with plot and characters. In
particular, they found that transportation could augment story-consistent beliefs and favorable
evaluations of protagonists. Highly transported readers will find fewer inconsistencies in the
story than less-transported readers. Transportation and the corresponding beliefs were
generally unaffected by labeling a story as fact or as fiction.
Moreover, Green and Brock (2004) explored the theoretical linkage between
transportation and media enjoyment and suggested that the experience of being immersed in a
narrative world can create an increase in enjoyment. According to their study, it is believed
that transportation somewhat resembles the experience of flow, which is brought about by
absorption in an activity and is often marked by a deep sense of enjoyment. Since individuals
are often drawn into stories that are frightening, the enjoyment of transportation experience
does not necessarily have to happen in the positive narrative context.
Transportation is conceived as a convergent process (Green & Brock 2000; Green &
Brock, 2002), where all mental systems and capacities become focused on events occurring

15

in the narrative. The first consequence of transportation is that the audience loses access to
real-world facts in favor of accepting the narrative world that the author has created. The
second consequence is that the transported audience may experience strong emotions, even
though they know the events in the story are not real. The third consequence is that people
who return from being transported will be somewhat changed by that experience.
Conceptually, the state of transportation is equivalent to the extent of engagement
with a narrative and is similar to Krugman’s (1965) situational media involvement.
Involvement is an important definition in psychology and has also received attention in
marketing and consumer science. In the literature of advertising and marketing, there are
three major types of involvement concepts, including product involvement, outcome
involvement, and situational media involvement.
The concept of product involvement is based on perceived personal relevance of an
object to an individual (Zaichkowsky, 1985; Zaichkowsky, 1986; Celsi & Olson, 1988). The
definition illustrates the extent to which people intrinsically devote themselves to a product.
Involvement can be explained as the degree or strength of the psychological correlation
between an individual and a stimulus object. It emphasizes the extent of an object’s
relatedness, correlations, or commitment to an individual’s self-concept, needs, and/or values.
Outcome involvement is a motivational state of an individual. According to Johnson
and Eagly (1989), outcome involvement focuses on the mental state evoked by stimuli. It
does not require enduring personal relevance or the arousal of central values as a necessary
prerequisite for involvement. It indicates the amount or state of perceived importance, interest,
emotional attachment, arousal, drive, activation, and/or motivation.

16

Krugman (1965) described situational media involvement as, “By this we do not
mean attention, interest, or excitement but the number of conscious bridging experiences,
connections, or personal references per minute that the viewer makes between his own life
and the stimulus”(p.584).

The content of involvement is the actualized reactions to an

individual in a specific stimulus context. The stimulus context could either be a marketing
communication situation or a specific medium situation. One typical example of this
situational media involvement is the audience’s state of immersion into narrative messages,
such as a story or a movie.
Moreover, the Transportation Theory is conceptually relevant to Slater’s idea of the
Extended Elaboration Likelihood Model (EELM). Slater (2002) proposed the Extended
Elaboration Likelihood Model on the basis of Petty and Cacioppo’s classic Elaboration
Likelihood Model (ELM). While the traditional ELM only describes responses to overtly
persuasive messages addressing outcome-relevant topics, the EELM was adopted to address
persuasion processes in entertainment or narrative persuasion contexts, such as movies,
television programs and stories.
According to Slater (2002), both the ELM and the EELM posit that attitude change
may occur through one of two different processing routes: the central route or the peripheral
route (Petty & Cacioppo, 1986). Slater (2002) stated that one of the contrasts between the
ELM and the EELM concerns the variables that predict the amount of central processing. In
the ELM, central processing is predicted by outcome involvement, while in the EELM, such
processing is predicted by narrative interest and identification with protagonists. In other
words, in overtly persuasive messages argument strength is relevant, but in the context of

17

movies, story plot interest and identification with protagonists will be influential.
According to Tang (2009), the contrast between the Transportation Theory and the
Elaboration Likelihood Model (ELM) also lies in the understanding of the relationship
between the central route and the peripheral route for information processing. Although both
the ELM and the Transportation Theory state that involvement is a moderator in information
processing, their emphasis is different. According to the ELM, it is believed that when
involvement is high, central processing will have a stronger impact than peripheral
processing. When involvement is low, peripheral processing will have a stronger impact. In
comparison, the Transportation Theory states that when involvement is high, viewers will
align their perceptions of peripheral content (e.g. place relevant information through movie
scanning) with perceptions of central content (e.g. the narrative meaning the story has
projected).
The Adapted Meaning Transfer Model
In the literature of product placement, Russel (1998) proposed the Adapted Meaning
Transfer Model (AMTM) to understand how product placement works in the context of
movies or television shows. The major idea of this model is to assess the effectiveness of
product placement in terms of transformation. According to Russel, the overall process of
product placement can be identified as a form of transformation. The notion of transformation
was first raised by Puto and Wells (1984) in their research on transformational advertising.
According to their study, transformational advertising is defined as advertising which would
make the experience of using a brand richer and more enjoyable by connecting the experience
of the ad with that of using the brand. When “consumers cannot remember that brand without

18

recalling the experience generated by the advertisement” (Puto and Wells 1984, p.638),
transformational advertising is regarded as effective and successful.
By drawing the parallel between advertising and product placement, the Adapted
Meaning Transfer Model suggests that successful product placement can intimately connect
the experience of using a brand with what is shown in movies or television shows. The movie
or the television show can be seen as a long drama advertisement, which endorses a brand
with a specific story and all the members of the cast. When a place is embedded as a brand in
the movie or the television show, the story and all the characters will work together to
endorse the place image and transform the physical location into something laden with
emotional and symbolic meanings.
Moreover, Russel (1998) suggested that affective conditioning will drive most of the
product placement process. She proposed that “the pairing of a product with an emotionally
rich show (television or movie) conditions a transfer of affect from the show to the product”
(Russel 1998, p.363). This suggests that placing products within shows that elicit
positive/negative emotional responses will translate into a similar emotional response to the
product.
Relevant Constructs
Movie Transportation
Transportation has been studied with great interest in the advertising and marketing
field in recent years, because media engagement is an increasingly important factor that
marketers need to consider when making advertising placements. For example, Wang and
Calder (2006) adopted the Transportation Theory to evaluate the impact of media context on

19

the effectiveness of advertising. In their study, transportation was defined as a process of
narrative information processing in which a person not only attends to information but is also
absorbed into the flow of a story in a pleasurable and active way (Wang & Calder 2006).
They focused on print media and found that transportation into magazine stories positively
affected advertising that did not intrude on the transportation process, but negatively affected
advertising that interrupted the transportation experience.
According to Green et al. (2008), transportation also happens in the audio/visual
context in addition to the print context, although the original explorations of transportation
used only written materials. They stated that mental engagement can be created regardless of
the way in which the narrative is conveyed. Some empirical evidence has already found that
movie and print can be equivalently engaging. For example, Dal Cin, Zanna, and Fong (2004)
examined transportation into both movies and print. Their individual difference scale geared
toward movies predicted transportation into both movies and print, and the individual
difference scale assessing tendencies to become transported to written stories also predicted
transportation into both media.
Consistent with the literature, in this study, movie transportation is defined as the state
of immersion into a movie. It measures how well a movie can engage the viewers.
Transported viewers are completely focused on the story presented by the movie. The viewers
may lose track of time, fail to notice events going on around them, and experience vivid
mental images of settings and characters.
Movie Genre
Genre is a French word meaning “type” or “kind”. It has a lengthy origin in literary

20

criticism long before the advent of the cinema. The meaning of genre varies considerably and
it is very difficult to identify a tenuous school of thought on the subject (Grant, 1995). As
Wellek and Warren (1956) advocated, genre “should be conceived…as a grouping of literary
works based theoretically upon both outer form (specific meter or structure) and also upon
inner form (attitude, tone, purpose-more crudely, subjects and audience)” (p. 260).
As for the cinematic equivalents, Tudor (1995) stated that, from the perspective of
movie critics, movie genre is associated with the notion of conventions and codes, such as
themes, actions, and characteristic mannerisms. He argued that although it is difficult to
categorize movies into mutually exclusive groups, it seems that movies do have something in
common, which can make two stories part of the same genre. For example, the western genre
can be defined by both outer form (iconography) and inner form (content). The first outer
form for western genre is the setting, which usually features the very particular kinds of
country: deserts, mountains, plains, and woods. The second outer form is the clothes, which
is often characterized by wide-brimmed hats, open-neck shirts with scarves, tight jeans, and
high-heeled boots. Another essential outer form is the various tools of the trade, principally
guns and horses. When it comes to the inner form, basically it refers to what kind of story the
western genre will present. Typical western movies will be associated with the stories about
the survival of man in the hostile natural environment, and about the establishment of
civilization.
While movie genre is defined based on iconography and content from the perspective
of movie critics, it can also be defined based on the audience’s responses and expectations
(Langford, 1995). According to Altman (1996), a cinema based on genre films depends on a

21

stable, generically trained audience, sufficiently knowledgeable about genre systems to
recognize genre cues, sufficiently familiar with genre plots to exhibit generic expectations. In
practice, the online movie rental company Netflix and numerous online movie reviewers have
defined movie genres based on the general mood a movie can create among the audience. On
one hand, the feel-good movies, such as romance and comedy, usually are associated with a
delightful watching experience. On the other hand, the dark movies, such as crime and horror,
are usually associated with anxiety and fearful excitement.
As for the feel-good movies, the romance genre enjoys a sustained preference from
the American movie goers (Preston, 2000). According to Preston (2000), a very general
definition of romance is a “film in which the development of love between the two main
characters is the primary narrative thread, the main story line” (p.227). Romance films are
often seen to be hybrids of different theorized genres. Based on a very large narrative
distinction for the purposes of analysis, the romance genre can be categorized into romantic
comedy, screwball comedy, drama, and the hybrid.

Particularly, as for romantic drama, the

primary narrative line usually deals with the development and recognition of love between
the two main characters. Unlike romantic comedy, romantic drama equally involves both
romantic and serious content. Typical examples of romantic dramas are Bridges of Madison
County, One Fine Day, Up Close and Personal, and Lost in Translation. These movies
approach romance and love as something that happens to people in the midst of their going
about their lives rather than characterize romance as an event out of ordinary life.
Among the dark movies, the American public has shown an increasing interest in
crime films (Wilson, 2000). According to Wilson (2000), violence is an important convention

22

in the crime genre. For this movie genre, violence is the motivation and it can give the
maximum definition to a story. It has become a part of the nation’s covert culture. Typical
examples are L.A. Confidential, Pulp Fiction, Mulholland Falls, The Usual Suspects, and Kill
Bill (Vol.1 and Vol. 2). For the most part of these movies, it was concerned with the criminal
element and the criminal act itself, in all its various and violent manifestations.
According to the genre theory (Fowler, 1982), texts have attributes specific to one
genre but not others. Russel (1998) proposed that placing products within shows that elicit a
positive or a negative emotional response will translate into a similar emotional response to
the product. Moreover, there is a debate in the industry as to whether dark movies will
adversely influence place images and visitation interest. For example, according to Weekend
Australian (2005), the release of the horror movie Wolf Creek in 2005 had brought doubts
about the movie’s potential impact on the country’s rural tourism. The movie’s director Greg
Mclean confessed that it would be difficult for him to come up with a rationale to win
government funding to finance the movie. As a result, it would be interesting to see if movie
genres based on the audience’s response, particularly dark movies versus feel-good movies,
will have any different influence on people’s perceptions on embedded place and visitation
interest. For this study, I chose two movies as the stimuli, a violent crime thriller (Kill Bill
Vol.1) and a romantic drama (Lost in Translation), and conducted an experiment to
investigate the movie genre’s impact on the viewers’ perception of place images and
visitation interest.
Place Image and Visitation Interest
Place image is an important construct in the place branding and destination marketing

23

literature. In the destination marketing field, place image is commonly termed as destination
image and is defined as the sum of beliefs, ideas and impressions that people have of a place
or destination based on information processing from a variety of sources over time, resulting
in an internally accepted mental construct (Baloglu & McCleary, 1999; Gartner, 1993).
Despite its importance, destination image studies have been criticized as lacking a theoretical
and conceptual framework (Echtner & Ritchie, 1993; Gartner, 1993). The concept of
destination image has not been understood in a unified way (Kim & Richardson, 2003).
Gartner (1993) proposed a typology of eight place or destination image formation
agents relating to the degree of control by the promoter and credibility with the target market.
He categorized the image formation agents into two groups: 1) induced and 2) non-induced.
The four “induced” categories are overt induced I (traditional advertising), overt induced II
(information received from tour operators), covert induced I (second-party endorsement of
products through traditional forms of advertising), covert induced II (secondary-party
endorsement through unbiased reports such as newspaper articles). These four “induced”
categories are within greater control of place marketers, but are less credible. On the other
hand, the “non-induced” categories are autonomous (news and popular culture), unsolicited
organic (unsolicited information received from friends and relatives), solicited organic
(solicited information received from friends and relatives), and organic (information based on
actual visitation). These four “non-induced” categories are somewhat out of the place
marketers’ control, but are authoritative and credible.
In the destination marketing literature, destination images have been described as
consisting of different components. For example, when measuring the destination image of

24

Mexico held by US citizens, Crompton (1979) conceptualized destination image as the sum
of cognitive beliefs and affective impressions that an individual possesses of a particular
destination. Similarly, Baloglu and Bringerg (1997) summarized that destination image is
characterized by subjective perceptions that consist of both high levels of cognitive aspects
(belief) and affective aspects (feeling). Moreover, Yueksel and Akgeuel (2007) reported that
the affective component of image has a substantial impact on travelers’ evaluations and
choice of destinations. Gartner (1993) stated that the interrelationship of cognitive and
affective image components will eventually determine the predisposition for visiting a
destination.
Based on these indications, place image in this study is an evaluative attitudinal
judgment that was comprised of cognitive and affective elements. The measurement of place
image in this study will include both the cognitive and affective aspects. Affective place
image refers to the people’s emotional responses to places and environmental features, while
cognitive place image is based on the evaluation of physical attributes of places. Affective
place image is defined based on the work by Baloglu and Brinberg (1997), which consists of
two basic dimensions, including pleasant-unpleasant and sleepy-arousing. Cognitive place
image is defined based on the work by Baloglu and McClearly (1999), which also consists of
two basic dimensions, including tourism attraction and community quality.
Furthermore, according to Gartner (1993), destination image also consists of the
conative component in addition to the cognitive and affective aspects. Since the conative
image is analogous to behavioral intention, it can be considered as the likelihood of visiting a
destination within a certain time period (Pike & Ryan, 2004). According to Chen and Tsai

25

(2007), destination image had a direct effect on behavioral intentions and an indirect effect on
behavioral intentions through trip quality, perceived value, and satisfaction. Moreover,
Alcaniz, Sanchez, and Blas (2009) also found a direct effect of cognitive destination image
on tourism behavioral intentions. Consistent with the literature, the conative component of
place image in this study is treated as a behavioral intention variable as visitation interest.
Initial Place familiarity
Familiarity with a place is believed to play an important role in influencing an
individual’s perceptions. On one hand, the previous studies have demonstrated that place
familiarity has a positive influence on the people’s perceptions. For example, according to
Olsen, McAlexander and Roberts (1986), as tourists become more knowledgeable about a
place, they have more feelings of security and comfort, which leads to increased confidence
in destination choice. Hunt (1975) also suggested that people who had visited the United
States generally had a more favorable opinion of the United States than those who had not
visited the United States. However, on the other hand, as noted by MacKay and Fesenmaier
(1997), place familiarity might also have negative effects. They introduced the concept of
“optimal familiarity”, which indicates that place familiarity and attractiveness are only
positively related to a point, after which they are negatively related because the novelty of
travel is reduced.
As for the definition of place familiarity, there is not a unified agreement in the
literature. According to Hu and Ritchie (1993), place familiarity can be influenced by a
number of factors, such as geographic distance, previous personal visitation experience, and
the level of overall knowledge about a place. In this study, the concept of place familiarity is

26

treated as an attitudinal variable, similar to Kim and Richardson’s study (2003). The concept
is operationally measured with four statements about the target place’s physical attributes,
including historical sites, cultural attractions, natural landscapes, and night entertainment life.
Hypotheses and Research Questions
In the literature of product placement, Russel (1998) proposed the Adapted
Meaning Transfer Model (AMTM) to understand how product placement works in the
context of movies or television shows. By drawing the parallel between advertising and
product placement, the Adapted Meaning Transfer Model suggests that a successful product
placement can intimately connect the experience of using a brand with what is shown in the
movies or television shows. The movie or the television show can be regarded as an
extremely long drama advertisement, which endorses a brand with a specific story and all the
members of the cast. When a place is embedded as a brand in the movie or the television
show, the story and all the characters will work together to endorse the place image and
transform the physical location into something laden with emotional and symbolic meanings.
Moreover, Russel (1998) suggested that affective conditioning will drive most of the
product placement process. She proposed that “the pairing of a product with an emotionally
rich show (television or movie) conditions a transfer of affect from the show to the product”
(p. 363). Staats (1996) demonstrated how the conditioned emotion-eliciting properties of the
stimulus work. He accomplished this by pairing simple words with a person. The example
demonstrated that the pairing of positive words such as pretty, honest, smart, rich, and so on
with a person would increase the degree of positive reinforcement associated with the person.
Based on this logic, it is rational to anticipate that if the emotional response associated with a

27

stimulus is negative, the reinforcement can go in the other direction and generate a negative
affective transfer. When it comes to place embedment in movies, it is reasonable to
hypothesize that feel-good movies will generate a positive impact on place image and
visitation interest, while dark movies may bring about adverse influences. So my hypotheses
are:
H1:

The movie viewers who are exposed to feel-good movies will generate
more favorable images of the embedded place than those who are in the
control group, which will be followed by the viewers who are exposed to
dark movies.

H2:

The movie viewers who are exposed to feel-good movies will generate
more visitation interest to the embedded place than those who are in the
control group, which will be followed by the viewers who are exposed to
dark movies.

Additionally, Green and Brock (2000) demonstrated how persuasive influence is a
function of transportation and relies on a narrative world with plot and characters. According
to the Transportation Theory, transportation into a narrative world is a state of immersion into
a story. It was proposed as a mechanism whereby narratives can affect beliefs. Green and
Brock (2004) explored the theoretical linkage between transportation and media enjoyment.
They suggested that the audience’s enjoyment positively correlates with their experience of
being immersed in a narrative world and also positively correlates with the consequences of
that immersion. According to their study, it is believed that transportation somewhat
resembles the experience of flow, which is brought about by absorption in an activity and is

28

often marked by a deep sense of enjoyment. Moreover, the study stated that the enjoyment of
transportation experience does not necessarily have to happen in the positive narrative
context, since individuals are often drawn into stories that are frightening. As a result, I also
hypothesize that:
H3: For both feel-good movies and dark movies, the more the movie
viewers are transported to the movie, the more favorable place image
they will have.
H4: For both feel-good movies and dark movies, the more the movie
viewers are transported to the movie, the more visitation interest they
will have.
According to the Adapted Meaning Transfer Model, the viewers exposed to dark
movies will have less favorable place images and visitation interest than the viewers exposed
to feel-good movies. Conversely, the Transportation Theory indicates that regardless of the
movie genre, the more the movie viewers are transported, the more favorable place images
and visitation interest they will have. The question is, which theory will determine if the dark
movie will hurt the place image or improve the place image? In other words, is the effect of
movie genre stronger or is the effect of movie transportation stronger? If movie genre is more
influential, than almost all dark movies should be avoided for place marketing purposes,
because no matter if they can transport the audience or not, they are doomed to hurt the place
images. On the other hand, if the effect of movie transportation is stronger, then all movies,
regardless of genre, can be considered by the place marketers, as long as they could transport
the audience with the stories. Consequently, the following research question is put forth:

29

Research Question 1:

Does the transportation effect weaken the influence
brought about by movie genre?

Furthermore, in the literature of tourism studies, place familiarity has been found to
be an important factor influencing consumers’ perceptions of place images. Place familiarity
is influenced by many factors, including the overall knowledge of a place, geographic
distance to the location, and previous personal visitation experience (Hu & Richitie, 1993).
Based on the literature, place familiarity could have positive effects on the people’s
perceptions of place images. Anand, Holbrook, and Stephens (1988) proposed that an
increase in knowledge about an object might cause an increase in feelings toward the object.
In line with this logic, I hypothesize that initial place familiarity will have a positive
relationship with people’s perceptions and visitation interest. Thus:
H5: Initial place familiarity will have a main effect such that the more
people are initially familiar with an embedded place, the more
favorable place images of it they will have.
H6: Initial place familiarity will have a main effect such that the more
people are initially familiar with an embedded place, the more
visitation interest they will have.
On the other hand, the literature also proposed that place familiarity might have
negative effects, but only under specific conditions. MacKay and Fesenmaier (1997)
introduced the concept of “optimal familiarity” and proposed that destination familiarity and
attractiveness should be positively related to a point, after which they will be negatively
related because the novelty of the place is reduced. It is believed that exposure to the movies

30

can provide the audience with vicarious visiting experience of the target place (Kim &
Richardson, 2003). As a result, the more the viewers are transported by the movies, the more
details they should be able to learn about the featured historical sites, landscapes, and local
people’s lifestyles. Therefore, the audience’s initial place familiarity should interact with
movie transportation to influence the viewers’ place images and visitation interest. So I
hypothesize:
H7: Movie transportation and initial place familiarity will have two-way
interaction such that for the viewers who are initially unfamiliar with
the embedded place, the more they are transported by the movie, the
more positive impact the movie will have on their place images;
while for the viewers who are initially familiar with the embedded
place, the more they are transported by the movie, the more negative
impact the movie will have on their place images.
H8: Movie transportation and initial place familiarity will have two-way
interaction such that for the viewers who are initially unfamiliar with
the embedded place, the more they are transported by the movie, the
more positive impact the movie will have on their visitation interest;
while for the viewers who are initially familiar with the embedded
place, the more they are transported by the movie, the more negative
impact the movie will have on their visitation interest.
Finally, in the literature of movie-induced tourism, although it has been found that
movies can significantly influence the viewers’ perceptions of place images right after movie

31

exposure (e.g. Kim & Richardson, 2003; Shani, Wang, Hudson, & Gil 2008: Tasci, 2009), it
is still unclear how long a movie’s impact can endure. The question is, can a movie by itself
generate a long-term impact on the viewers’ perceptions of place images or does it have to be
reinforced by prior or subsequent promotion messages? As a result, the following research
question is put forth:
Research Question 2:

Does the movie’s impact on place images and
visitation interest change over time?

32

Method
Design: The study was conducted using a posttest only control group
experimental design. The advantage of this design is that it eliminates pretesting effects,
including effects of prior observation on later observation and of potential sensitization of
subjects to experimental manipulation (Campbell & Stanley, 1963). The equality of the
subjects was accomplished through random assignment of the subjects to the two experiment
groups and the control group by using randomly generated numbers. A detailed report of the
subjects’ characteristics is reported in Table 1. Based on the information below, the subjects
were not different in any of the following factors across groups, including gender, age, year in
school, major, ethnicity, movie watch frequency, domestic travel frequency, and international
travel frequency.
Table 1: Respondent Characteristics
Romantic
Drama
(n=95)

Violent
Crime
Movie
(n=85)
Gender

Male
Female

Statistical
Analysis

27(32.1%) 38(40.4%) 32(31.1%) Chi-square(2)
57(67.9%) 56(59.6%) 71(68.9%) =2.20, n.s.
20.19

Age

Year in
School

Control
Group
(n=106)

Freshman
Sophomore
Junior
Senior
Graduate

20.77

20.40

25(29.4%)
12(14.1%)
36(42.2%)
12(14.1%)
0(.0%)

15(16.0%)
21(22.3%)
36(38.3%)
21(22.3%)
1(1.1%)

15(14.4%) Chi-square(8)
18(17.3%) =11.66, n.s.
45(43.3%)
25(24.0%)
1(1.0%)

33

F(2,282)=2.42
n.s.

Table 1 (cont’d)
Major

Advertising
Communication
Packaging
Business
Media Arts
Other

41(25.0%)
12(38.7%)
1(25.0%)
15(44.1%)
1(14.3%)
15(32.6%)

63(38.4%)
7(22.6%)
2(50.0%)
7(20.6%)
2(28.6%)
14(30.4%)

60(36.6%) Chi-square(10)
12(38.7%) =10.66, n.s.
1(25.0%)
12(35.3%)
4(57.1%)
17(37.0%)

Ethnicity

American Indian
Black, non-Hispanic
White, non-Hispanic
Asian
Hispanic
Other

0(.0%)
7(8.2%)
65(76.5%)
12(14.1%)
1(1.2%)
0(.0%)

0(.0%)
12(12.6%)
61(64.2%)
18(18.9%)
4(4.2%)
0(.0%)

5(4.7%)
Chi-square(10)
12(11.3%) =17.65, n.s.
70(66.0%)
16(15.1%)
1(.9%)
2(1.9%)

Movie
watch
frequency

11.12

11.59

10.25

F(2,270)=.84
n.s.

Domestic
travel
frequency

10.00

7.77

8.00

F(2,283)=.23
n.s.

Internatio
-nal travel
frequency

1.80

2.12

1.59

F(2, 282)=.84
n.s.

Movie Stimuli: The experimental treatments were two entertainment movies. An
entertainment movie was defined as a film produced for the entertainment of the general
public employing plot and characters (Kim & Richardson, 2003).
To select the movie stimuli, a survey was conducted among the college students
during the early stages of the research project and course extra credit was offered as incentive
to the participants. The students were asked to identify at least one movie that had generated
their travel interests. They were also asked to identify the featured place name in the movie
and a brief explanation. Among the 312 students, 265 of them returned the questionnaire. The

34

top 10 most frequently mentioned movies and featured locations are reported in Table 2.
Table 2: Most Frequently Mentioned Movies and Locations
Movies
Sex and the City
Eat, Pray, Love
Lord of the Rings
Harry Potter
Sisterhood of the Traveling Pants Forgetting Sarah Marshal
Couples Retreat
Locations
Italy
Greece
Ireland

New York
Las Vegas
London
France

P.S. I Love You
The Hangover
Gladiator

England
Rome
Hawaii

Note: Each location can be featured in multiple movies. As a result, the movies do
not match locations in this table.
For this study, infrequently mentioned movies that featured moderate popular
locations among the college students were chosen as potential stimuli. In this way, existing
positive bias in terms of movie transportation and place image perception among the
audience could be avoided. A list of infrequently mentioned movie examples and moderate
popular locations is reported in Table 3.

Table 3: Infrequently Mentioned Movies and Locations
The Departed
Gran Torino
Kill Bill
Sweet Home Alabama

Boston
Michigan
Tokyo
Alabama

Movies
Vickie Christina Barcelona
Man on Fire
Field of Dreams

City of God
Beerfest
Across the Universe

Locations
Barcelona
Mexico
Iowa

Rio De Janeiro
Germany
Liverpool

35

Moreover, based on the study of Hudson and Ritchie (2006), several movie factors
were also taken into consideration when narrowing down the choices of movie stimuli. First,
the movie should have featured an identifiable and accessible location. Second, there is a
clear link between the story and the location. Finally, the movie has a substantial amount of
exposure of the location throughout the story.
With all these criteria in mind, one feel-good movie and one dark movie about Tokyo
were chosen as the stimuli for this study. One is Kill Bill: Vol. 1, a violent crime thriller and
the other is Lost in Translation, a romantic drama. Kill Bill: Vol. 1 is a movie about revenge.
The leader character, “The Bride,” a professional assassin, seeks revenge on a group of
people, crossing them off a list one by one as she kills them. The movie describes how “The
Bride” survives an attack during her wedding some years ago and how she confronts her first
target, O-Ren Ishii, in Tokyo. While the movie is about bloody killing and the underworld of
Tokyo, it also demonstrates Japan’s martial arts tradition, shows beautiful oriental gardens,
and highlights the nightlife entertainment.
As for Lost in Translation, it is a movie about two Americans in Tokyo. The male
character Bob is a movie star traveling to Japan to shoot a whiskey commercial, while the
female character Charlotte is a young woman tagging along with her workaholic
photographer husband. Unable to sleep, Bob and Charlotte meet each other by chance one
night in the luxury bar and become friends. Later they venture through Tokyo, having often
hilarious encounters with the local people of Japan, and ultimately discover a new belief in
life’s possibilities. The movie is about dislocation and disorientations, while it also shows
Japan’s high-rise architecture, city entertainment, temples, and beautiful countryside.

36

Procedures: The sample for this study was a convenience sample comprised of
college students enrolled in a major public university in the United States. Students from six
different undergraduate courses were recruited for the research project. The reason for
choosing college students as the subjects is because college students fall into the key
demographic segment for both the entertainment movie industry and the tourism industry. A
national Harris study showed that 61% of college students travel during a given school year,
spending almost $5 billion on travel (Harris, 2002).
The sample size for the first exploratory survey to identify movie stimuli was 265
with a response rate of 84.94%. For the main experiment, the sample size was 286 with a
response rate of 74.81%. When it comes to the second posttest, the sample size was 245 with
a response rate of 85.67%.
In the first step of the experiment, recruitment e-mails were sent to the students by the
researcher 4 weeks before the main study, and the subjects were told that the purpose of the
project was to find out how movies could influence people’s perceptions in general. In return
for their time and effort, each participant who completed all three steps of the study received
extra course credits from their instructors and a chance to win one of 17 gift cards. In these
e-mails, links to the questions about the subjects’ basic information were included. The
questions were asked in the following sequence: general movie genre preference, previous
movie stimuli exposure, movie watching frequency, initial place familiarity, domestic travel
frequency, international travel frequency, potential movie watching availability,
demographics, and contact information (Appendix 1). To find out whether the subjects had
watched the two movie stimuli before, 10 additional entertainment movies were listed along

37

with the two movie stimuli, and the subjects were asked to indicate their watching experience
for all the movies. In this way, the subjects would not know which movies were of interest
for this study in advance.
In response to the recruitment e-mail, 405 students signed up for this study. Among
these students, 42 of them had watched Kill Bill: Vol. 1 and 31 of them had watched Lost in
Translation. As a result, these 73 subjects were excluded from the experiment groups. The
remaining subjects were randomly assigned to the two experiment groups and the control
group. Finally, 135 subjects were assigned to each of the three groups.
In the second step, the subjects assigned to the two movie groups were shown the
movie stimuli respectively in a classroom theater setting. For the violent crime thriller group,
88 out of the 135 subjects showed up as scheduled. As for the romantic drama, 109 out of the
135 subjects showed up as scheduled. During the movie screening time, soft drinks and
snacks were provided. Right after the movie exposure, the subjects were handed the first
posttest questionnaire. The questionnaire recorded their movie transportation level, affective
place image, cognitive place image, visitation interest, a movie fact recall test, and some filter
questions such as character evaluation, character empathy involvement, and movie theme
relevant beliefs (Appendix 2).
Meanwhile, the subjects assigned to the control group were informed that they only
needed to complete two surveys to receive the extra credit and the chance to win a gift card.
Among the 135 registered subjects, 110 of them agreed to continue with the research project
and completed the first survey, which recorded their general perceptions of the city images of
Detroit, Tokyo, and Boston (Appendix 3).

38

Two weeks before the second step took place, an earthquake struck off the coast of
Japan on March 11, 2011 and churned up a tsunami that swept over cities and farmland in the
northern part of the country. Japan later faced a nuclear emergency. Explosions and leaks of
radioactive gas took place in three reactors at the Fukushima Daiichi Nuclear Power Station.
Because of this unexpected natural disaster, the subjects’ responses from all three groups in
this study might be negated. To monitor how much the earthquake was on the subjects’ minds
when they answered the questionnaires, a 3-item scale was created to measure the
earthquake’s impact on the subjects’ place perceptions and visitation interest.
In the third step, the subjects from all three groups were asked to complete the second
posttest (Appendix 4) one month after the first posttest. In the survey, the subjects were asked
to evaluate their perceptions of Tokyo again. The measurements of affective place image,
cognitive place image, and visitation interest were the same as those that appeared in the first
posttest. After the second posttest was completed, the subjects were debriefed and thanked for
their participation in this project.
Measures: The following section identifies the proposed items that were used to
measure the independent and dependent variables.
Movie Transportation: Movie transportation was measured by the 13-item, 7-point
bipolar scale developed by Green and Brock (2000). The wording of the items was adjusted
to fit into the specific movie-viewing contexts. Scores on each of the individual items were
summed and averaged to create the final measure. Since the control group did not watch any
movie, the transportation scale was analyzed only for the two experiment groups. The
Cronbach’s alpha coefficient for the scale was .75. The mean score was 3.78 for the violent

39

movie group and 3.81 for the romantic drama group. This indicates that the movie viewers
had a moderate level of movie transportation for both the violent crime thriller and the
romantic drama (Table 4).
Table 4: Descriptive Statistics of Transportation Scale Items

Item

Scale

a

Violent
Crime
(n=85)
Mean
(SD)

Romantic
Drama
(n=95)
Mean
(SD)

b

Movie transportation (.75)

1. While I was watching the movie, I could easily
picture the events in it taking place.
2. While I was watching the movie, activity going
on in the room around me was on my mind. (R)
3. I could picture myself in the scene of the events
described in the movie.
4. I was mentally involved in the movie while
watching it.
5. After watching the movie, I found it easy to put it
out of my mind. (R)
6. I wanted to learn how the movie ended.
7. The movie affected me emotionally.
8. I found myself thinking of ways the movie could
have turned out differently.
9. I found my mind wandering while watching the
movie. (R)
10. The events in the movie are relevant to my
everyday life.
11. The events in the movie have changed my life.
12. While watching the movie I had a vivid image of
the leader character A.
13. While watching the movie I had a vivid image of
the leader character B.

a

b

3.62(1.94)

5.04(1.62)

1-7

4.84(1.80)

4.45(1.76)

1-7

2.46(1.68)

3.95(1.77)

1-7

4.71(1.79)

4.33(1.80)

1-7

4.27(1.54)

3.54(1.66)

1-7
1-7
1-7

5.14(1.94)
3.71(1.76)
3.65(2.07)

4.75(1.92)
3.07(1.84)
4.63(2.12)

1-7

4.13(1.97)

3.26(1.72)

1-7

1.40(1.12)

2.63(1.73)

1-7
1-7

1.41(.93)
4.99(1.50)

1.68(1.02)
4.00(2.06)

1-7

4.88(1.51)

4.15(2.11)

3.78(.84)

Grand Mean

1-7

3.81(.95)

Scale: 1=Not at all and 7=Very much. Number in parentheses indicates Cronbach’s
alpha coefficient.
40

Initial Place Familiarity: Initial place familiarity was measured with the 4-item
7-point bipolar scale used by Kim and Richardson (2003). This scale estimated the subjects’
familiarity with the physical environment and local lifestyle in Tokyo before movie exposure.
Descriptive statistics for four familiarity items are reported in Table 5. Overall, all the
subjects rated the items below 3.5, indicating that the subjects did not consider Tokyo as a
familiar place. Principal components analysis suggested that all four items were loaded on
one factor and a Cronbach’s alpha coefficient of .93 indicated high internal consistency of
items.
Table 5: Descriptive Statistics of Initial Place Familiarity Items

Item

Scale

Violent
Crime
(n=85)
Mean(SD)

Romantic
Drama
(n=95)
Mean (SD)

a

Control Group
(n=106)
Mean (SD)

Initial Place Familiarity
b

(.93 )

1. How familiar are you with
the lifestyle of people in
Tokyo?
2. How familiar are you with
the cultural/historical
attractions in Tokyo?
3. How familiar are you with
the landscape in Tokyo?
4. How familiar are you with
the nighttime
entertainment in Tokyo?

1-7

3.49(1.75)

3.42(1.87)

3.73(1.81)

1-7

2.98(1.68)

3.09(1.82)

3.32(1.78)

1-7

3.09(1.74)

3.11(1.78)

3.42(1.74)

1-7

2.60(1.54)

2.89(1.83)

3.08(1.76)

3.04(1.51)

3.13(1.69)

3.38(1.58)

Grand Mean
a

b

Scale: 1=Extremely Unfamiliar and 7=Extremely Familiar. Number in parentheses
indicates Cronbach’s alpha coefficient.

41

Cognitive Place Image: To measure the cognitive component of the subjects’ images
of Tokyo, the scale items developed by Russel and Pratt (1980) were used with some
modifications. After some changes in wording, 13 image attributes, which were measured
using a 7-point Likert scale, were used in the questionnaire. These items evaluated the
subjects’ perceptions of historical and natural attractions, atmosphere of the community, and
lifestyle of the local people.
Principal components analysis with varimax rotation was performed on the cognitive
image items to identify dimensions underlying the original 13 items. To examine the
suitability of the data for factor analysis, Kaiser’s measures of sampling adequacy were
checked for the overall data set and each variable. Overall, this was .76, which was
acceptable (Tabachnik & Fidell, 2007). When all 13 items were entered into the principal
components analysis, an eigenvalue of 1.0 was utilized for factor extraction and loadings
of .60 were used for item inclusion (Tabachnik & Fidell, 2007). Five items (“It seems to me
that Tokyo’s standards of cleanliness and hygiene are low,” “Tokyo has appealing local food,”
“Tokyo offers quality nighttime entertainment,” “Reliable local transportation is available in
Tokyo,” and “A trip to Tokyo is good value for the money”) were dropped because of low
factor loading scores. Thus, the eight cognitive image items from the questionnaire resulted
in two factors that accounted for 59.31% of the total variance (Table 6). Factors were labeled
based on highly loaded items and the common characteristics when grouped together.
Accordingly, they were labeled as tourism attractions (Factor 1) and community quality
(Factor 2). Table 6 shows that eigenvalues of these factors ranged from 1.47 to 3.27 and all
the loadings were greater than .60, indicating a good correlation between the items and the

42

factor on which they were loaded. Cronbach’s alpha coefficients were analyzed to check the
internal consistency of the scale and coefficients were above the satisfactory level (above .70)
in tourism attractions and community quality. As a result, these two factors were used as the
cognitive image variables in the subsequent hypothesis tests.

Table 6: Factor Analysis of Cognitive Place Image Items
Violent
Crime
(n=85)
Mean(SD)

Romantic
Drama
(n=95)
Mean
(SD)

Control
Group
(n=106)
Mean (SD)

Factor
Loading

a

b

(. 76)

Interesting
cultural
attractions

4.60(1.55)

5.43(1.48)

5.61(1.05)

.80

Interesting
historical
attractions

4.22(1.48)

5.04(1.59)

4.92(1.31)

.77

Quality
accommodations

5.01(1.40)

5.71(1.46)

5.19(1.33)

.70

Impressive
natural sceneries

5.07(1.45)

5.22(1.67)

4.75(1.64)

.69

43

% of
Variance

3.27

Factor 1:
Tourism
attractions

Eigen
Value

30.98

Table 6 (cont’d)
1.47

Factor 2:
Community

28.33

b

quality (.72)
Unpolluted
environment

3.59(1.11)

3.49(1.45)

3.39(1.44)

.75

Good climate

4.05(1.07)

4.37(1.19)

4.40(1.18)

.76

Friendly local
people

4.11(1.25)

5.06(1.41)

4.47(1.18)

.72

Safe place to
visit

3.85(1.48)

4.81(1.40)

4.42(1.26)

.65
59.31

a

b

Scale: 1=Strongly disagree and 7= Strongly agree. Number in parentheses indicates
Cronbach’s alpha coefficient.

Affective Place Image: Affective evaluations of Tokyo were measured by the scales
developed by Baloglu and Brinberg (1997). Two 10-item scales assessed the two basic
bipolar affective dimensions of places: unpleasant–pleasant and sleepy–arousing. Following
the instructions, eight interval positions were provided for rating the extent to which each
adjective described feelings toward Tokyo. Affective image scores were calculated according
to the instructions provided by Russel and Pratt (1980). From each dimension, scores on
positively and negatively keyed items were totaled separately. The sums of negative items
were then subtracted from those of positive ones. For example, for the unpleasant–pleasant
dimension, scores of the five positive items (pleasant, nice, pleasing, pretty, and beautiful)
and the five negative items (dissatisfying, displeasing, repulsive, unpleasant, and
uncomfortable) were summed respectively. Then the total of negative bipolar items was
44

subtracted from that of positives within the same dimension. This procedure produced two
bipolar dimensions (unpleasant–pleasant and sleepy–arousing), which captured the basic
affective image of Tokyo (Table 7). Cronbach’s alpha coefficients were analyzed to check the
internal consistency of the scales and the coefficients were .92 for the dimension of
unpleasant–pleasant and .86 for the dimension of sleepy–arousing. As a result, the two
dimensions of affective image were used as two affective image variables in the subsequent
hypothesis tests.
Table 7:

Measures of Affective Place Image

a

Violent Crime
Movie
(n=85)
Mean(SD)

Dimension

Unpleasant-Pleasant

Romantic
Drama
(n=95)
Mean (SD)

Control
Group
(n=106)
Mean (SD)

Total
(N=286)

.28(1.40)

1.04(1.35)

1.12(1.01)

.83(1.30)

1.71(.94)

1.59(1.33)

1.53(.83)

1.61(1.05)

b

Quality (.92)

Sleepy-Arousing
b

Quality(.86 )

a
b

Scale: 1=Extremely Inaccurate and 8=Extremely Accurate.
Number in the parentheses indicates Cronbach’s alpha coefficient.

Visitation Interest: The variable of visitation interest was measured by the 3-item,
7-point scale used by Putrevu and Lord (1994). The subjects were asked to evaluate the
following three statements: “It is very likely that I am going to travel to Tokyo,” “I would like
to travel around Tokyo,” and “I would like to travel to Tokyo for my next vacation.” The

45

Cronbach’s alpha coefficient for the scale was .84.
Earthquake’s Impact: Earthquake’s impact was measured by a 3-item 7-point Likert
scale. The statements assessed the participants’ self-reported evaluation of how much the
earthquake was on their minds when they were answering the questions about Tokyo’s
cognitive image, affective image, and visitation interest. The statements were listed at the
very end of the questionnaire and the subjects were asked not to change any existing answers
after reading these statements. Principal components analysis suggested that all 3 items were
loaded on one factor and a Cronbach’s alpha coefficient of .95 indicated high internal
consistency. Descriptive statistics for these three statements are reported in Table 8. The
means were all below 3.0, which indicated that the earthquake in Japan generally had a weak
impact on the subjects’ evaluations of Tokyo’s image in this study. Meanwhile, it also
demonstrated that the earthquake had more negative impact on the subjects’ evaluations in the
control group than in the two experiment groups. As a result, earthquake’s impact was
included as a covariate in the subsequent hypothesis tests.

46

Table 8: Descriptive Statistics of Earthquake’s Impact Item

Item

Earthquake’s Impact (.95)

Violent
Crime
(n=85)
Mean(SD)

Romantic
Drama
(n=95)
Mean(SD)

Control
Group
(n=106)
Mean(SD)

1-7

2.04(1.26)

2.13(1.44)

2.81(1.90)

1-7

1.91(1.17)

2.03(1.40)

2.63(1.71)

1-7

2.14(1.57)

2.18(1.54)

2.80(1.86)

Scale

b

1. The recent earthquake in Japan has
negatively influenced my answers to
question A. (Questions A asked
about your general impression of
Tokyo’s cultural attractions,
nighttime entertainment, local
transportation, etc.)
2. The recent earthquake in Japan has
negatively influenced my answers to
question B. (Question B asked you
to rate how accurately some words
describe Tokyo, such as pleasant,
dissatisfying, active, drowsy, etc.)
3. The recent earthquake in Japan has
negatively influenced my answers to
question C. (Question C asked about
your general interest to visit Tokyo.)

a

Scale: 1=Strongly Disagree and 7=Strongly Agree.

b

a

Number in parentheses indicates Cronbach’s alpha coefficient.

47

Results Summary
In this chapter, I have summarized the test results for the study. The results are
organized in order by the hypotheses and research questions. For each hypothesis and
research question, I have described the results in details and provided the relevant tables and
figures.
H1: The movie viewers who are exposed to feel-good movies will generate more
favorable images of the embedded place than those who are in the control group, which will
be followed by the viewers who are exposed to dark movies.
Affective image dimension 1: pleasant–unpleasant
An analysis of covariance (ANCOVA) was carried out with movie group as the
independent variable and the earthquake’s impact as the covariate. The following output
(Table 9) shows that this hypothesis is partially supported, F(2, 282)=16.45, p<.0001. In
particular, the movie viewers who were exposed to the violent crime thriller (mean=.19)
generated more unfavorable (unpleasant) place images than the viewers who were in the
control group (mean=1.20). However, viewers exposed to the romantic drama (mean=1.00)
did not have more favorable (pleasant) place images than those from the control group
(mean=1.20). This result indicated that, as expected, the violent crime movie had a
significantly negative impact on the audience’s place images on the dimension of
pleasant–unpleasant, but the romantic drama did not have a positive impact on the audience’s
perceptions.

48

Table 9: H1 Test Result for Affective Image-Pleasant/Unpleasant
Between-Subjects Factors
Movie Group

N

1.Violent Crime

85

2. Romantic Drama

95

3. Control

106

Tests of Between-Subjects Effects
Dependent Variable: Affective Image (Pleasant-Unpleasant)
Source
Corrected Model

Type III Sum
of Squares

Movie Group
Earthquake’s
Impact
Error
Total
Corrected Total
a

a

3

Mean
Square
19.302

118.874
49.819
15.827

1
2
1

118.874
24.910
15.827

427.077
684.040
484.984

282
286
285

1.514

57.907

df

F
12.745

Sig.
.000

78.493
16.448
10.451

.000
.000
.001

R Squared = .119 (Adjusted R Squared = .110)

Estimates
Dependent Variable: Affective Image (Pleasant-Unpleasant)
95% Confidence Interval
a
Mean
Movie Group
Std. Error
Lower Bound
Upper Bound
1
.134
-.070
.459
.194
2
.127
.754
1.252
1.003
3
.122
.957
1.435
1.196
a

Covariates appearing in the model are evaluated at the following values:

Earthquake’s Impact = 2.28.

49

Table 9 (cont’d)
Pairwise Comparisons
Dependent Variable: Affective Image (Pleasant-Unpleasant)
95% Confidence
Interval for

(I) movie
group
1
2
3
a

(J) movie
group
2
3
1
3
1
2

Mean
Differenc Std.
a
Sig.
e (I-J)
Error
-.809
.184
.000
-1.002
.183
.000
.809
.184
.000
-.193
.177
.829
1.002
.183
.000
.193
.177
.829

Difference

a

Lower
Bound
-1.251

Upper
Bound
-.366

-1.443
.366
-.619
.560
-.233

-.560
1.251
.233
1.443
.619

Adjustment for multiple comparisons: Bonferroni

Affective place image dimension 2: sleepy–arousing
An analysis of covariance (ANCOVA) was carried out with movie group as the
independent variable and the earthquake’s impact as the covariate. The following output
(Table 10) shows that this hypothesis is not supported, F(2, 282)=.22, p=.80. The movie
viewers from all three groups had similar place images on the affective image dimension of
sleepy–arousing right after movie exposure. For violent crime movie, the mean was 1.67,
while the mean for romantic drama was 1.57 and the mean for the control group was 1.60.
This result indicated that contrary to the expectations, neither the violent crime movie nor the
romantic drama had any significant influences on the viewers’ place images on the dimension
of sleepy–arousing.

50

Table 10: H1 Test Result for Affective Image-Sleepy/Arousing
Between-Subjects Factors
Movie Group

N

1.Violent Crime

85

2. Romantic Drama

95

3.Control

106

Tests of Between-Subjects Effects
Dependent Variable: Affective Image (Sleepy-Arousing)
Type III Sum
Mean
Source
of Squares
df
Square
Corrected Model
3
3.758
a
11.275
Movie Group
Earthquake’s
Impact
Error
Total
Corrected Total
a

319.932
.473
9.810

1
2
1

300.976
1049.540
312.251

282
286
285

F
3.521

Sig.
.016

319.932 299.761
.237
.222
9.810
9.191

.000
.801
.003

1.067

R Squared = .036 (Adjusted R Squared = .026)

Estimates
Dependent Variable: Affective Image (Sleepy-Arousing)
95% Confidence Interval
Std.
a
Mean
movie group
Error
Lower Bound
Upper Bound
1
.113
1.445
1.889
1.667
2
.106
1.359
1.777
1.568
3
.102
1.389
1.791
1.590
a

Covariates appearing in the model are evaluated at the following values:

Earthquake’s Impact= 2.28.

51

Table 10 (cont’d)
Pairwise Comparisons
Dependent Variable: Affective Image (Sleepy-Arousing)
95% Confidence Interval
(I)
movie (J) movie
group group
1
2
2
3
a

for Difference
Mean Difference
(I-J)
.099

3
1
3
1
2

.077
-.099
-.022
-.077
.022

Std.
a
Sig.
Error
.154 1.000
.154 1.000
.154 1.000
.149 1.000
.154 1.000
.149 1.000

a

Lower
Bound
-.273

Upper
Bound
.470

-.294
-.470
-.380
-.447
-.336

.447
.273
.336
.294
.380

Adjustment for multiple comparisons: Bonferroni
Cognitive place image dimension 1: tourism attraction
An analysis of covariance (ANCOVA) was carried out with movie group as the

independent variable and the earthquake’s impact as the covariate. The following output
(Table 11) shows that this hypothesis is partially supported, F(2, 282)=17.84, p<.0001. In
particularly, the movie viewers who were exposed to the violent crime movie (mean=4.41)
generated significantly more unfavorable cognitive place images on the dimension of tourism
attraction than the viewers from the control group (mean= 5.34). However, the romantic
drama (mean=5.07) did not generate significantly more favorable perceptions of tourism
attraction than the viewers from the control group (mean= 5.34). This result indicated that, as
expected, the violent crime movie had a significantly negative impact on the audience’s place
images on the dimension of tourism attraction, but the romantic drama did not have positive
impact on the audience’s perceptions.

52

Table 11: H1 Test Result for Cognitive Image-Tourism Attraction
Between-Subjects Factors
Movie Group

N

1.Violent Crime

85

2. Romantic Drama

95

3.Control

106

Tests of Between-Subjects Effects
Dependent Variable: Cognitive Place Image (Tourism Attraction)
Source
Type III Sum of Squares df Mean Square
Corrected Model

Movie group
Earthquake’s Impact
Error
Total
Corrected Total
a

41.559

a

3

2288.273
1
40.944
2
3.623
1
323.594 282
7447.813 286
365.153 285

13.853

F

12.073 .000

2288.273 1994.145 .000
20.472
17.841 .000
3.623
3.157 .077
1.147

R Squared = .114 (Adjusted R Squared = .104)

Estimates
Dependent Variable: Cognitive Place Image (Tourism Attraction)
95% Confidence Interval
a
Mean
movie group
Std. Error Lower Bound Upper Bound
1
.117
4.181
4.642
4.411
2
.110
4.856
5.289
5.073
3
.106
5.136
5.551
5.343
a

Covariates appearing in the model are evaluated at the following

values: Earthquake’s Impact = 2.22.

53

Sig.

Table 11 (cont’d)
Pairwise Comparisons
Dependent Variable: Cognitive Place Image (Tourism Attraction)
95% Confidence Interval for
(I) movie
group
1

3
1
3
1
2

2
3
a

(J) movie
group
2

Mean Difference
(I-J)
-.661
-.932
.661
-.271
.932
.271

Difference

a

Std.
a
Sig.
Error
Lower Bound Upper Bound
.160 .000
-.976
-.346
.159 .000
-1.245
-.619
.160 .000
.346
.976
.154 .079
-.573
.031
.159 .000
.619
1.245
.154 .079
-.031
.573

Adjustment for multiple comparisons: Least Significant Difference (equivalent to no

adjustments).
Cognitive place image dimension 2: community quality
An analysis of covariance (ANCOVA) was carried out with movie group as the
independent variable and the earthquake’s impact as the covariate. The following output
(Table 12) shows that this hypothesis is partially supported, F(2, 282)=7.66, p=.001. In
particular, the movie viewers who were exposed to the violent crime thriller (mean=3.87)
generated significantly more unfavorable cognitive place images on the dimension of
community quality than the viewers from the control group (mean=4.21). However, the
viewers who were exposed to the romantic drama (mean=4.42) did not generate significantly
more favorable images on community quality than the viewers from the control group
(mean=4.21). This result indicates that, as expected, the violent crime movie had a
significantly negative impact on the audience’s place images on the dimension of community
quality, but the romantic drama did not have a positive impact on the audience’s perceptions.

54

Table 12: H1 Test Result for Cognitive Image-Community Quality
Between-Subjects Factors
Movie Group

N

1.Violent Crime

85

2. Romantic Drama

95

3.Control

106

Tests of Between-Subjects Effects
Dependent Variable: Cognitive Image (Community Quality)
Type III Sum
Source
of Squares
df
Mean Square
Corrected Model
3
6.109
a
18.328
Movie Group
Earthquake’s
Impact
Error
Total
Corrected Total
a

1672.847
13.637
5.372

1
2
1

251.031
5256.188
269.359

282
286
285

F
6.863

Sig.
.000

1672.847 1879.222
6.818
7.660
5.372
6.035

.000
.001
.015

.890

R Squared = .068 (Adjusted R Squared = .058)

Estimates
Dependent Variable: Cognitive Image (Community Quality)
95% Confidence Interval
Lower
Upper
a
Mean
Movie Group
Std. Error
Bound
Bound
1
.103
3.665
4.071
3.868
2
.097
4.226
4.608
4.417
3
.093
4.023
4.390
4.207
a

Covariates appearing in the model are evaluated at the following

values: Earthquake’s Impact = 2.22.

55

Table 12 (cont’d)
Pairwise Comparisons
Dependent Variable: Cognitive Image (Community Quality)
95% Confidence Interval
(I)
movie (J) movie
group group
1
2
3
a

2
3
1
3
1
2

for Difference
Mean Difference
(I-J)
-.549
-.339
.549
.210
.339
-.210

Std.
Error
.141
.140
.141
.135
.140
.135

Sig.

a

.000
.016
.000
.121
.016
.121

Lower
Bound
-.826
-.615
.272
-.056
.063
-.477

a

Upper
Bound
-.272
-.063
.826
.477
.615
.056

Adjustment for multiple comparisons: Least Significant Difference (equivalent to no

adjustments).
H2: The movie viewers who are exposed to feel-good movies will generate more
visitation interest to the embedded place than those who are in the control group, which will
be followed by the viewers who are exposed to dark movies.
An analysis of covariance (ANCOVA) was carried out with movie group as the
independent variable and the earthquake’s impact as the covariate. The following output
(Table 13) shows that this hypothesis is not supported, although there is a significant
difference among the three groups, F(2, 282) =12.52, p<.0001. Interestingly, the viewers
exposed to the romantic drama (mean=3.48) did not generate more visitation interest than the
viewers who were exposed to the violent movie (mean=3.21) or the people from the control
group (mean=4.31). Instead, the movie viewers from both the violent movie group
(mean=3.21) and the romantic drama group (mean=3.48) generated significantly less
visitation interest than the people in the control group (mean=4.31) right after movie

56

exposure.
Table 13: H2 Test Result for Visitation Interest
Between-Subjects Factors
Movie Group

N

1.Violent Crime

85

2. Romantic Drama

95

3.Control

106

Tests of Between-Subjects Effects
Dependent Variable: Visitation Interest at Time 1
Source
Corrected Model

Movie Group
Earthquake’s
Impact
Error
Total
Corrected Total
a

Type III Sum
of Squares
64.134

df

a

Mean Square
3
21.378

1431.664
62.232
7.704

1
2
1

1431.664
31.116
7.704

700.918
4698.667
765.052

282
286
285

F
8.601

Sig.
.000

576.000
12.519
3.099

.000
.000
.079

2.486

R Squared = .084 (Adjusted R Squared = .074)

Estimates
Dependent Variable: Visitation Interest at Time 1
movie group
1
2
3
a

Mean

a

Std. Error
.172
3.210
.162
3.484
.155
4.310

95% Confidence Interval
Lower Bound
Upper Bound
2.872
3.548
3.164
3.803
4.006
4.615

Covariates appearing in the model are evaluated at the following values:

Earthquake’s Impact= 2.40.

57

Table 13 (cont’d)
Pairwise Comparisons
Dependent Variable: Visitation Interest at Time 1
95% Confidence Interval
for Difference

(I)
movie (J) movie
group group
1
2
3
1
3
1
2

2
3
a

Mean Difference
(I-J)
-.274
-1.100
.274
-.827
1.100
.827

Std.
a
Sig.
Error
.235 .738
.233
.235
.225
.233
.225

a

Lower
Bound
Upper Bound
-.841
.293
-1.660
-.293
-1.370
.540
.284

.000
.738
.001
.000
.001

-.540
.841
-.284
1.660
1.370

Adjustment for multiple comparisons: Bonferroni.
H3: For both feel-good movies and dark movies, the more the movie viewers are

transported by the movie, the more favorable place image they will have.
Affective place image dimension 1: pleasant-unpleasant
Multiple regression tests were conducted, and movie transportation and the
earthquake’s impact were entered as the predictors. The output (Table 14) shows that there
was significant main effect of movie transportation, β=.18, t=2.45, p=.015, but the effect was
2

2

relatively weak because R =.06 (Adjusted R =.05) when both movies were considered.
Interestingly, when the data were split based on movie group, the output shows that there was
significant interaction between movie transportation and movie group. In particular, for the
2

2

violent crime movie, β=.34, t=3.47, p=.001, R =.19 (Adjusted R =.17), which indicates that
the more the viewers were transported by the movie, the more favorable (pleasant) place
2

images the viewers had. In contrast, for the romantic drama, β=.06, t=.55, p=.58, R =.02
2

(Adjusted R =-.003), which indicates that movie transportation did not significantly influence
58

the viewers’ affective place images on the dimension of pleasant–unpleasant.
Table 14: H3 Test Result for Affective Image-Pleasant/Unpleasant
Main Effect
b

Model Summary
Model
1
a
b

R
.245

R Square
a

Adjusted R Square

.060

Predictors: (Constant), Earthquake’s Impact, Transportation (based on 13 items)
Dependent Variable: Affective Image (Pleasant/Unpleasant)

Model
1
Regression
Residual
Total

b

Sum of
Squares
21.911

df

342.444
364.355

b

Mean Square
2
10.956

177
179

F
5.663

Sig.
.004

a

1.935

Predictors: (Constant), Earthquake’s Impact, Transportation (based on 13 items)
Dependent Variable: Affective Image (Pleasant/Unpleasant)

Coefficients

Model
1 (Constant)
Transportation
Earthquake’s
Impact
a

1.39094

.050

ANOVA

a

Std. Error of the Estimate

Unstandardized
Coefficients
B
Std. Error
-.085
.490
.285
.116
-.165
.078

a

Standardized
Coefficients
Beta

Dependent Variable: Affective Image (Pleasant/Unpleasant)

59

t
Sig.
-.174 .862

.179 2.447 .015
-.155 -2.121 .035

Table 14 (cont’d)
Interaction Effect
b

Model Summary
Movie Group

Model

1

1

2

1

a
b

R

R Square

.436
.137

Adjusted R
Square

Std. Error of
the Estimate

a

.190

.170

1.27283

a

.019

-.003

1.35575

Predictors: (Constant), Earthquake’s Impact, Transportation (based on 13 items)
Dependent Variable: Affective Image (Pleasant/Unpleasant)

ANOVA

b

Movie
Group
1

Model
1 Regression

132.847 82
164.032 84
3.241 2

1.620

2

Residual
Total
1 Regression
Residual

169.100 92

1.838

Total

172.341 94

a
b

Sum of Squares df Mean Square
F
Sig.
31.185 2
15.592 9.624
a
.000

1.620

.882

a

.418

Predictors: (Constant), Earthquake’s Impact, Transportation (based on 13 items)
Dependent Variable: Affective Image (Pleasant/Unpleasant)

60

Table 14 (cont’d)
Coefficients

a

Unstandardized
Coefficients

Movie
Group
1

Model
1 (Constant)

B
Std. Error
-1.363
.681

Transportation

a

.575

.166

Earthquake’s
Impact
1 (Constant)
Transportation
Earthquake’s
Impact

2

Standardized
Coefficients

-.292

.111

.959
.082
-.113

.637
.148
.100

Beta

t
Sig.
-2.002 .049
.344 3.465 .001
-.263 -2.643 .010
1.504 .136
.057 .553 .582
-.118 -1.134 .260

Dependent Variable: Affective Image (Pleasant/Unpleasant)
Affective image dimension 2: sleepy-arousing
Multiple regression tests were conducted and movie transportation and the

earthquake’s impact were entered as the predictors. The output (Table 15) shows that there
was no significant main effect of transportation, β=.12, t=1.68, p=.095. However, when the
data were split based on movie group, there was significant interaction between movie
transportation and movie group. In particular, for the violent crime movie, movie
2

transportation had significant impact, β=.24, t=2.25, p=.027. However, because R was .07
2

(Adjusted R was .04), it suggests that the impact from movie transportation was relatively
weak. As for the romantic drama, movie transportation had no significant impact, β=.06,
t=.56, p=.58.

61

Table 15: H3 Test Result for Affective Image-Sleepy/Arousing
Main effect
Model Summary
Model
1
a

R

Adjusted R
Square

R Square

.193

.037

a

.026

Model
1
Regression
Residual
Total

Sum of
Squares
8.906
229.643
238.550

b

Mean
Square
4.453

df
2
177
179

F
3.432

Sig.
.034

a

1.297

Predictors: (Constant), Earthquake’s Impact, Transportation

b

Dependent Variable: Affective Image (Sleepy/Arousing)

Coefficients

Model
1 (Constant)
Transportation based
on 13 items
Earthquake’s Impact
a

1.13904

Predictors: (Constant), Earthquake’s Impact, Transportation

ANOVA

a

Std. Error of the
Estimate

a

Unstandardized
Coefficients
B
Std. Error
1.282
.401
.160
.095
-.120

Standardized
Coefficients
Beta

.064

Dependent Variable: Affective Image (Sleepy/Arousing)

62

t
Sig.
3.194 .002

.124 1.679 .095
-.139 -1.881 .062

Table 15 (cont’d)
Interaction Effect
Model Summary

Movie Group
1

Model
1

2

1

a

a

.065

Std. Error of
the Estimate
.91052
.043

a

.036

.015

R

R Square

.256
.190

Adjusted R
Square

1.31579

Predictors: (Constant), Earthquake’s Impact at Time 1, Transportation

ANOVA

b

Movie Group
1

Model
1 Regression

2

Residual
Total
1 Regression

67.981 82
72.744 84
5.947 2

Residual

159.281 92

Total

165.227 94

a
b

Sum of Squares df Mean Square
F
Sig.
4.763 2
2.382 2.873
a
.062
.829
2.973 1.717
1.731

Predictors: (Constant), Earthquake’s Impact at Time 1, Transportation
Dependent Variable: Affective Image (Sleepy/Arousing)

63

.185

a

Table 15 (cont’d)
Coefficients

movie
group_p1
1

Unstandardized
Coefficients
Model
1 (Constant)
Transportation
Earthquake’s
Impact
1 (Constant)
Transportation

2

Earthquake’s
Impact
a

a

Standardized
Coefficients

B
Std. Error
.816
.487
.268
.119

Beta

t
Sig.
1.675 .098

-.062

.079

.241 2.254 .027
-.084 -.785 .435

1.628
.080

.619
.144

2.632 .010
.057 .555 .580

-.164

.097

-.174 -1.690 .094

Dependent Variable: Affective Image (Sleepy/Arousing)

Cognitive image dimension 1: tourism attraction
Multiple regression tests were conducted and movie transportation and the
earthquake’s impact were entered as the predictors. The output (Table 16) shows that there
2

was significant main effect of movie transportation, β =.28, t=3.89, p<.0001, R = .08
2

(Adjusted R =.07). This means the more the viewers were transported to the movie, the
more favorable images about tourism attraction they had. When the data were split by movie
2

2

groups, for the violent crime movie, β =.34, t =3.28, p =.002, R =.12 (Adjusted R =.10). As
2

2

for the romantic drama group, β =.25, t =2.46, p =.016, R =.06 (Adjusted R =.04). This
indicates that movie transportation’s main effect was slightly stronger for the violent crime
movie group than the romantic drama group.

64

Table 16: H3 Test Result for Cognitive Image-Tourism Attraction

Model

R

1
a

Model Summary
Adjusted R
R Square
Square

.281

.079

a

Predictors: (Constant), Earthquake’s Impact, Transportation

Model
1 Regression

Sum of
Squares
19.660

Residual
Total

b

229.874
249.534

b

Mean
Square
9.830

df
2
177
179

F
7.569

Sig.
.001

a

1.299

Predictors: (Constant), Earthquake’s Impact, Transportation
Dependent Variable: Cognitive Image (Tourism Attraction)

Coefficients

Model
1 (Constant)
Transportation
Earthquake’s Impact
a

1.13962

.068

ANOVA

a

Std. Error of
the Estimate

a

Unstandardized
Coefficients
B
Std. Error
3.342
.407
.371
.096
.014
.066

Standardized
Coefficients
Beta

Dependent Variable: Cognitive Image (Tourism Attraction)

65

t
Sig.
8.209 .000

.282 3.886 .000
.016 .217 .828

Table 16 (cont’d)
Model Summary
Movie
Group Model
1

1

2

1

a

R Square

Adjusted R
Square

a

.118

.097

1.05645

a

.062

.041

1.13678

R
.344
.249

Predictors: (Constant), Earthquake’s Impact, Transportation

ANOVA
Movie Group Model
1

Std. Error of
the Estimate

b

Sum of Squares df Mean Square

F

1 Regression

12.249 2

6.125 5.488

Residual

91.520 82

1.116

Total
2

1 Regression

7.832 2
118.889 92

Total

b

.006

a

103.769 84

Residual
a

Sig.

3.916 3.030
1.292

126.721 94

Predictors: (Constant), Earthquake’s Impact, Transportation
Dependent Variable: Cognitive Image (Tourism Attraction)

66

.053

a

Table 16 (cont’d)
Coefficients

a

Unstandardized
Coefficients

Standardized
Coefficients

Movie
Group

Model

B

1

1 (Constant)

2.775

.576

4.816 .000

13 items

.453

.138

.341 3.280 .002

cog japan
image

-.028

.099

-.029 -.282 .779

1 (Constant)

3.888

.539

7.218 .000

13 items

.306

.124

.251 2.457 .016

cog japan
image

.017

.084

.020 .199 .842

2

a

Std. Error

Beta

t

Sig.

Dependent Variable: Cognitive Image (Tourism Attraction)

Cognitive place image dimension 2: community quality
Multiple regression tests were conducted and movie transportation and the
earthquake’s impact were entered as the predictors. The output (Table 17) shows that movie
transportation did not have significant effect on community quality, Beta=.11, t=1.49, p=.14.
In other words, whether the movie viewers were transported to the movie or not, there was no
difference in their perception of community quality.
Table 17: H3 Test Result for Cognitive Image-Community Quality

Model
1
a

R
.164

Model Summary
Adjusted R Std. Error of
R Square
Square
the Estimate
.027
.016
.97569
a

Predictors: (Constant), Earthquake’s Impact,

Transportation

67

Table 17 (cont’d)
ANOVA
Sum of
Squares

Model
1

Mean
Square

df
2

168.498

177
179

2.434

.091

a

Predictors: (Constant), Earthquake’s Impact, Transportation
Dependent Variable: Cognitive Image (Community Quality)

Coefficients

Model
1 (Constant)
Transportation
Earthquake’s
Impact
a

Sig.

.952

173.132

F

2.317

Total

b

4.634

Residual
a

Regression

b

Unstandardized
Coefficients
B
Std. Error
3.882
.349
.122
.082
-.083
.057

a

Standardized
Coefficients
Beta

t
Sig.
11.140 .000

.111 1.486 .139
-.109 -1.464 .145

Dependent Variable: Cognitive Image (Community Quality)
H4: For both feel-good movies and dark movies, the more the movie viewers are

transported to the movie, the more visitation interest they will have.
Multiple regression tests were conducted and movie transportation and the
earthquake’s impact were entered as the predictors. The output (Table 18) shows that there
2

was significant main effect of movie transportation, β =.36, t =5.09, p<.0001, R =.13
2

(Adjusted R =.13).

This means that the more the viewers were transported by the movie,

the more visitation interest they exhibited regardless of which movie they saw. To some
degree, a violent crime movie can still have a significant positive impact on visitation interest
when the movie viewers are effectively transported.
68

Table 18: H4 Test Result for Visitation Interest

Model
1
a

Model Summary
Adjusted R
R Square
Square

R
.367

.134

a

Std. Error of
the Estimate

.125

1.42226

Predictors: (Constant), Earthquake’s Impact,

Transportation

ANOVA
Model
1
Regression
Residual
Total
a
b

Sum of
Squares
55.604

b

df
2

358.041
413.644

177
179

F
13.744

Sig.
.000

a

2.023

Predictors: (Constant), Earthquake’s Impact, Transportation
Dependent Variable: Visitation Interest

Coefficients
Unstandardized
Coefficients
Model
1 (Constant)
Transportation
Earthquake’s
Impact
a

Mean
Square
27.802

B
.816
.606
.121

Std. Error
.501
.119
.069

a

Standardized
Coefficients
Beta

t
Sig.
1.629 .105
.357 5.087 .000
.123 1.754 .081

Dependent Variable: Visitation Interest

Research Question 1: Does the transportation effect weaken the influence brought
about by movie genre?
First, a median split was performed for the variable movie transportation. The movie
69

viewers were categorized into two groups: highly transported audience and weakly
transported audience. Then a number of analysis of covariance (ANCOVA) tests were
performed with movie group as the predictor and the earthquake’s impact as the covariate.
Affective place image dimension 1: unpleasant–pleasant
The output below (Table 19) shows that when the audience was weakly transported,
the difference between the violent crime movie group (mean=-.21) and the romantic drama
group (mean=1.02) was significant, F(1, 86)=17.78, p<.0001. However, when the audience
was highly transported, the difference between the violent crime movie group (mean=.73)
and the romantic drama group (mean=1.07) was not significant, F(1, 88)=1.59, p=.21. This
indicates that among the highly transported audience, the violent crime movie did not
significantly decrease the audience’s perceptions of the pleasantness of the embedded place.
This suggests that movie transportation can weaken movie genre’s effect on the movie
viewers’ perceptions of place pleasantness.

Table 19: RQ 1 Result for Affective Place Image-Pleasant/Unpleasant
Highly Transported Audience
Between-Subjects Factors
Movie Group

N

1.Violent Crime

40

2.Romantic Drama

51

70

Table 19 (cont’d)
Tests of Between-Subjects Effects
Dependent Variable: Affective Place Image (Pleasant/Unpleasant)
Type III Sum
of Squares

Source
Corrected Model

df

Mean Square

F

Sig.

a

2

4.939

2.959

.057

44.296

1

44.296

26.537

.000

Movie Group

2.661

1

2.661

1.594

.210

Earthquake

6.475

1

6.475

3.879

.052

Error

146.892

88

1.669

Total

233.940

91

Corrected Total

156.770

90

9.879

Intercept

a

R Squared = .063 (Adjusted R Squared = .042)

Estimates
Dependent Variable: Affective Place Image
(Pleasant/Unpleasant)
Movie
Group
1
2
a

95% Confidence Interval
Mean
.727
1.073

Std. Error

Lower Bound

Upper Bound

a

.205

.320

1.134

a

.181

.713

1.433

Covariates appearing in the model are evaluated at the

following values: Earthquake’s Impact = 2.03.

Weakly Transported Audience
Between-Subjects Factors
Movie Group

N

1. Violent Crime

45

2. Romantic Drama

44

71

Table 19 (cont’d)
Tests of Between-Subjects Effects
Dependent Variable: Affective Place Image (Pleasant/Unpleasant)
Type III Sum
of Squares

Source
Corrected Model

df

Mean Square

F

Sig.

a

2

18.385

9.966

.000

Intercept

20.856

1

20.856

11.305

.001

Movie Group

32.795

1

32.795

17.776

.000

8.338

1

8.338

4.519

.036

Error

158.659

86

1.845

Total

209.750

89

Corrected Total

195.430

88

Earthquake

a

36.771

R Squared = .188 (Adjusted R Squared = .169)

Estimates
Dependent Variable: Affective Place Image (Pleasant/Unpleasant)
95% Confidence Interval
Movie
Group

Mean

1

Upper Bound

a

.204

-.612

.198

a

.206

.613

1.433

-.207

2
a

Std. Error

Lower
Bound

1.023

Covariates values: Earthquake’s Impact = 1.98.

Pairwise Comparisons
Dependent Variable: Affective Place Image (Pleasant/Unpleasant)
95% Confidence Interval for
(I)
(J)
movie movie
group group

Mean
Difference
(I-J)

Difference
Std. Error

Sig.

a

a

Lower Bound Upper Bound

1

2

-1.230

.292

.000

-1.810

-.650

2

1

1.230

.292

.000

.650

1.810

72

Affective place image dimension 2: sleepy-arousing
The output below (Table 20) shows that when the audience was highly transported,
the difference between the violent crime movie group (Mean=1.91) and the romantic drama
group (Mean=1.58) was not significant, F(1, 88)=1.72, p=.19. Meanwhile, when the audience
was weakly transported, the difference between the violent crime movie group (Mean=1.53)
and the romantic drama group (Mean=1.60) was not significant, either, F(1, 86)=.009, p=.77.
This indicates that movie transportation did not change movie genre’s effect on the movie
viewers’ perceptions of place image on the dimension of sleepy-arousing.
Table 20: RQ 1 Result for Affective Place Image-Sleepy/Arousing
Highly Transported Audience
Between-Subjects Factors
Movie Group

N

1. Violent Crime

40

2. Romantic Drama

51

Tests of Between-Subjects Effects
Dependent Variable: Affective Place Image (Sleepy/Arousing)
Source
Corrected Model

Type III Sum
of Squares

df

Mean Square

F

Sig.

a

2

4.441

3.079

.051

120.717

1

120.717

83.702

.000

Movie Group

2.474

1

2.474

1.716

.194

Earthquake

7.028

1

7.028

4.873

.030

Error

126.915

88

1.442

Total

407.010

91

Corrected Total

135.797

90

Intercept

a

8.881

R Squared = .065 (Adjusted R Squared = .044)

73

Table 20 (cont’d)
Estimates
Dependent Variable: Affective Place Image (Sleepy/Arousing)
Movie
Group

95% Confidence Interval
Mean

1
2
a

Std. Error

Lower Bound

Upper Bound

a

.190

1.535

2.291

a

.168

1.245

1.915

1.913
1.580

Covariates appearing in the model are evaluated at the following values:

Earthquake’s Impact = 2.03.
Weakly Transported Audience
Between-Subjects Factors
Movie Group

N

1. Violent Crime

45

2. Romantic Drama

44

Tests of Between-Subjects Effects
Dependent Variable: Affective Place Image (Sleepy/Arousing)
Source
Corrected Model

Type III Sum
of Squares

df

Mean Square

F

Sig.

2

.385

.328

.721

83.997

1

83.997

71.620

.000

Movie Group

.104

1

.104

.089

.766

Earthquake

.735

1

.735

.627

.431

Error

100.862

86

1.173

Total

320.600

89

Corrected Total

101.632

88

Intercept

a

.770

a

R Squared = .008 (Adjusted R Squared = -.016)

74

Table 20 (cont’d)
Estimates
Dependent Variable: Affective Place Image
(Sleepy/Arousing)
Movie
Group
1
2
a

95% Confidence Interval
Mean

Std. Error Lower Bound Upper Bound

a

.162

1.211

1.857

a

.164

1.277

1.930

1.534
1.604

Covariates appearing in the model are evaluated at the

following values: Earthquake’s Impact = 1.98.
Cognitive place image dimension 1: tourism attraction
The output below (Table 21) shows that when the audience was highly transported, the
difference between the violent crime movie group (mean=4.91) and the romantic drama
group (mean=5.20) was not significant, F(1, 88)=1.67, p=.20. However, when the audience
was weakly transported, the difference between the violent crime movie group (mean=4.06)
and the romantic drama group (mean=4.91) was significant, F(1, 86)=13.30, p<.0001. This
indicates that among the highly transported audience, the violent crime movie did not
significantly decrease the audience’s perceptions of tourism attraction. It suggests that movie
transportation can weaken movie genre’s effect on the movie viewers’ perceptions of tourism
attraction.

75

Table 21: RQ 1 Result for Cognitive Place Image-Tourism Attraction
Highly Transported Audience
Between-Subjects Factors
Movie Group

N

1. Violent Crime

40

2. Romantic Drama

51

Tests of Between-Subjects Effects
Dependent Variable: Cognitive Place Image (Tourism Attraction)
Type III Sum
of Squares

Source

df

Mean Square

F

Sig.

7.358a

2

748.556

1

Movie Group

1.975

1

1.975

1.662

.201

Earthquake

4.870

1

4.870

4.098

.046

Error

104.579

88

1.188

Total

2454.938

91

111.937

90

Corrected Model
Intercept

Corrected Total
a

3.679

3.096

.050

748.556 629.885

.000

R Squared = .066 (Adjusted R Squared = .044)

Estimates
Dependent Variable: Cognitive Place Image (Tourism Attraction)
95% Confidence Interval
Movie Group
1
2
a

Mean

Std. Error

Lower Bound

Upper Bound

a

.173

4.564

5.250

a

.153

4.901

5.509

4.907
5.205

Covariates appearing in the model are evaluated at the following values:

Earthquake’s Impact = 1.95.

76

Table 21 (cont’d)
Weakly Transported Audience
Between-Subjects Factors
Movie Group

N

1. Violent Crime

45

2. Romantic Drama

44

Tests of Between-Subjects Effects
Dependent Variable: Cognitive Place Image (Tourism Attraction)
Type III Sum
of Squares

Source
Corrected Model

19.518

Intercept

df

Mean Square
2

a

9.759

F

Sig.

8.223

.001

504.648 425.228

.000

504.648

1

15.783

1

15.783

13.299

.000

1.558

1

1.558

1.313

.255

Error

102.062

86

1.187

Total

1905.875

89

121.580

88

Movie Group
Earthquake

Corrected Total
a

R Squared = .161 (Adjusted R Squared = .141)

Estimates
Dependent Variable: Cognitive Place Image (Tourism Attraction)
Movie
Group
1
2
a

95% Confidence Interval
Mean

Std. Error

Lower Bound

Upper Bound

a

.164

3.730

4.380

a

.165

4.581

5.238

4.055
4.909

Covariates appearing in the model are evaluated at the following

values: Earthquake’s Impact = 2.00.

77

Cognitive place image dimension 2: community quality
The output below (Table 22) shows that when the audience was highly transported,
the difference between the violent crime movie group (mean=4.11) and the romantic drama
group (mean=4.50) was not significant, F(1, 88)=3.56, p=.06. However, when the audience
was weakly transported, the difference between the violent crime movie group (mean=3.69)
and the romantic drama group (mean=4.38) was significant, F(1, 86) =13.15, p<.0001. This
indicates that among the highly transported audience, the violent crime movie did not
significantly decrease the audience’s perceptions of community quality. In other words, when
the viewers are highly transported, movie transportation’s effect can weaken movie genre’s
effect on the movie viewers’ perceptions of community quality.

Table 22: RQ 1 Test Result for Cognitive Place Image-Community Quality
Highly Transported Audience
Between-Subjects Factors
Movie Group

N

1. Violent Crime

40

2. Romantic Drama

51

78

Table 22 (cont’d)
Tests of Between-Subjects Effects
Dependent Variable: Cognitive Image (Community Quality)
Type III Sum
of Squares

Source
Corrected Model

4.636

Intercept

df

Mean Square
2

a

2.318

F

Sig.

2.373

.099

506.396 518.329

.000

506.396

1

3.474

1

3.474

3.556

.063

.871

1

.871

.892

.348

Error

85.974

88

.977

Total

1796.500

91

90.610

90

Movie Group
Earthquake

Corrected Total
a

R Squared = .051 (Adjusted R Squared = .030)

Estimates
Dependent Variable: Cognitive Image (Community Quality)
Movie
Group
1
2
a

95% Confidence Interval
Mean

Std. Error

Lower Bound

Upper Bound

a

.157

3.797

4.419

a

.139

4.228

4.779

4.108
4.503

Covariates appearing in the model are evaluated at the following

values: Earthquake’s Impact = 1.95.
Weakly Transported Audience
Between-Subjects Factors
Movie Group

N

1. Violent Crime

45

2. Romantic Drama

44

79

Table 22 (cont’d)
Tests of Between-Subjects Effects
Dependent Variable: Cognitive Image (Community Quality)
Type III Sum
of Squares

Source
Corrected Model

df

Mean Square

F

Sig.

2

5.738

7.370

.001

505.818

1

505.818

649.699

.000

10.237

1

10.237

13.149

.000

2.647

1

2.647

3.400

.069

Error

66.955

86

.779

Total

1522.500

89

78.430

88

Intercept

11.475

Movie Group
Earthquake

Corrected Total
a

a

R Squared = .146 (Adjusted R Squared = .126)
Estimates

Dependent Variable: Cognitive Image (Community Quality)
95% Confidence Interval

Movie
Group

Mean

1

a

.132

3.425

3.951

a

.134

4.110

4.642

3.688

2
a

Std. Error Lower Bound Upper Bound

4.376

Covariates appearing in the model are evaluated at the

following values: Earthquake’s Impact = 2.00.
Pairwise Comparisons
Dependent Variable: Cognitive Image (Community Quality)
95% Confidence Interval for
(I)
(J)
movie movie
group group

Mean
Difference
(I-J)

Difference
Std. Error

Sig.

a

a

Lower Bound Upper Bound

1

2

-.688

.190

.000

-1.065

-.311

2

1

.688

.190

.000

.311

1.065

a

Adjustment for multiple comparisons: Bonferroni.

80

Visitation Interest
The output below (Table 23) shows that when the audience was highly transported,
the difference between the violent crime movie group (mean=3.85) and the romantic drama
group (mean=3.95) was not significant, F(1, 88) =.13, p=.72. Meanwhile, when the audience
was weakly transported, the difference between the violent crime movie group (mean=2.72)
and the romantic drama group (mean=2.96) was not significant, either, F(1, 86) =.57, p=.45.
This indicates that movie transportation did not change movie genre’s effect on the movie
viewers’ visitation interest.
Table 23: RQ 1 Test Result for Visitation Interest
Highly Transported Audience
Between-Subjects Factors
Movie Group

N

1. Violent Crime

40

2. Romantic Drama

51

Tests of Between-Subjects Effects
Dependent Variable: Visitation Interest
Source
Corrected Model
Intercept

Type III Sum
of Squares
.300

df

Mean Square
2

a

.150

F

Sig.

.075

.928

400.364 199.442

.000

400.364

1

Movie Group

.252

1

.252

.125

.724

Earthquake

.086

1

.086

.043

.837

Error

176.653

88

2.007

Total

1564.444

91

176.952

90

Corrected Total
a

R Squared = .002 (Adjusted R Squared = -.021)

81

Table 23 (cont’d)
Estimates
Dependent Variable: Visitation Interest
95% Confidence Interval

Movie
Group

Mean

1

3.845

2
a

3.952

Std. Error Lower Bound Upper Bound

a

.225

3.397

4.293

a

.199

3.556

4.348

Covariates appearing in the model are evaluated at the

following values: Earthquake’s Impact = 2.15.
Weakly Transported Audience
Between-Subjects Factors
Movie Group

N

1. Violent Crime

45

2. Romantic Drama

44

Tests of Between-Subjects Effects
Dependent Variable: Visitation Interest
Source
Corrected Model
Intercept

Type III Sum of Squares df Mean Square

F

Sig.

2

3.150 1.511 .226

207.135 1

207.135 99.362 .000

6.300

a

Movie Group

1.182 1

1.182

Earthquake

4.250 1

4.250 2.039 .157

Error

179.281 86

Total

902.889 89

Corrected Total

185.581 88

a

R Squared = .034 (Adjusted R Squared = .011)

82

2.085

.567 .454

Table 23 (cont’d)
Estimates
Dependent Variable: Visitation Interest
95% Confidence
Interval
Movie
Group

Mean

1

Lower
Bound

Upper
Bound

a

.217

2.293

3.154

a

.219

2.521

3.392

2.724

2
a

Std.
Error

2.957

Covariates appearing in the model are evaluated at the following

values: Earthquake’s Impact = 2.17.
H5: Initial place familiarity will have a main effect such that the more people are
initially familiar with the embedded place, the more favorable place images they will have.
Affective place image dimension 1: unpleasant-pleasant
Multiple regression tests were conducted and initial place familiarity and the
earthquake’s impact were entered as the predictors. The output (Table 24) shows that there
was a significant main effect of initial place familiarity, β=.21, t=3.56, p<.0001. However,
2

2

since R =.06 (Adjusted R =.05) was very low, it means that the main effect of initial place
familiarity accounted for very little variance in the dependent variable. Interestingly, when
the data were split based on movie group, the output shows that there was also a significant
interaction between initial place familiarity and movie group. In particular, for the violent
crime movie, the influence of initial place familiarity was not significant, β=.10, t=.94, p=.35;
for the romantic drama, the influence of initial place familiarity was significant, β=.23, t=2.14,
2

2

p=.035, R =.06 (Adjusted R = .04); for the control group, the influence of initial place

83

2

2

familiarity was significant, β=.30, t=3.26, p=.002, R =. 13 (Adjusted R =.12). This means
that the main effect of initial place familiarity was stronger for the viewers who were in the
control group than those who were in the experimental groups. As for the romantic drama
viewers, the more they were initially familiar with the place, the more favorable (pleasant)
place images they had. However, this effect did not take place for the viewers who were
exposed to the violent crime movie.
Table 24: H5 Test Result for Affective Image- Pleasant/Unpleasant

Model
1
a

R
.243

R Square
a

Main Effect
Model Summary
Adjusted R Square

.059

.052

Std. Error of the Estimate
1.26997

Predictors: (Constant), Earthquake’s Impact, Initial Place Familiarity
ANOVA

Model
1
Regression
Residual
Total
a

Sum of
Squares
28.555

df
2

456.429
484.984

283
285

Mean
Square
14.277

F
8.852

Sig.
.000

1.613

Predictors: (Constant), Earthquake’s Impact, Initial Place Familiarity

Coefficients

Model
1
(Constant)
Initial Place Familiarity
Earthquake’s Impact
a

a

Unstandardized
Coefficients
B
Std. Error
.594
.190
.170
.048
-.133
.048

Standardized
Coefficients
Beta

.208 3.562 .000
-.160 -2.746 .006

Predictors: (Constant), Earthquake’s Impact, Initial Place Familiarity

84

t
Sig.
3.133 .002

a

Table 24 (cont’d)
Interaction Effect
Model Summary
Movie
Group Model
1
b
1
2
3

b

1

b

1

a

a

.082

.059

Std. Error of the
Estimate
1.35548

a

.062

.042

1.32553

a

.134

.118

.94618

R
.286
.249
.367

R Square

Adjusted R
Square

Predictors: (Constant), Earthquake’s Impact, Initial Place Familiarity

b

1=crime movie, 2=romantic drama, 3=control
ANOVA

b

Movie
Group Model
1
1
Regression

Sum of
Squares
13.372
150.660
164.032
10.695

82
84
2

1.837

1

Residual
Total
Regression

161.646
172.341
14.319

92
94
2

1.757

1

Residual
Total
Regression

2

3

df

Mean
Square F
2 6.686 3.639

5.347 3.043

7.160 7.997

Residual
92.212
103
.895
Total
106.531
105
a. Predictors: (Constant), Earthquake’s Impact, Initial Place Familiarity
b. Dependent Variable: Affective Image (Pleasant/Unpleasant)

85

Sig.
.031

.052

.001

a

a

a

Table 24 (cont’d)
Coefficients

Movie
Group
1

a

Unstandardized
Coefficients

3

a

B
.568

Std. Error
.384

Initial Place
Familiarity

2

Model
1 (Constant)

Standardized
Coefficients
Beta

t
Sig.
1.477 .143

.093

.099

Earthquake’s
Impact
1 (Constant)
Initial Place
Familiarity
Earthquake’s
Impact
1 (Constant)
Initial Place
Familiarity
Initial Place
Familiarity

-.311

.119

-.280 -2.623 .010

.855
.182

.314
.085

2.722 .008
.227 2.136 .035

-.188

.102

-.196 -1.843 .068

.804
.191

.259
.059

3.110 .002
.298 3.256 .002

-.119

.051

-.215 -2.344 .021

.101

.943 .348

Dependent Variable: Affective Image(Pleasant/Unpleasant)

Affective place image dimension 2: sleepy-arousing
Multiple regression tests were conducted and initial place familiarity and the
earthquake’s impact were entered as the predictors. The output (Table 25) shows that there
was no significant main effect of initial place familiarity, β=.10, t=1.72, p=.087. However,
when the data were split based on movie group, there was significant interaction between
initial place familiarity and movie group. In particularly, for the violent crime movie, there
was no significant impact from initial place familiarity, β=-.01, t=-.11, p=.91. Similarly, for
the romantic drama, there was no significant impact from initial place familiarity, β=.10,
t=.96, p=.34. However, for the control group, initial place familiarity had significant impact,

86

2

2

β=.23, t=2.49, p=.014, R =.12 (Adjusted R =.10). This means that for the control group, the
more the viewers were initially familiar with the place, the more arousing place image they
had, but this effect was not significant for either the violent crime movie group or the
romantic drama group.
Table 25: H5 Test Result for Affective Image-Sleepy/Arousing
Main Effect
Model Summary
Model
1
a

R
.211

a

R Square
.045

Adjusted R
Std. Error of the
Square
Estimate
.038
1.02674

Predictors: (Constant), Earthquake’s Impact, Initial Place Familiarity

ANOVA
Model
1
Regression
Residual
Total
a

Sum of
Squares
13.912
298.339
312.251

df
2
283
285

F
6.598

Sig.
.002

a

1.054

Predictors: (Constant), Earthquake’s Impact, Initial Place Familiarity

Coefficients

Model
1 (Constant)
Initial Place
Familiarity
Earthquake’s
Impact
a

Mean
Square
6.956

Unstandardized
Coefficients
B
Std. Error
1.699
.153
.066
.039
-.134

a

Standardized
Coefficients
Beta

.039

Dependent Variable: Affective Image(Sleepy/Arousing)

87

.101

t
Sig.
11.079 .000
1.718 .087

-.201 -3.421 .001

Table 25 (cont’d)
Interaction Effect
Model Summary
Movie Group
1
2
3
a
b

Model

b

1

b

1

b

1

R
.088
.206
.344

R Square

Adjusted R
Square

Std. Error of
the Estimate

a

.008

-.016

.93822

a

.042

.022

1.31142

a

.118

.101

.78957

Predictors: (Constant), Earthquake’s Impact, Initial Place Familiarity
1=crime movie, 2=romantic drama, 3=control

ANOVA

b

Movie
Group
1

Model
1 Regression

Sum of
Squares
.563

2

Residual
Total
1 Regression

72.182
72.744
7.004

82
84
2

158.224
165.227
8.603

92
94
2

1.720

3

Residual
Total
1 Regression
Residual
Total

64.212
72.814

103
105

.623

a
b

df
2

Mean Square
.281

F
.320

88

.727

a

.880
3.502 2.036

4.301 6.900

Predictors: (Constant), Earthquake’s Impact, Initial Place Familiarity
Dependent Variable: Affective Image(Sleepy/Arousing)

Sig.

.136

.002

a

a

Table 25 (cont’d)
Coefficients

a

Unstandardized
Coefficients

Movie
Group
1

Model
1 (Constant)

Standardized
Coefficients

a

-.007

.068

-.012

-.110 .913

Earthquake’s
Impact
1 (Constant)
Initial Place
Familiarity
Earthquake’s
Impact
1 (Constant)
Initial Place
Familiarity
Earthquake’s
Impact

3

Std. Error
.266

Initial Place
Familiarity

2

B
1.854

Beta

t
Sig.
6.969 .000

-.063

.082

-.086

-.773 .442

1.754
.081

.311
.084

5.646 .000
.103 .962 .339

-.200

.101

-.213 -1.986 .050

1.442
.122

.216
.049

6.683 .000
.231 2.494 .014

-.117

.042

-.256 -2.769 .007

Dependent Variable: Affective Image(Sleepy/Arousing)

Cognitive place image dimension 1: tourism attractions
Multiple regression tests were conducted and initial place familiarity and the
earthquake’s impact were entered as the predictors. The output (Table 26) shows that there
2

was significant main effect of initial place familiarity, β=.34, t=6.44, p<.0001, R =.21
2

(Adjusted R =.21). This means that for all three groups, meaning the violent crime movie
group, the romantic drama group, and the control group, the more the viewers were initially
familiar with the embedded place, the more favorable cognitive place images they had.

89

Table 26: H5 Test Result for Cognitive Image-Tourism Attraction

Model
1
a

R
Square

R
.460

Model Summary
Adjusted R
Square

a

.211

.206

1.00867

Predictors: (Constant), Earthquake’s Impact, Initial Place Familiarity

ANOVA
Model
1 Regression
Residual
Total
a
b

b

Sum of Squares df Mean Square
F
77.226
2
38.613 37.952
287.927 283
365.153 285

Sig.
.000

a

1.017

Predictors: (Constant), Earthquake’s Impact, Initial Place Familiarity
Dependent Variable: Cognitive Image (Tourism Attraction)

Coefficients

Model
1 (Constant)
Initial Place
Familiarity
Earthquake’s
Impact
a

Std. Error of the
Estimate

Unstandardized
Coefficients
B
Std. Error
3.399
.195
.242
.038
.388

a

Standardized
Coefficients
Beta

.074

t
Sig.
17.440 .000

.341

6.438 .000

.279

5.272 .000

Dependent Variable: Cognitive Image (Tourism Attraction)

Cognitive place image dimension 2: community qualities
Multiple regression tests were conducted and initial place familiarity and the
earthquake’s impact were entered as the predictors. The output (Table 27) shows that there
90

was a significant main effect of initial place familiarity, β=.20, t=3.37, p=.001. This means
that for all three groups, the more the viewers were initially familiar with the embedded place,
2

the more favorable cognitive place images they had. However, since R =.06 (Adjusted
2

R =.05), the main effect of initial place familiarity was relatively weak.
Table 27: H5 Test Result for Cognitive Image-Community Quality
Model Summary
Model
1
a

R
.235

a

Adjusted R
Std. Error of the
Square
Estimate
.049
.94820

R Square
.055

Predictors: (Constant), Earthquake’s Impact, Initial Place Familiarity

ANOVA
Model
1 Regression
Residual
Total
a

Sum of
Squares
14.919
254.439
269.359

df

Mean Square
7.460
.899

2
283
285

F
Sig.
8.297 .000a

Predictors: (Constant), Earthquake’s Impact, Initial Place Familiarity

Coefficients
Unstandardized
Coefficients
Std.
B
Error
4.034
.141
.121
.036

a

Standardized
Coefficients

Model
Beta
t
Sig.
1 (Constant)
28.692 .000
Initial Place
.198 3.373 .001
Familiarity
Earthquake’s
-.110
.038
-.170 -2.883 .004
Impact
a Dependent Variable: Cognitive Image (Community Quality)

91

H6: Initial place familiarity will have a main effect such that the more people are
initially familiar with the embedded place, the more visitation interest they will have.
Multiple regression tests were conducted and initial place familiarity and the
earthquake’s impact were entered as the predictors. The output (Table 28) shows that there
2

was significant main effect of initial place familiarity, β=.45, t=8.29, p<.0001, R =.20
2

(Adjusted R =.19). This means that for the subjects from all three groups, the more they were
initially familiar with the embedded place, the more travel interest they had right after movie
exposure.
Table 28: H6 Test Result for Visitation Interest

Model
1
a

R
.444

a

Model Summary
Adjusted R
R Square
Square
.197
.192

Predictors: (Constant), Earthquake’s Impact, Initial Place Familiarity

ANOVA
Model
1
Regression
Residual
Total
a
b

Std. Error of the
Estimate
1.47309

Sum of
Squares
150.942

b

df
2

614.110
765.052

283
285

Mean
Square
75.471

F
34.779

2.170

Predictors: (Constant), Earthquake’s Impact, Initial Place Familiarity
Dependent Variable: Visitation Interest

92

Sig.
.000

a

Table 28 (cont’d)
a

Coefficients
Unstandardized
Coefficients
Model

B

Standardized
Coefficients

Std. Error

1 (Constant)
Initial Place Familiarity

2.555
.462

.215
.056

Earthquake’s Impact

-.135

Beta

.053

a

t

Sig.

11.873 .000
.450 8.287 .000
-.139 -2.566 .011

Dependent Variable: Visitation Interest

H7: Movie transportation and initial place familiarity will have two-way interaction
such that for the viewers who are initially unfamiliar with the embedded place, the more they
are transported by the movie, the more positive impact the movie will have on their place
images; while for the viewers who are initially familiar with the embedded place, the more
they are transported by the movie, the more negative impact the movie will have on their
place images.
Affective place dimension 1: pleasant-unpleasant
Multiple regression tests were conducted and initial place familiarity, movie
transportation, the interaction term of familiarity and transportation, and the earthquake’s
impact were entered as the predictors. The output below (Table 29) shows that this hypothesis
2

2

is supported, β= -1.13, t= -3.03, p=.003, R =.12 (Adjusted R =.10). The scatter plot (Figure 1)
shows that, for the movie viewers who were initially unfamiliar with the embedded place, the
more they were transported by the movies, the more favorable (pleasant) place images they
had. On the other hand, for the movie viewers who were initially familiar with the embedded
place, the effect of movie transportation on their place images was reversed. In particular, the
more the movie viewers were transported by the movies, the less favorable (pleasant) place
93

2

images they had. However, since R was small, it means that the impact from the interaction
term was relatively weak. In this sense, the results should be interpreted with caution.
Table 29: H7 Test Result for Affective Image-Pleasant/Unpleasant
Model Summary
Model
1
a

R
.352

Adjusted R
Std. Error of the
Square
Estimate
.104
1.35031

R Square
.124

a

Predictors: (Constant), Earthquake’s Impact, Initial Place Familiarity,

Transportation, Familiarity x Transportation

ANOVA
Model
1 Regression

Sum of Squares
45.271

Residual
Total
a

df

Mean Square
11.318

4

319.084
364.355

175
179

F
6.207

Sig.
.000

a

1.823

Predictors: (Constant), time 1: Earthquake’s Impact, Initial Place Familiarity,

Transportation, Familiarity x Transportation

Coefficients

Model
1 (Constant)
Transportation
Initial Place
Familiarity
Familiarity x
Transportation
Earthquake’s Impact
a

a

Unstandardized
Coefficients
B
Std. Error
-2.353
.850
.813
.221
1.008
.298

Standardized
Coefficients
Beta

t
-2.769
.511 3.676
1.133 3.377

Sig.
.006
.000
.001

-.226

.075

-1.129 -3.031 .003

-.236

.079

-.221 -2.990 .003

Dependent Variable: Affective Image (Pleasant/Unpleasant)

94

Figure 1: Interaction between Movie Transportation and Initial Place Familiarity
on Affective Place Image-Pleasant/Unpleasant

Affective place image dimension 2: sleepy-arousing
Multiple regression tests were conducted and initial place familiarity, movie
transportation, the interaction term of familiarity and transportation, and the earthquake’s
impact were entered as the predictors. The output below (Table 30) shows that this hypothesis
2

2

is supported, β=-1.10, t=-2.89, p=.004, R =. 08 (Adjusted R =.06). The scatter plot (Figure 2)
shows that for the movie viewers who were initially unfamiliar with the embedded place, the
more they were transported by the movies, the more arousing place images they had. On the
95

other hand, for the movie viewers who were initially familiar with the embedded place, the
effect of movie transportation on place image was reversed. In particular, the more the movie
viewers were transported by the movies, the more sleepy place images they had. However,
2

since R was small, it means that the impact from the interaction term was relatively weak. In
this sense, the results should be interpreted with caution.
Table 30: H7 Test Result for Affective Image-Sleepy/Arousing
Model Summary
Model
1
a

R
.286

R Square
a

Adjusted R Square

.082

.061

Std. Error of the
Estimate
1.11872

Predictors: (Constant), time 1: Earthquake’s Impact, Transportation, Initial

Place Familiarity, Transportation x Place Familiarity

ANOVA
Model
1
Regression
Residual
Total
a

Sum of
Squares
19.531
219.018
238.550

b

df
4

Mean
Square
4.883

175
179

Sig.
.005

1.252

Predictors: (Constant), time 1: Earthquake’s Impact, Transportation, Initial

Place Familiarity, Transportation x Place Familiarity
b

F
3.901

Dependent Variable: Affective Image(Sleepy/Arousing)

96

a

Table 30 (cont’d)
Coefficients

Model

Unstandardized
Coefficients
B
Std. Error

Standardized
Coefficients
Beta

1 (Constant)
Transportation

-.418
.605

.704
.183

-.594 .554
.470 3.304 .001

Initial Place Familiarity

.717

.247

.996 2.899 .004

Transportation x Place
Familiarity
Earthquake’s Impact

-.179

.062

-1.103 -2.892 .004

-.152

.065

-.176 -2.325 .021

t

Sig.

Figure 2: Interaction between Movie Transportation and Initial Place Familiarity
on Affective Place Image-Sleepy/Arousing

97

Cognitive place image dimension 1: tourism attractions
Multiple regression tests were conducted and initial place familiarity, movie
transportation, the interaction term of familiarity and transportation, and the earthquake’s
impact were entered as the predictors. The output below (Table 31) shows that this hypothesis
is not supported, β=-2.22, t=-.61, p=.54. This finding indicates that initial place familiarity
did not moderate movie transportation’s influence on tourism attractions, although it did
moderate movie transportation’s influence on affective place image.
Table 31: H7 Test Result for Cognitive Image-Tourism Attraction

Model
1
a

R
.417

a

Model Summary
Adjusted R Std. Error of
R Square
Square
the Estimate
.174
.155
1.08536

Predictors: (Constant), Earthquake’s Impact,

Transportation, Initial Place Familiarity, Transportation x
Place Familiarity

ANOVA
Model
1
Regression
Residual
Total
a

Sum of
Squares
43.383
206.152
249.534

b

df
4

Mean
Square
10.846

175
179

F
9.207

Sig.
.000

a

1.178

Predictors: (Constant), Earthquake’s Impact, Transportation, Initial Place

Familiarity, Transportation x Place Familiarity

98

Table 31 (cont’d)
Coefficients

a

Unstandardized
Coefficients
Model
1 (Constant)
Transportation

B
2.766
.375

Std. Error
.684
.178

.383
-.037

.240
.060

-.222

-.078

Initial Place Familiarity
Transportation x Place
Familiarity
Earthquake’s Impact
a

Standardized
Coefficients
Beta

t
Sig.
4.044 .000
.285 2.113 .036

.066

-.085 -1.166 .245

.520 1.592 .113
-.613 .540

Dependent Variable: Cognitive Image (Tourism Attraction)
Cognitive place image dimension 2: community quality
Multiple regressions were conducted and initial place familiarity, movie transportation,

the interaction term of familiarity and transportation, and the earthquake’s impact were
entered as the predictors. The output below (Table 32) shows that this hypothesis is not
supported, β=-.01, t=-.03, p=.98. This finding indicates that initial place familiarity did not
moderate movie transportation’s influence on cognitive place image on the second dimension,
either.
Table 32: H7 Test Result for Cognitive Image-Community Quality

Model
1
a

R
.283

a

Model Summary
Adjusted R Std. Error of
R Square
Square
the Estimate
.080
.059
.95411

Predictors: (Constant), Earthquake’s Impact,

Transportation, Initial Place Familiarity, Transportation x
Place Familiarity

99

Table 32 (cont’d)
ANOVA
Sum of
Squares

Model
1

Mean
Square

df

13.824

4

3.456

Residual

159.308

175

173.132

F

Sig.

.910

Total
a

Regression

b

3.797

179

.005

a

Predictors: (Constant), Earthquake’s Impact, Transportation, Initial Place

Familiarity, Transportation x Place Familiarity
b

Dependent Variable: Cognitive Image (Community Quality)

Coefficients

Model
1 (Constant)
Transportation
Initial Place Familiarity
Transportation x Place
Familiarity
Earthquake’s Impact
a

a

Unstandardized
Coefficients
B
Std. Error
3.723
.601
.070
.156
.156
.211
-.002
.053
-.137

Standardized
Coefficients
Beta

.058

t
6.191
.064 .447
.255 .740

-.011

Sig.
.000
.655
.460

-.030 .976

-.181 -2.349 .020

Dependent Variable: Cognitive Image (Community Quality)

H8: Movie transportation and initial place familiarity will have two-way interaction

such that for the viewers who are initially unfamiliar with the embedded place, the more they
are transported by the movie, the more positive impact the movie will have on their visitation
interest; while for the viewers who are initially familiar with the embedded place, the more
they are transported by the movie, the more negative impact the movie will have on their
visitation interest.

100

Multiple regression tests were conducted and initial place familiarity, movie
transportation, the interaction term of familiarity and transportation, and the earthquake’s
impact were entered as the predictors. The output below (Table 33) shows that this hypothesis
is not supported, β=-.50, t=-1.45, p=.15. This finding suggests that initial place familiarity did
not significantly change movie transportation’s influence on people’s visitation interest.

Table 33: H8 Test Result for Visitation Interest
Model Summary
Model
1
a

R
.502

R Square
a

Adjusted R Square

.252

Std. Error of the
Estimate

.235

1.32928

Predictors: (Constant), Earthquake’s Impact, Transportation, Initial Place

Familiarity, Transportation x Place Familiarity

ANOVA
Model
1 Regression
Residual
Total
a

Sum of
Squares
104.422
309.223
413.644

df

b

Mean Square
4
26.105

175
179

Sig.
.000

1.767

Predictors: (Constant), Earthquake’s Impact, Transportation, Initial Place

Familiarity, Transportation x Place Familiarity
b

F
14.774

Dependent Variable: Visitation Interest

101

a

Table 33 (cont’d)
Coefficients

a

Unstandardized Standardized
Coefficients
Coefficients
Model
1 (Constant)
Transportation
Initial Place Familiarity
Transportation x Place Familiarity
Earthquake’s Impact
a

B

Std.
Error

-.516
.752

.835
.218

.754
-.107
-.003

.297
.074
.069

Beta

t

Sig.

.443

-.617 .538
3.449 .001

.795

2.536 .012

-.501
-.003

-1.448 .149
-.038 .970

Dependent Variable: Visitation Interest

Research Question 2: Does the movie’s impact on place images and visitation
interest change over time?
First of all, a number of analysis of covariance (ANCOVA) tests were performed to
explore whether the movie genre’s effect as hypothesized in H1 and H2 still held after 4
weeks. Moreover, multiple regressions were performed to explore whether the main effect of
movie transportation as hypothesized in H3 and H4 still held over time. Finally, a number of
ANCOVA tests were performed to explore whether the effect of movie transportation still can
weaken movie genre’s effect on a long term basis.
I. Movie genre’s effect over time
Affective place image dimension 1: unpleasant-pleasant
An analysis of covariance (ANCOVA) was conducted with movie group as the
independent variable, the measurement of the earthquake’s impact at time 2 as the covariate,
and affective image (pleasant-unpleasant) measured at time 2 as the dependent variable. The
102

output (Table 34) shows that the main effect of movie group still holds after a month, F(2,
241)=6.37, p=.002. In particularly, the movie viewers who were exposed to the romantic
drama (mean=.727) still had significantly more favorable place images than the viewers who
were exposed to the violent crime movie (mean=.214). However, the movie viewers who
were exposed to the romantic drama (mean=.727) did not have significantly more favorable
place images than the viewers who were in the control group (mean=.947).
Table 34: Genre’s Long-Term Effect on Affective Image-Pleasant/Unpleasant

Between-Subjects Factors
Movie Group

N

1. Violent Crime

71

2. Romantic Drama

83

3.Control

91

Tests of Between-Subjects Effects
Dependent Variable: Affective Image (Pleasant/Unpleasant)
Source
Type III Sum of Squares df Mean Square
Corrected Model
3
13.283
a
39.850
90.704
1
Movie Group
21.397
2
Earthquake_T2
24.421
1
Error
404.998 241
Total
551.570 245
Corrected Total
444.848 244
a. R Squared = .090 (Adjusted R Squared = .078)

103

F
Sig.
7.904 .000

90.704 53.975 .000
10.699 6.366 .002
24.421 14.532 .000
1.680

Table 34 (cont’d)
Estimates
Dependent Variable: Affective Image (Pleasant/Unpleasant)
95% Confidence Interval

2. Romantic Drama
3. Control
a

a

Lower Bound
-.091

Upper
Bound
.518

a

.143

.445

1.009

a

Movie Group
1. Violent Crime

Std.
Error
.155

.138

.674

1.220

Mean
.214
.727
.947

Covariates appearing in the model are evaluated at the following values:

Earthquake’s Impact at Time 2= 2.62.
Pairwise Comparisons
Dependent Variable: Affective Image (Pleasant/Unpleasant)
95% Confidence Interval
for Difference
(I) movie
group
1
2
3
a

(J) movie
group
2
3
1
3
1
2

Mean
Difference (I-J)
-.514
-.733
.514
-.220
.733
.220

Std.
Error
.210
.209
.210
.201
.209
.201

Sig.

a

.045
.002
.045
.827
.002
.827

Lower
Bound
-1.019
-1.238
.008
-.705
.229
-.265

a

Upper
Bound
-.008
-.229
1.019
.265
1.238
.705

Adjustment for multiple comparisons: Bonferroni.

Affective place image dimension 2: sleepy-arousing

An analysis of covariance (ANCOVA) was conducted with movie group as the
independent variable, the measurement of the earthquake’s impact at time 2 as the covariate,
and affective image (sleepy-arousing) measured at time 2 as the dependent variable. The
output at time 2 (Table 35) shows that the three movie groups still remain similar in terms of
104

affective image on the dimension of sleepy-arousing, F(2, 241)=2.92, p=.06.
Table 35: Genre’s Long-Term Effect on Affective Image-Sleepy/Arousing
Between-Subjects Factors
Movie Group

N

1. Violent Crime

71

2. Romantic Drama

83

3.Control

91

Tests of Between-Subjects Effects
Dependent Variable: Affective Image-Sleepy/Arousing

Source
Corrected Model

Movie Group
Earthquake_T2
Error
Total
Corrected Total
a

Type III Sum
of Squares
36.886

df

a

293.963
6.346
34.873
261.780
806.410
298.666

Mean Square
3
12.295

1
2
1
241
245
244

293.963
3.173
34.873
1.086

F
11.319

Sig.
.000

270.628
2.921
32.105

.000
.056
.000

R Squared = .124 (Adjusted R Squared = .113)

Estimates
Dependent Variable: Affective Image-Sleepy/Arousing
Movie Group
1. Violent Crime
2. Romantic Drama
3. Control
a

Mean
1.430
1.239
1.630

95% Confidence Interval
Lower Bound
Upper Bound
1.185
1.674

a

Std. Error
.124

a

.115

1.012

1.466

a

.111

1.411

1.849

Covariates appearing in the model are evaluated at the following values:

Earthquake’s Impact at Time 2 = 2.62.

105

Cognitive place image dimension 1: tourism attraction

An analysis of covariance (ANCOVA) was conducted with movie group as the
independent variable, the measurement of the earthquake’s impact at time 2 as the covariate,
and cognitive image (tourism attraction) measured at time 2 as the dependent variable. The
output (Table 36) shows that the main effect of movie group still holds after a month, F(2,
241)=5.75, p=.004. In particular, the movie viewers who were exposed to the romantic drama
(mean=5.10) still had significantly more favorable place images than the viewers who were
exposed to the violent crime movie (mean=4.60).
Table 36: Genre’s Long-Term Effect on Cognitive Image-Tourism Attraction
Between-Subjects Factors
Movie Group

N

1. Violent Crime

71

2. Romantic Drama

83

3.Control

91

Tests of Between-Subjects Effects
Dependent Variable: Cognitive Image(Tourism Attraction)
Source
Type III Sum of Squares df Mean Square
Corrected Model
3
5.549
a
16.646
Movie Group
Earthquake_T2
Error
Total
Corrected Total
a

1792.453
1
15.852
2
1.648
1
332.075 241
6426.313 245
348.720 244

R Squared = .048 (Adjusted R Squared = .036)

106

F
Sig.
4.027 .008

1792.453 1300.856 .000
7.926
5.752 .004
1.648
1.196 .275
1.378

Table 36 (cont’d)
Estimates
Dependent Variable: Cognitive Image (Tourism Attraction)
95% Confidence Interval
Movie Group
1. Violent
Crime
2. Romantic
Drama
3. Control
a

Mean
4.585
5.095
5.185

a

Std. Error Lower Bound Upper Bound
.140
4.310
4.860

a

.129

4.840

5.350

a

.124

4.940

5.431

Covariates appearing in the model are evaluated at the following values:
Earthquake’s Impact at Time 2= 2.46.
Pairwise Comparisons

Dependent Variable: Cognitive Image (Tourism Attraction)
95% Confidence Interval
for Difference
(I) movie
group
1
2
3

(J) movie
group
2
3
1
3
1
2

Mean
Std.
a
Sig.
Difference (I-J) Error
-.510
.190 .023
-.601
.188 .005
.510
.190 .023
-.091
.181 1.000
.601
.188 .005
.091
.181 1.000

Lower
Bound
-.968
-1.054
.052
-.527
.147
-.345

a

Upper
Bound
-.052
-.147
.968
.345
1.054
.527

Based on estimated marginal means
a

Adjustment for multiple comparisons: Bonferroni.

Cognitive place image dimension 2: community quality

An analysis of covariance (ANCOVA) was conducted with movie group as the
independent variable, the measurement of the earthquake’s impact at time 2 as the covariate,
and cognitive image (community quality) measured at time 2 as the dependent variable. The
107

output (Table 37) shows that the main effect of movie group still holds after a month, F(2,
241)=5.80, p=.003. In particularly, the movie viewers who were exposed to the romantic
drama (mean=4.35) still had significantly more favorable place images than the viewers who
were exposed to the violent crime movie (mean=3.77).
Table 37: Genre’s Long-Term Effect on Cognitive Image-Community Quality

Between-Subjects Factors
Movie Group

N

1. Violent Crime

71

2. Romantic Drama

83

3.Control

91

Tests of Between-Subjects Effects
Dependent Variable: Cognitive Image (Community Quality)
Source
Type III Sum of Squares df Mean Square
Corrected Model
3
5.193
a
15.578
Movie Group
Earthquake_T2
Error
Total
Corrected Total
a

1254.008
1
13.740
2
2.109
1
285.250 241
4470.688 245
300.828 244

R Squared = .052 (Adjusted R Squared = .040)

108

F
Sig.
4.387 .005

1254.008 1059.478 .000
6.870
5.804 .003
2.109
1.782 .183
1.184

Table 37 (cont’d)
Estimates

Movie Group
1
2
3
a

95% Confidence Interval
Lower
Upper
Bound
Bound
3.512
4.021

a

Std.
Error
.129

a

.120

4.110

4.582

a

.115

3.977

4.432

Mean
3.766
4.346
4.205

Covariates appearing in the model are evaluated at the following

values: Earthquake’s Impact at Time 2= 2.46.
Pairwise Comparisons
Dependent Variable: Cognitive Image (Community Quality)
95% Confidence
Interval for
Mean
(I) movie (J) movie
Difference
Std.
a
Sig.
group
group
(I-J)
Error
1
2
-.579
.176 .003
3
-.438
.174 .038
2
1
.579
.176 .003
3
.141
.168 1.000
3
1
.438
.174 .038
2
-.141
.168 1.000
Based on estimated marginal means
a

Difference
Lower
Bound
-1.004
-.858
.155
-.263
.018
-.546

a

Upper
Bound
-.155
-.018
1.004
.546
.858
.263

Adjustment for multiple comparisons: Bonferroni.

Visitation Interest
An analysis of covariance (ANCOVA) was conducted with movie group as the
independent variable, the measurement of the earthquake’s impact at time 2 as the covariate,
and visitation interest measured at time 2 as the dependent variable. The output (Table 38)

109

shows that the initial difference among the three movie groups in terms of visitation interest
at time 1 goes away after a month, F(2, 241)=1.80, p=.17. The movie viewers exposed to the
violent crime movie (mean=3.37) and the romantic drama (mean=3.62) have similar levels of
visitation interest with the people from the control group (mean=3.86) at time 2.
Table 38: Genre’s Long-Term Effect on Visitation Interest
Between-Subjects Factors
Movie Group

N

1. Violent Crime

71

2. Romantic Drama

83

3.Control

91

Tests of Between-Subjects Effects
Dependent Variable: Visitation Interest
Source
Type III Sum of Squares df Mean Square
Corrected Model
3
3.170
a
9.509
Movie Group
Earthquake_T2
Error
Total
Corrected Total
a

947.017
1
9.501
2
.207
1
634.938 241
3884.778 245
644.447 244

R Squared = .015 (Adjusted R Squared = .002)

110

F
Sig.
1.203 .309

947.017 359.454 .000
4.750
1.803 .167
.207
.079 .779
2.635

Table 38 (cont’d)
Estimates
Dependent Variable: Visitation Interest

Movie Group
1
2
3
a

95% Confidence Interval
Lower
Upper
Bound
Bound

Mean

Std. Error

a

.193

2.985

3.747

a

.179

3.270

3.974

a

.172

3.522

4.201

3.366
3.622
3.861

Covariates appearing in the model are evaluated at the following

values: Earthquake’s Impact at Time 2= 2.80.
Pairwise Comparisons
Dependent Variable: Visitation Interest
95% Confidence Interval
Mean
(I) movie (J) movie
Difference
Std.
a
Sig.
group
group
(I-J)
Error
1
2
-.256
.262 .991
3
-.495
.261 .177
2
1
.256
.262 .991
3
-.239
.250 1.000
3
1
.495
.261 .177
2
.239
.250 1.000
Based on estimated marginal means
a

for Difference
Lower
Bound
-.889
-1.124
-.377
-.842
-.134
-.364

a

Upper
Bound
.377
.134
.889
.364
1.124
.842

Adjustment for multiple comparisons: Bonferroni.

II. Movie transportation’s main effect over time
Affective place image dimension 1: pleasant-unpleasant
Multiple regression tests were conducted. Movie transportation and the earthquake’s
impact at time 2 were entered as the predictors and the affective image (pleasant-unpleasant)
at time 2 was entered as the dependent variable. The output (Table 39) shows that the main
111

2

effect of movie transportation still held after a month, β=.25, t=3.21, p=.002, R =.07
2

(Adjusted R =.06). When the data were split based on movie group, the output shows that the
effect of movie transportation still held according to the same pattern. In particularly, for the
2

2

violent crime movie, β=.41, t=3.66, p<.0001, R =.17 (Adjusted R =.14); while for the
2

2

romantic drama, β=1.33, t=1.20, p=.234, R =. 02 (Adjusted R = -.002).
Table 39: Transportation’s Long-Term Effect on Affective image-Pleasant/Unpleasant
Main Effect
Model Summary
Model
1
a

R

Adjusted R
Square

R Square

.262

a

.068

Std. Error of
the Estimate

.056

13.47100

Predictors: (Constant), Earthquake’s Impact at time 2,

Transportation

ANOVA

Model
1

df

Mean Square
2

1007.428

27401.637

151

29416.494

F

181.468

Total

b

2014.857

Residual

a

Regression

Sum of
Squares

b

5.552

153

Predictors: (Constant), Earthquake’s Impact at time 2, Transportation
Dependent Variable: Affective image(Pleasant/Unpleasant)

112

Sig.
.005

a

Table 39 (cont’d)
Coefficients

a

Unstandardized
Coefficients

Standardized
Coefficients

Model

B

1 (Constant)

-7.220

4.894

-1.475 .142

Transportation

3.905

1.217

.253 3.209 .002

Earthquake

-.853

.754

-.089 -1.131 .260

a

Std. Error

Beta

t

Sig.

Dependent Variable: Affective image(Pleasant/Unpleasant)
Interaction Effect
Model Summary

Movie
Group Model
1
2
a

1
1

R
.410
.150

R Square

Adjusted R
Square

Std. Error of
the Estimate

a

.168

.144

12.80731

a

.023

-.002

13.52291

Predictors: (Constant), Earthquake’s Impact at Time 2,

Transportation

ANOVA
movie group Model
1

b

Sum of Squares df Mean Square
2259.339 2

Residual

11153.844 68

Total
2

1 Regression

1 Regression

338.507 2
14629.517 80

Total

1129.669 6.887

Sig.

13413.183 70

Residual
a

F

a

164.027
169.254 .926

14968.024 82

182.869

Predictors: (Constant), Earthquake’s Impact at time 2, Transportation

113

.002

.401

a

Table 39 (cont’d)
Coefficients

Movie
Group
1

a

Unstandardized
Coefficients
Model

Std. Error

Beta

t

Sig.

7.248

-2.752 .008

6.887

1.881

.409 3.661 .000

Earthquake_T2

-1.252

1.111

-.126 -1.126 .264

2.343

6.451

.363 .717

Transportation

1.865

1.556

.133 1.199 .234

Earthquake_T2
a

-19.949

Transportation
2

1 (Constant)

B

Standardized
Coefficients

-.674

.994

-.075 -.677 .500

1 (Constant)

Dependent Variable: Affective image(Pleasant/Unpleasant)

Affective place image dimension 2: sleepy-arousing

Multiple regression tests were conducted. Movie transportation and the earthquake’s
impact at time 2 were entered as the predictors and the affective image (sleepy-arousing) at
time 2 was entered as the dependent variable. The output (Table 40) shows that the main
2

effect of movie transportation emerged at time 2, β=.22, t=2.81, p=.006. The Adjusted R

increased from .03 at time 1 to .11 at time 2. When the data were split based on movie group,
the output shows that the effect of movie transportation still held according to the same
2

pattern. In particular, for the violent crime movie, β=.31, t=2.66, p=.01, R =.11 (Adjusted
2

2

2

R =.08); while for the romantic drama, β=.12, t=1.56, p=.122, R =.16 (Adjusted R =.14).

114

Table 40: Transportation’s Long-Term Effect on Affective Image-Sleepy/Arousing
Main Effect
Model Summary
Model
1
a

R

Adjusted R
Square

R Square

.342

.117

a

Std. Error of
the Estimate

.105

11.02704

Predictors: (Constant), Earthquake’s Impact at time 2,

Transportation

ANOVA
Sum of
Squares

Model
1

df

Mean Square
2

1213.481

18360.934

151

20787.896

Sig.

153

9.980

.000

a

Predictors: (Constant), Earthquake’s Impact at time 2, Transportation
Dependent Variable: Affective Image(Sleepy/Arousing)

Coefficients

a

Unstandardized
Coefficients
Model
1

F

121.596

Total

b

2426.962

Residual

a

Regression

b

B

Std. Error

Standardized
Coefficients
Beta

t

Sig.

(Constant)

8.677

4.006

2.166

.032

Transportation

2.799

.996

.216 2.810

.006

-2.267

.617

-.282 -3.671

.000

Earthquake’s
Impact
a

Dependent Variable: Affective Image(Sleepy/Arousing)

115

Table 40 (cont’d)
Interaction Effect
Model Summary
Movie
Group Model
1
2
a

1
1

R

Adjusted R
Square

R Square

.326
.398

a

.106

.080

9.80038

a

.159

.138

11.85017

Predictors: (Constant), Earthquake’s Impact at time 2, Transportation

ANOVA
Movie
Group Model
1

df
2

388.823

6531.228

68

7308.873
2117.428

2

1058.714

11234.114

80

140.426

13351.542

4.048

82

Sig.

70

Regression

F

96.047

Total

b

777.645

Residual

1

Regression

Mean
Square

Total
2

1

Sum of
Squares

b

Residual

a

Std. Error of the
Estimate

7.539

Predictors: (Constant), Earthquake’s Impact at time 2, Transportation
Dependent Variable: Affective Image(Sleepy/Arousing)

116

.022

.001

a

a

Table 40 (cont’d)
Coefficients
Unstandardized
Coefficients

Movie
Group

Standardized
Coefficients

Model

B

1 (Constant)

1

3.472

5.546

.626 .533

3.830

1.440

.308 2.661 .010

-1.172

.851

-.160 -1.378 .173

12.219

5.653

2.162 .034

2.128

1.363

.160 1.561 .122

-3.136

.871

-.369 -3.598 .001

Transportation
Earthquake
2

1 (Constant)
Transportation
Earthquake

a

a

Std. Error

Beta

t

Sig.

Dependent Variable: Affective Image(Sleepy/Arousing)

Cognitive place image dimension1: tourism attraction
Multiple regression tests were conducted. Movie transportation and the earthquake’s
impact at time 2 were entered as the predictors and the cognitive image (tourism attraction) at
time 2 was entered as the dependent variable. The output (Table 41) shows that the main
2

effect of movie transportation still held at time 2, β=.28, t=3.61, p<.0001, R =. 08 (Adjusted
2

R =.07). When the data were split based on movie group, the output shows that the effect of
2

movie transportation still held for the violent crime movie, β=.35, t=3.09, p=.03, R =.17
2

2

(Adjusted R =.14), but not for the romantic drama, β=.21, t=1.90, p=.06, R =.06 (Adjusted
2

R =.03). This means, as far as tourism attraction is concerned, that movie transportation’s
impact lasts longer for the violent crime movie group than for the romantic drama group.

117

Table 41: Transportation’s Long-Term Effect on Cognitive Image-Tourism Attraction
Main Effect
Model Summary
Model
1
a

R
Square Adjusted R Square Std. Error of the Estimate

R
.284

a

.081

.069

1.20561

Predictors: (Constant), Earthquake’s Impact at time 2, Transportation

ANOVA
Sum of
Squares

Model
1

Mean
Square

df

F

2

9.651

219.478

151

238.780

153

.002

a

Dependent Variable: Cognitive Image (Tourism Attraction)

a

Unstandardized
Coefficients
Model

B
(Constant)

Std. Error

3.347
.392

.109

Earthquake’s
Impact

.022

.072

Standardize
d
Coefficients
Beta

.441

Transportation

a

6.640

Predictors: (Constant), Earthquake’s Impact at time 2, Transportation

Coefficients

1

Sig.

1.453

Total

b

19.302

Residual

a

Regression

b

Sig.

7.584

.000

.282

3.606

.000

.024

.310

.757

Dependent Variable: Cognitive Image (Tourism Attraction)

118

t

Table 41 (cont’d)
Interaction Effect
Model Summary
Movie
Group Model
1

1

2

1

a

R
.409
.236

Adjusted R
Square

R Square

Std. Error of
the Estimate

a

.167

.142

1.18003

a

.056

.032

1.16533

Predictors: (Constant),

Earthquake’s

Impact at time 2,

Transportation

ANOVA

Movie Group Model
1

1 Regression
Residual
Total

2

1 Regression

b

Sum of Squares df
18.982

2

94.688 68
6.384

2

108.640 80

Total

b

F

9.491 6.816

Sig.
.002

a

1.392

113.671 70

Residual
a

Mean
Square

3.192 2.351

.102

1.358

115.024 82

Predictors: (Constant), Earthquake’s Impact at time 2, Transportation
Dependent Variable: Cognitive Image (Tourism Attraction)

119

a

Table 41 (cont’d)
Coefficients

a

Unstandardized
Coefficients

Standardized
Coefficients

Movie
Group

Model

B

1

1 (Constant)

2.243

.666

3.370 .001

Transportation

.537

.174

.347 3.091 .003

Earthquake

.158

.106

.168 1.495 .139

4.353

.567

7.676 .000

.255

.134

.206 1.900 .061

-.096

.094

-.111 -1.023 .309

2

1 (Constant)
Transportation
Earthquake

a

Std. Error

Beta

t

Sig.

Dependent Variable: Cognitive Image (Tourism Attraction)

Cognitive place image dimension 2: community quality
Multiple regression tests were conducted. Movie transportation and the earthquake’s
impact at time 2 were entered as the predictors and the cognitive image (community quality)
at time 2 was entered as the dependent variable. Interestingly, the output (Table 42) shows
that the main effect of movie transportation emerged at time 2, β=.25, t=3.15, p=.002, R2=.06
(Adjusted R2=.05). When the data were split based on movie group, the output shows that the
effect of movie transportation was significant at time 2 for the violent crime movie group,
β=.29, t=2.54, p=.01, R2=. 12 (Adjusted R2=.10), but not for the romantic drama group, β=.19,
t=1.73, p=.09, R2=.08 (Adjusted R2=.05). This means, as far as community quality is
concerned, movie transportation’s impact was significant on a long- term basis, although it
was not meaningful right after movie exposure.

120

Table 42: Transportation’s Long-Term Effect on Cognitive Image-Community Quality
Main Effect
Model Summary
Model
1
a

R

Adjusted R
Square

R Square

.250

.062

a

Std. Error of
the Estimate

.050

1.11815

Predictors: (Constant), Earthquake’s Impact at Time 2,

Transportation

b

ANOVA
Sum of
Squares

Model
1

df

F

2

6.289

188.791

151

201.369

153

.008

a

Dependent Variable: Cognitive Image (Community Quality)

a

Unstandardized
Coefficients
Model
(Constant)
Transportation
Earthquake
a

5.030

Predictors: (Constant), Earthquake’s Impact at Time 2, Transportation

Coefficients

1

Sig.

1.250

Total

b

12.578

Residual

a

Regression

Mean
Square

B

Std. Error

2.982

.101

-.038

.067

Beta

.409

.318

Standardize
d
Coefficients
t
7.285

.000

.249

3.149

.002

-.044

-.563

.574

Dependent Variable: Cognitive Image (Community Quality)

121

Sig.

Table 42 (cont’d)
Interaction Effect
Model Summary
Movie
Group Model
1

a

1

2

R

1

.349
.278

Adjusted R
Square

R Square

Std. Error of
the Estimate

a

.122

.096

1.04891

a

.077

.054

1.09066

Predictors: (Constant), Cognitive Image (Community Quality)

ANOVA
Movie Group Model
1

Sum of Squares df Mean Square

1 Regression

2

5.190

Residual

74.814

68

85.194

70

1 Regression

7.993

2

3.996

Residual

95.164

80

1.190

103.157

82

Total
a

Sig.

1.100

Total
2

10.380

F
4.717

3.360

.012

.040

a

a

Predictors: (Constant), Earthquake’s Impact at Time 2, Transportation

Coefficients

Movie
Group Model
1

B

Std. Error

2.046
.393

.154

Earthquake

.122

.094

3.910

.125

-.169

.088

Beta

t

.531

.216

Standardized
Coefficients

.592

Transportation
2

1 (Constant)

Unstandardized
Coefficients

a

1 (Constant)
Transportation
Earthquake

122

Sig.

3.458

.001

.293

2.541

.013

.149

1.294

.200

7.368

.000

.185

1.725

.088

-.206

-1.914

.059

Visitation Interest
Multiple regression tests were conducted. Movie transportation and the earthquake’s
impact at time 2 were entered as the predictors and visitation interest at time 2 was entered as
the dependent variable. The output (Table 43) shows that the main effect of movie
2

2

transportation still held at time 2, β=.19, t=2.37, p=.02, R =.04 (Adjusted R =.03). When the
data were split based on movie group, the output shows that the effect of movie transportation
2

2

still held for the violent crime movie, β=.27, t=2.33, p=.02, R =.10 (Adjusted R =.08), but
2

2

not for the romantic drama, β=.11, t=.99, p=.32, R =.01 (Adjusted R =-.01). This means, as
far as visitation interest is concerned, that movie transportation’s impact lasts longer for the
violent crime movie group than for the romantic drama group.
Table 43: Transportation’s Long-Term Effect on Visitation Interest
Main Effect
Model Summary
Model

R

1
a

R Square

.196

Adjusted R Square

.038

a

Std. Error of the
Estimate

.026

1.62956

Predictors: (Constant), Earthquake’s Impact, Transportation

ANOVA
Sum of
Squares

Model
1

df

15.958

2

7.979

Residual

400.975

151

416.933

F

2.655

Total
a

Regression

Mean
Square

3.005

153

Predictors: (Constant), Earthquake’s Impact, Transportation

123

Sig.
.053

a

Table 43 (cont’d)
Coefficients

a

Unstandardized Coefficients Standardized Coefficients
Model

B

1 (Constant)

2.090

.595

3.513 .001

Transportation

.347

.147

.189 2.365 .019

Earthquake

.045

.078

.046 .572 .568

a

Std. Error

Beta

t

Sig.

Dependent Variable: Visitation Interest
Interaction Effect
Model Summary
Movie
Group Model
1

1

2

1

R
.322
.118

R Square

Adjusted R
Square

Std. Error of
the Estimate

a

.104

.078

1.50015

a

.014

-.011

1.73291

a. Predictors: (Constant), Earthquake’s Impact, Transportation

ANOVA
Movie
Group
1

Model

Sum of Squares

Mean
Square

df
2

153.030

68

Total

170.789
3.405

2

240.239

80

Total

243.644

82

1 Regression

8.879 3.946

Sig.

70

Residual
a

17.759

Residual
2

1 Regression

F

a

2.250
1.703 .567
3.003

Predictors: (Constant), Earthquake’s Impact, Transportation

124

.024

.569

a

Table 43 (cont’d)
Coefficients

Movie
Group Model
1

B

Standardized
Coefficients

Std. Error

1.144
.518

.223

.122

.113

(Constant)

2.948

.200

-.032

.110

Transportatio
n
Earthquake
a

t

.852

.199

Beta

.837

Earthquake
1

(Constant)

Unstandardized
Coefficients

Transportatio
n
2

1

a

Sig.

1.367

.176

.273

2.328

.023

.126

1.077

.285

3.459

.001

.111

.997

.322

-.032

-.289

.773

Dependent Variable: Visitation Interest

III. Movie transportation’s moderation effect over time
Affective place image dimension 1: unpleasant-pleasant
First, a median-split analysis was performed and the audience was divided into two
categories: highly transported audience and weakly transported audience. Then two
ANCOVA tests were conducted with movie group as the independent variable, the
measurement of the earthquake’s impact at time 2 as the covariate, and affective image
(unpleasant-pleasant) measured at time 2 as the dependent variable.
The output (Table 44) shows that movie transportation’s effect was enduring and it still
could weaken movie genre’s effect after 4 weeks. In particular, for the highly transported
audience, the difference between the violent crime movie group (mean=8.98) and the
romantic drama group (mean=7.88) was not significant, F(1, 71) =.13, p=.72. However, for
the weakly transported audience, the difference between the violent crime movie group
125

(mean=-.1.59) and the romantic drama group (mean=7.56) was significant, F(1, 77) =9. 05,
p=.004.
Table 44: Transportation’s Long-Term Moderation Effect on
Affective Image-Pleasant/Unpleasant

Highly Transported Audience
Between-Subjects Factors
Movie Group

N

1. Violent Crime

30

2. Romantic Drama

44

Tests of Between-Subjects Effects
Dependent Variable: Affective Image(Pleasant/Unpleasant)

Source
Corrected Model

Type III Sum
of Squares

df

Mean Square

F

Sig.

2

389.612

2.383

.100

3684.584

1

3684.584

22.539

.000

21.603

1

21.603

.132

.717

776.261

1

776.261

4.748

.033

Error

11606.992

71

163.479

Total

17514.000

74

Corrected Total

12386.216

73

Intercept
Movie Group
Earthquake

a

779.224

a

R Squared = .063 (Adjusted R Squared = .037)

126

Table 44 (cont’d)
Estimates
Dependent Variable: Affective Image(Pleasant/Unpleasant)
95% Confidence Interval
Movie Group
1
2
a

Mean

Std. Error

8.982
7.876

Lower Bound

Upper Bound

a

2.342

4.312

13.652

a

1.932

4.024

11.728

Covariates appearing in the model are evaluated at the following values:

Earthquake’s Impact= 2.47.
Pairwise Comparisons
Dependent Variable: Affective Image(Pleasant/Unpleasant)
95% Confidence Interval for
(I)
(J)
movie movie
group group

Mean
Difference
(I-J)

Difference
Std. Error

Sig.

a

a

Lower Bound Upper Bound

1

2

1.107

3.044

.717

-4.963

7.177

2

1

-1.107

3.044

.717

-7.177

4.963

Based on estimated marginal means
a

Adjustment for multiple comparisons: Bonferroni.
Weakly Transported Audience
Between-Subjects Factors
Movie Group

N

1. Violent Crime

41

2. Romantic Drama

39

127

Table 44 (cont’d)
Tests of Between-Subjects Effects
Dependent Variable: Affective Image(Pleasant/Unpleasant)
Type III Sum
of Squares

Source
Corrected Model

df

Mean Square

F

Sig.

2

853.206

4.632

.013

119.737

1

119.737

.650

.423

1667.422

1

1667.422

9.053

.004

14.854

1

14.854

.081

.777

Error

14182.338

77

184.186

Total

16550.000

80

Corrected Total

15888.750

79

1706.412

Intercept
Movie Group
Earthquake

a

a

R Squared = .107 (Adjusted R Squared = .084

Estimates
Dependent Variable: Affective Image(Pleasant/Unpleasant)
Movie
Group
1

Mean

Std. Error Lower Bound Upper Bound

a

2.121

-5.809

2.639

a

2.175

3.233

11.895

-1.585

2
a

95% Confidence Interval

7.564

Covariates value: Earthquake’s Impact = 2.15.

Pairwise Comparisons
Dependent Variable: Affective Image(Pleasant/Unpleasant)
95% Confidence Interval for
(I)
(J)
movie movie
group group

Mean
Difference
(I-J)

Difference
Std. Error

Sig.

a

a

Lower Bound Upper Bound

1

2

-9.149

3.041

.004

-15.204

-3.094

2

1

9.149

3.041

.004

3.094

15.204

a

Adjustment for multiple comparisons: Bonferroni.

128

Cognitive place image dimension 1: tourism attraction

First, a median-split analysis was performed and the audience was divided into two
categories: highly transported audience and weakly transported audience. Then two
ANCOVA tests were conducted with movie group as the independent variable, the
measurement of the earthquake’s impact at time 2 as the covariate, and cognitive image
(tourism attraction) measured at time 2 as the dependent variable.
The output (Table 45) shows that movie transportation’s effect was enduring and it
still could weaken movie genre’s effect after 4 weeks. In particular, for the highly transported
audience, the difference between the violent crime movie group (mean=5.14) and the
romantic drama group (mean=5.17) was not significant, F(1, 71) =.02, p=.90. However, for
the weakly transported audience, the difference between the violent crime movie group
(mean= 4.24) and the romantic drama group (mean=4.99) was significant, F(1, 77) =7.27,
p=.009.
Table 45: Transportation’s Long-Term Moderation Effect on
Cognitive Image-Tourism Attraction

Highly Transported Audience
Between-Subjects Factors
Movie Group

N

1. Violent Crime

30

2. Romantic Drama

44

129

Table 45 (cont’d)
Tests of Between-Subjects Effects
Dependent Variable: Cognitive Image (Tourism Attraction)
Type III Sum
of Squares

Source
Corrected Model

2.202

Intercept

df

Mean Square
2

a

F

1.101

Sig.

.895

.413

551.076 447.823

.000

551.076

1

.019

1

.019

.015

.902

2.073

1

2.073

1.685

.198

Error

87.370

71

1.231

Total

2058.938

74

89.572

73

Movie Group
Earthquake

Corrected Total
a

R Squared = .025 (Adjusted R Squared = -.003)

Estimates
Dependent Variable: Cognitive Image (Tourism Attraction)
Movie
Group
1
2
a

95% Confidence Interval
Mean
5.139
5.172

Std. Error

Lower Bound

Upper Bound

a

.204

4.733

5.546

a

.168

4.837

5.507

Covariates appearing in the model are evaluated at the following

values: Earthquake’s Impact = 2.43.

Weakly Transported Audience
Between-Subjects Factors
Movie Group

N

1. Violent Crime

41

2. Romantic Drama

39

130

Table 45 (cont’d)
Tests of Between-Subjects Effects
Dependent Variable: Cognitive Image (Tourism Attraction)
Type III Sum
of Squares

Source
Corrected Model

15.306

Intercept

df

Mean Square
2

a

7.653

F

Sig.

4.824

.011

381.555 240.491

.000

381.555

1

11.536

1

11.536

7.271

.009

2.727

1

2.727

1.719

.194

Error

122.166

77

1.587

Total

1834.875

80

137.472

79

Movie Group
Earthquake

Corrected Total
a

R Squared = .111 (Adjusted R Squared = .088)

Estimates
Dependent Variable: Cognitive Image (Tourism Attraction)
Movie
Group
1
2
a

95% Confidence Interval
Mean
4.235
4.997

Std. Error

Lower Bound

Upper Bound

a

.197

3.842

4.627

a

.202

4.595

5.399

Covariates appearing in the model are evaluated at the following

values: Earthquake’s Impact= 2.05.

Cognitive place image dimension 2: community quality

First, a median-split analysis was performed and the audience was divided into two
categories: highly transported audience and weakly transported audience. Then two
ANCOVA tests were conducted with movie group as the independent variable, the
measurement of the earthquake’s impact at time 2 as the covariate, and cognitive image
131

(community quality) measured at time 2 as the dependent variable.
The output (Table 46) shows that movie transportation’s effect was enduring and it still
could weaken movie genre’s effect after 4 weeks. In particular, for the highly transported
audience, the difference between the violent crime movie group (mean=4.27) and the
romantic drama group (mean=4.44) was not significant, F(1, 71)=.47, p=.49. However, for
the weakly transported audience, the difference between the violent crime movie group
(mean=3.48) and the romantic drama group (mean=4.21) was significant, F(1, 77) =9.25,
p=.003.
Table 46: Transportation’s Long-Term Moderation Effect on Community Quality
Highly Transported Audience
Between-Subjects Factors
Movie Group

N

1. Violent Crime

30

2. Romantic Drama

44

Tests of Between-Subjects Effects
Dependent Variable: Cognitive Image (Community Quality)

Source
Corrected Model
Intercept

Type III Sum
of Squares
9.131

df

Mean Square
2

a

4.565

F

Sig.

4.108

.020

447.735 402.906

.000

447.735

1

.525

1

.525

.473

.494

7.785

1

7.785

7.006

.010

Error

78.900

71

1.111

Total

1502.250

74

88.030

73

Movie Group
Earthquake

Corrected Total
a

R Squared = .104 (Adjusted R Squared = .078)
132

Table 46 (cont’d)
Estimates
Dependent Variable: Cognitive Image (Community Quality)
Movie
Group
1
2
a

95% Confidence Interval
Mean

Std. Error

4.268
4.442

Lower Bound

Upper Bound

a

.194

3.882

4.655

a

.160

4.124

4.760

Covariates appearing in the model are evaluated at the following

values: Earthquake’s Impact = 2.43.
Weakly Transported Audience
Between-Subjects Factors
Movie Group

N

1. Violent Crime

41

2. Romantic Drama

39

Tests of Between-Subjects Effects
Dependent Variable: Cognitive Image (Community Quality)
Source

Type III Sum
of Squares

df

Mean Square

F

Sig.

Corrected Model

14.605a

2

7.302

6.416

.003

Intercept

254.413

1

254.413

223.530

.000

10.529

1

10.529

9.251

.003

3.031

1

3.031

2.663

.107

Error

87.638

77

1.138

Total

1278.438

80

102.243

79

Movie Group
Earthquake

Corrected Total
a

R Squared = .143 (Adjusted R Squared = .121)

133

Table 46 (cont’d)
Estimates
Dependent Variable: Cognitive Image (Community Quality)
95% Confidence Interval
Movie Group
1
2
a

Mean

Std. Error Lower Bound

Upper Bound

a

.167

3.147

3.812

a

.171

3.867

4.548

3.479
4.208

Covariates appearing in the model are evaluated at the following

values: Earthquake’s Impact = 2.05.
Pairwise Comparisons
Dependent Variable: Cognitive Image (Community Quality)
95% Confidence Interval for
(I)
(J)
movie movie
group group

Mean
Difference
(I-J)

Difference
Std. Error

Sig.

a

a

Lower Bound Upper Bound

1

2

-.728

.239

.003

-1.205

-.251

2

1

.728

.239

.003

.251

1.205

Based on estimated marginal means
a

Adjustment for multiple comparisons: Bonferroni.

134

Discussion and Conclusion
The primary purpose of the study was to investigate how entertainment movies can
influence viewers’ perceptions of depicted places and subsequent visitation interest. What
distinguishes this study from other research conducted on movie-induced tourism is that it
focused simultaneously on two movies about the same place and considered both instant
effect and long-term effect. Moreover, this study also estimated two psychological
mechanisms by which movies can potentially influence people’s perceptions, including
meaning transfer and movie transportation.
The key findings of the study can be summarized in three ways: movie genre, movie
transportation, and initial place familiarity could significantly affect the place image
components and visitation interest; movie transportation could significantly moderate movie
genre and initial place familiarity’s effects; and movie genre and movie transportation’s
effects are found to be meaningful both on short-term and long-term bases.
Findings
First of all, H1 and H2 explored whether movie genre, specifically feel-good movie
and dark movie, will have any different influence on peoples’ perceptions. As expected, the
results of H1 indicate that the violent crime movie had significant negative impact on the
viewers’ affective and cognitive place images immediately after the movie exposure.
Interestingly, based on the results of research question 2, this negative impact from the
violent crime movie was not merely a short-term effect. The longitudinal data indicate that
for the viewers exposed to the violent crime movie, their comparatively low ratings of
affective and cognitive place images still held 4 weeks after their initial movie exposure. This

135

means that the movie associated with killings and violence had the potential to substantially
hurt the image of a featured place, even though the audience only had one exposure to this
experience.
On the other hand, some of the findings are contrary to the expectations. The results
of H1 show that a single exposure to the romantic drama did not have any significant positive
impact on the viewers. Basically, the data show that the viewers exposed to the romantic
drama did not have higher image ratings compared with the subjects from the control group.
This indicates that if a romantic drama can have positive influence on the place image as
predicted by theories, it may not always be a one-shot effect, but may need to be reinforced
with multiple exposures on a long-term basis or from a variety of media channels. For this
specific study, another possible explanation lies in Tokyo’s established global reputation.
Since Tokyo’s image is in general positive in the public’s mind, it would be very hard for a
single movie exposure to modify its current state.
As for H2, the results indicate that the violent crime movie had significant negative
impact on the viewers’ visitation interest as predicted by the Adapted Meaning Transfer
Model. This means that the violent crime movie’s influence is powerful enough to alter
people’s immediate travel intentions. On the other hand, the results also demonstrate that the
romantic drama did not generate the expected significant positive impact on the viewers’
visitation interest immediately after the movie exposure. Contrary to the expectation, the
subjects exposed to the romantic drama had less visitation interest than the subjects who did
not watch a movie. This indicates that apart from movie-induced feelings, some other
moderating variables might be more influential on the people’s visitation interest, such as

136

geographical distance, cost of travel, language barriers, and time. The audiences’ vicarious
experience of the target place in the movies might have increased the salience of these factors
in the subjects’ minds and consequently, the movie viewers generated less visitation interest
compared with the subjects from the control group.
Based on the results of H1 and H2, it is apparent that meaning transfer is an essential
psychological mechanism by which movies can influence people’s place perceptions and
visitation interest. The findings confirm Beeton’s (2005) notion that movies do not always
have the same capacity to improve the embedded place images. Particularly, feel-good
movies may not necessarily be able to improve the embedded place image with a single
exposure. It may take additional messages from other media channels to realize the wanted
positive impact. Moreover, what is unique for this study is, it finds that the influence from the
violent crime movie is instant and enduring, but it has a negative impact. In other words,
movies can not only improve embedded place images, but can also hurt them unintentionally.
From the perspective of place marketers, this is an important piece of information to keep in
mind when their marketing programs will involve dark movies.
Secondly, H3 and H4 explored the question of whether movie transportation is also
a key mechanism by which movies can influence place perceptions and visitation interest.
What is special for this study is that it made an attempt to extend the Transportation Theory
into the research field of product placement. According to the literature, the Transportation
Theory was adopted to investigate whether narratives could influence story-relevant beliefs in
overt persuasive messages (e.g. Wang & Calder, 2006; Green & Brock, 2000; Green, 2004;
Escalas, 2004), but had not yet been tested in the subtle persuasive context. This study

137

estimated whether transportation could also influence the audience’s perceptions when the
persuasive message was hidden and indirect.
As expected, the test results indicate that movie transportation had significant
impact on the viewers’ affective place images, cognitive places images, and visitation interest.
As far as the affective place image is concerned, the more the viewers exposed to the violent
crime movie were transported, the more favorable place images they would have. Meanwhile,
for both experiment groups, movie transportation had positive impact on the viewers’
perceptions of tourism attraction and visitation interest. In other words, the test results
demonstrated that the more the movie viewers were immersed in the stories, the more
favorable impressions they would have for the featured tourism sites and consequently the
more interested they were in traveling to the target place.
Considering all the results for H3 and H4, it shows that movie transportation is an
essential mechanism under which movie viewing can enhance persuasion in the context of
tourism. It shows that movie transportation is a key linkage between movie viewing,
perception changes of depicted places, and visitation interest. At the same time, from a
theoretical perspective, it also shows that the Transportation Theory can be successfully
applied to the hidden persuasive contexts in addition to the overt persuasive contexts.
However, the statistical results in this study should be interpreted with caution,
because the R Squares in the multiple regression analyses were generally small. Considering
the total sample size of posttest 1 was 286 and posttest 2 was 245, it means that movie
transportation’s impact on place image and visitation interest was not very strong. This
indicates that the transportation experience with only one exposure to a movie may not be

138

able to change a place’s image or visitation interest dramatically. In other words, if a movie
can have significant impact on place image and visitation interest in reality, it most likely
requires repeated movie exposure or additional reinforcement messages from other media
channels. As a result, future studies can explore the repeated movie exposure effect and the
combined impact from the reinforcement media messages.
Furthermore, research question 1 explored which mechanism, meaning transfer or
movie transportation, will determine if the violent crime movie will hurt the place image or
improve it. The test results demonstrate that movie transportation could to a large degree
weaken movie genre’s influence. Particularly, for the highly transported audience, there were
no significant differences between the movie groups (violent crime vs. romantic drama) in
terms of their perceptions of place pleasantness, tourism attraction, and community quality.
However, significant differences did exist for the audience that was not well transported. This
indicates that the effect of movie transportation is more powerful than the effect of movie
genre. In this sense, violent crime movies may still have the potential to have a positive
impact on the place images if they are extremely successful in storytelling. In other words, if
a large proportion of the audience can be transported, violent crime movies can still enhance
the images of the embedded places.
H5 and H6 explored initial place familiarity’s role in the relationship between movie
watching and place perceptions. The results indicate that initial place familiarity generally
had a significant positive impact on the audiences’ cognitive place images and visitation
interest, regardless of movie genre. Meanwhile it also had a meaningful positive impact on
the affective place images for the viewers who were exposed to the romantic drama. This

139

indicates that when a movie is shot at a relatively unknown location (e.g., a little, exotic
European town for the American audience), it is always a good idea to have a little
introduction of the target place to the movie goers through advertising or other information
channels before their actual movie exposure.
In addition, H7 and H8 tested whether the phenomenon of “optimal familiarity” had
taken any effect on the subjects’ perceptions of place images and visitation interest. The
findings suggest that movie transportation is a significant moderator of initial place
familiarity’s influence on affective images, but not on cognitive images or visitation interest.
Interestingly, when the movie viewers were relatively unfamiliar with the embedded place
before movie exposure, movie transportation and affective place image had a positive
relationship. However, when the movie viewers were relatively familiar with the embedded
place before movie exposure, movie transportation and affective place image’s relationship
became negative. To some degree, this finding supports MacKay and Fesenmaier’s (1997)
notion of “optimal familiarity” in the context of tourism.
Finally, research question 2 explored movies’ long-term impact, which had not been
addressed by the previous studies in the literature. The results indicate that movie genres’
main effect as hypothesized in H1 and H2, movie transportation’s main effect as
hypothesized in H3 and H4, and movie transportation’s moderation effect as tested in
research question 1, are all meaningful on a long term basis.
As for H1 and H2, the violent crime movie did have an enduring negative impact
on the subjects’ affective place images as well as cognitive place images. The ANCOVA tests
show that even after a month, the difference between the violent crime movie and the control

140

group still held. It suggests that if a violent crime movie has adversely affected a place image,
the damage is not temporary. In other words, once something in the violent crime movie goes
wrong (for example, the movie is not successful) or the target audience does not like the story,
the potential negative influence brought about by the movie will not be easily eliminated.
Moreover, the results demonstrate that the movie genre’s effect on people’s visitation
interest is merely a short-term effect. The ANCOVA tests show that the difference between
the experiment groups and the control group at time 1 went away in a month. This means that,
in general, the movies’ impact on people’s behavioral intentions will have to be reinforced by
follow-up messages; otherwise the effect will not be sustained over time.
As for H3 and H4, when multiple regressions were performed again at time 2, the
results show that movie transportation’s main effect still held for the violent crime movie
after a month. Particularly, for both dimensions of affective place image, movie
transportation’s impact at time 1 remained significant at time 2 for the violent crime movie.
As for the perception of community quality, movie transportation’s effect interestingly
emerged at time 2, although it was not significant at time 1. When it comes to tourism
attraction and visitation interest, movie transportation’s impact remained meaningful for the
violent crime movie group, but not for the romantic comedy group. In this sense, we can
conclude that, on a long-term basis, the more the viewers are transported by violent crime
movies, the more favorable place images and more visitation interest they will potentially
exhibit. This indicates that if a violent crime movie can generate a positive impact on
people’s perceptions, its impact can be sustained longer than it can be for a romantic comedy.
Last but not least, the longitudinal data shows that movie transportation’s moderation

141

effect on movie genre can also last for a long time. The ANCOVA tests show that even after 4
weeks, the effect of movie transportation can still weaken the impact of movie genre on place
pleasantness, tourism attraction, and community quality. This means that, movie
transportation is a pivotal mechanism to explain the relationship between movie watching and
tourism.
Practical Implications
The findings from this study not only support and expand upon previous academic
studies on product placement and movie-induced tourism, but also have unique place
branding and destination marketing implications. First, the findings suggest that movie
transportation is the key link between movie watching and tourism. This means that to a large
degree, movies can help marketers to build “stories” in the place and go beyond the
traditional marketing programs which simply add logos, slogans, and other persuasive
messages next to the place names.
From the perspective of branding strategies, it appears that movies have the power to
help marketers enhance place branding from the level of mind-share branding to cultural
branding. According to Holt (2004), conventional mind-share branding creates content that
shapes perceptions by emphasizing consistency in brand identity communication. Consumers
will presumably discard the rhetorical materials once they believe that the communication
was designed to persuade them. In contrast, cultural branding is a strategy that extends
persuasion to myth making. An effective cultural branding strategy can create a storied
product, which has distinctive brand features through which customers experience identity
myths. The findings of this study suggest that movies can tell place-relevant stories that

142

resonate in the audience’s mind for a long time. This means that movies can effectively create
iconic brands and build up the most valuable assets for a place. In this sense, movies are
valuable motivators of mass tourism and are certainly effective media through which
marketers can approach place branding in general.
Second, from a target audience segmentation standpoint, the results supported Bolan
and Williams’ (2008) notion that there is an overlap between the target audience for the
entertainment movie industry and for the tourism industry. This is because, regardless of the
movie genre, the more the movie viewers are transported, the more favorable place images
and visitation interest they will have. In accordance with this, place marketers can predict the
possible tourism segment attracted by a movie and consequently optimize their segmentation
strategy. As social media grows rapidly, place marketers can effectively integrate travel
platforms with movie platforms so as to locate their potential target audience and even create
interrelated place brand communities and movie brand communities.
Moreover, this study shows that different movie genres will have different impacts on
place image and visitation interest, which can be either positive or negative. Since place
marketers usually do not have complete control over the way a place is portrayed in the
movie, it is important to develop or adjust their place brand management strategy depending
on the way the place is projected. For example, if a film is a dark movie, then the place
marketers need to be very careful to evaluate whether it is “appropriate” to invite the movie
to be filmed at their location or have direct sponsorships. This is because this study shows
that if a dark movie is not successful in storytelling, the film may hurt a place’s image both
instantly and over time. Basically, the task for place marketers is not simply to increase the

143

quantities of place embedment, but, more importantly, to find the appropriate movies for the
location.
In addition, as for whether dark movies will attract more visitors or will drive visitors
away, this study indicates that dark movies still can be considered to be part of a marketing
plan if the movie can successfully transport the target audience. The reason being, if a dark
movie can transport the audience, its positive impact on place images and visitation interest
will last longer than a feel-good movie. Also, when a dark movie can successfully transport
the audience, the negative effect from movie genre will be negligible. When a place is
associated with a dark movie, the only concern for the marketers is how much risk they can
take and want to take.
Limitations and Future Studies
This study clearly has its limitations that need to be addressed. First, this study
employed only two movie subgenres to explore the difference between feel-good movies and
dark movies. Since the extent of impacts and the relevant variables vary as the movie genre
changes, future studies could explore the potential difference with other movie subgenres,
such as horror versus romance, or horror versus comedy. According to Buscombe (1995), the
horror genre is usually associated with disturbing content, such as monsters, coffins, teeth,
and castles. It would be interesting to explore whether these genre conventions and codes will
attract or drive away visitors. The findings could substantially advance our understanding of
the relationship between movie viewing and tourism.
Second, the target place for this study is Tokyo, a city which enjoys a relatively
positive reputation in general. It is possible that because of the existing stereotype in the

144

audience’s mind, the study did not find a positive impact from the romantic drama, which
was an expected result. Future studies could choose a less known location among the
audience and a replication of this study could offer insights as to whether feel-good movies
could instantly improve place image.
Moreover, from the methodological perspective, a synthetic experiment was
employed in this study to accomplish the proposed purpose. This comes with inherent
weakness in terms of external validity of the study findings. For example, movie watching in
most cases is done voluntarily to seek entertainment. In this sense, forced movie exposure
under a lab condition might affect the viewers’ degrees of involvement and relevant emotions.
This might be another reason why this study did not find the expected positive impact from
the romantic drama. Therefore, future research could include field experiments conducted in
natural settings, such as commercial movie theaters.
In addition, this study focused on college students, which is only one of the target
segments of the movie and tourism industry. In the real world, movie audiences are more
diversified in terms of age, occupation, and other characteristics. Therefore, future research
could replicate this project among a different target audience, such as parents or retired
seniors. This should extend our understanding of movies’ impact on the other equally
important demographic segments.
Finally, although this study considers movies’ long-term effects, it only had one
follow-up measurement 1 month after the initial movie exposure. In order to have a better
picture of how the movie-induced effect develops over a longer period, future longitudinal
studies are needed. The findings offer practitioners valuable insights for the timing of

145

reinforcement messages. Consequently, they can best sustain the positive impacts from a
movie and ultimately improve the entire place marketing program.

146

APPENDICES

147

APPENDIX 1
Pretest Questions
To establish your condition for this research project, please answer the following
questions:
1.

Please indicate your preference of movies in terms of categories.
Not at all
Neutral
Interested
Action/Adventures
Romance/comedy
Crime Thrillers
Horror
Drama

1
1
1
1
1

2
2
2
2
2

3
3
3
3
3

4
4
4
4
4

5
5
5
5
5

Extremely
Interested
6
6
6
6
6

7
7
7
7
7

2. Please rate the movies you have watched before based on your memory. If you did
not watch a certain movie before, simply choose "Does Not Apply" and move to
the next movie name.
(Scale: 1=Dislike it a lot and 7=Like it a lot)
a. Eat, Love, Pray
b. The Hangover
c. The Departed
d. 127 Hours
e. Kill Bill (vol.1)
f. Sisterhood of the Traveling Pants
g. Fever Pitch
h. The Matador
i. Man on fire
j. Forgetting Sarah Marshall
k. Lost in Translation
l. Into the Wild
3. Approximately, how many movies have you watched since February 1st, 2011?
Please consider all the movies you have watched in the theater, on DVD, and online.
Please fill in a number as accurate as you can.
Approximately ____________ times
4. How many times have you been to any of the following cities in the past 5 years
(2006-2011)? Please check a box that applies to each city.
Rome
None
1 time
2 times or more
Tokyo
None
1 time
2 times or more
Boston
None
1 time
2 times or more
London
None
1 time
2 times or more
148

5. How familiar are you with Rome?

Life style and people
Cultural/historical attractions
Landscapes
Nightlife entertainment

Extremely
Unfamiliar
1
1
1
1

6. How familiar are you with Toyko?
Extremely
Unfamiliar
Life style and people
1
Cultural/historical attractions
1
Landscapes
1
Nightlife entertainment
1
7. How familiar are you with Boston?
Extremely
Unfamiliar
Life style and people
1
Cultural/historical attractions
1
Landscapes
1
Nightlife entertainment
1
8. How familiar are you with London?
Extremely
Unfamiliar
Life style and people
1
Cultural/historical attractions
1
Landscapes
1
Nightlife entertainment
1

Neutral
2
2
2
2

3
3
3
3

4
4
4
4

5
5
5
5

6
6
6
6

Extremely
Familiar
7
7
7
7

6
6
6
6

Extremely
Familiar
7
7
7
7

6
6
6
6

Extremely
Familiar
7
7
7
7

6
6
6
6

Extremely
Familiar
7
7
7
7

Neutral
2
2
2
2

3
3
3
3

4
4
4
4

5
5
5
5

Neutral
2
2
2
2

3
3
3
3

4
4
4
4

5
5
5
5

Neutral
2
2
2
2

3
3
3
3

4
4
4
4

5
5
5
5

9. Approximately, how many domestic (within the continental United States) leisure
trips have you taken in the past 3 years (2008-present)? Domestic leisure trips
refer to overnight trips that are more than 100 miles away from home. Please fill in
a number.
Approximately _____________trips
10. Approximately, how many international (overseas) leisure trips have you taken in
the past 3 years (2008-present)? Please fill in a number.
Approximately _____________trips

149

11. Please check your availability to watch a movie and participate in the study. Check as
many time slots as possible because we have limited time slots and locations. You
will spend about 2.5 hours to 3 hours in a classroom.

Thursday

Friday

1:30pm – 4:30 pm
6:30pm – 9:30pm
Your Name_________
MSU Email _____________
Please create a multiple-digit number for this study below for cash prizes (i.e. 2345, 5688,
etc).
_______________________
Age ________
Sex: (M/F)
Academic Status________________
a. Non-degree
b. Freshman year
c. Sophomore year
d. Junior year
e. Senior year
f. Graduate school
g. Other (Please specify): ______________________
Major _______________________
Nationality___________________
We will contact you by email with regard to the next steps including the time and
location. Thank you very much for your interest.

150

APPENDIX 2: Posttest Questionnaire 1 for the Experimental Groups
Movie Kill Bill
Please read each question carefully before responding. Please answer to the best of your
ability. Simply circle your choice for each question and thank you very much for your
help.
1. Have you seen this movie before?
1. No

2. Yes

2. Please indicate to which degree each adjective reflects your perception of the leader
character “The Bride”.
Neutral
a. Unattractive

1

2

3

4

5

6

7

Attractive

b. Bad

1

2

3

4

5

6

7

Good

c. Irresponsible

1

2

3

4

5

6

7

Responsible

d. Unpleasant

1

2

3

4

5

6

7

Pleasant

3. Please indicate to which degree each adjective reflects your perception of the female
assassin O-Ren Ishii.
Neutral
a. Unattractive

1

2

3

4

5

6

7

Attractive

b. Bad

1

2

3

4

5

6

7

Good

c. Irresponsible

1

2

3

4

5

6

7

Responsible

d. Unpleasant

1

2

3

4

5

6

7

Pleasant

4. Please indicate to which degree each adjective reflects your perception of Tokyo.
Neutral
a. Gloomy

1

2

3

4

5

6

7

Exciting

b. Distressing

1

2

3

4

5

6

7

Relaxing

c. Sleepy

1

2

3

4

5

6

7

Arousing

d. Unpleasant

1

2

3

4

5

6

7

Pleasant

151

5. Please indicate to what extent you have experienced the following conditions while
you were watching the movie.
a. While I was watching the movie, I could easily picture the events in it taking place.
Neutral
Not at all
1
2
3
4
5
6
7
Very much
b. While I was watching the movie, activity going on in the room around me was on my
mind.
Neutral
Not at all
1
2
3
4
5
6
7
Very much
c. I could picture myself in the scene of the events described in the movie.
Neutral
Not at all
1
2
3
4
5
6
7
Very much
d. I was mentally involved in the movie while watching it.
Neutral
Not at all
1
2
3
4
5
6

7

Very much

e. After watching the movie, I found it easy to put it out of my mind.
Neutral
Not at all
1
2
3
4
5
6
7

Very much

f. I wanted to learn how the movie ended.
Not at all

1

2

3

Neutral
4
5

6

7

Very much

Neutral
4
5

6

7

Very much

g. The movie affected me emotionally.
Not at all

1

2

3

h. I found myself thinking of ways the movie could have turned out differently.
Neutral
Not at all
1
2
3
4
5
6
7
Very much
i. I found my mind wandering while watching the movie.
Neutral
Not at all
1
2
3
4
5
6

7

Very much

j. The events in the movie are relevant to my everyday life.
Neutral
Not at all
1
2
3
4
5
6

7

Very much

k. The events in the movie have changed my life.
Neutral
Not at all
1
2
3
4
5

7

Very much

6

l. While watching the movie I had a vivid image of the leader character “The Bride”.
Neutral
Not at all
1
2
3
4
5
6
7
Very much
m. While watching the movie I had a vivid image of the female assassin O-Ren Ishii
from Japan.
Neutral
Not at all
1
2
3
4
5
6
7
Very much

152

6. Please indicate to what degree you agree or disagree with the following statements
about Tokyo after watching the movie.
a. Tokyo has interesting cultural attractions.
Neutral
Strongly disagree
1
2
3
4
5

6

7

Strongly agree

b. Tokyo has interesting historical attractions.
Neutral
Strongly disagree
1
2
3
4
5

6

7

Strongly agree

c. It seems to me that Tokyo does NOT have impressive beautiful natural sceneries.
Neutral
Strongly disagree
1
2
3
4
5
6
7
Strongly agree
d. Quality accommodations are NOT available in Tokyo.
Neutral
Strongly disagree
1
2
3
4
5
6

7

Strongly agree

e. Tokyo has appealing local food (cuisine).
Neutral
Strongly disagree
1
2
3
4
5

7

Strongly agree

6

f. It seems to me that Tokyo’s standards of cleanliness and hygiene are low.
Neutral
Strongly disagree
1
2
3
4
5
6
7
Strongly agree
g. Tokyo offers quality nighttime entertainment.
Neutral
Strongly disagree
1
2
3
4
5

6

7

Strongly agree

h. Reliable local transportation is available in Tokyo.
Neutral
Strongly disagree
1
2
3
4
5

6

7

Strongly agree

i. In general, Tokyo is a safe place to visit.
Neutral
Strongly disagree
1
2
3
4
5

6

7

Strongly agree

j. I think Tokyo’s people are friendly and hospitable.
Neutral
Strongly disagree
1
2
3
4
5

6

7

Strongly agree

6

7

Strongly agree

6

7

Strongly agree

6

7

Strongly agree

k. The climate in Tokyo is good.
Neutral
Strongly disagree
1
2
3
4
5
l. Tokyo has unpolluted/unspoiled environment.
Neutral
Strongly disagree
1
2
3
4
5
m. A trip to Tokyo is good value for the money.
Neutral
Strongly disagree
1
2
3
4
5
153

7. Please indicate to what degree you agree or disagree with the following statements.
a. I tried to understand the characters in the movie by imagining how things look from
their perspective.
Neutral
Strongly disagree
1
2
3
4
5
6
7
Strongly agree
b. I really got involved with the feelings of the characters in the movie.
Neutral
Strongly disagree
1
2
3
4
5
6
7
Strongly agree
c. While watching the movie, I easily put myself in the place of one of the leading
characters.
Neutral
Strongly disagree
1
2
3
4
5
6
7
Strongly agree
d. While watching the movie, I felt as if the characters’ thoughts and feelings were my
own.
Neutral
Strongly disagree
1
2
3
4
5
6
7
Strongly agree
e. While watching the movie, I imagined how I would feel if the events in the story were
happening to me.
Neutral
Strongly disagree
1
2
3
4
5
6
7
Strongly agree
f. While watching the movie, I tried to imagine what the characters were thinking.
Neutral
Strongly disagree
1
2
3
4
5
6
7
Strongly agree
g. I became very involved in what the characters were experiencing throughout the story.
Neutral
Strongly disagree
1
2
3
4
5
6
7
Strongly agree
h. While watching the movie, I experienced many of the same feelings that the
characters portrayed.
Neutral
Strongly disagree
1
2
3
4
5
6
7
Strongly agree

154

8. We would like you to rate how accurately each word below describes Tokyo after
watching the movie. Please be sure that you have given an answer for each word.
1=extremely inaccurate
2= very inaccurate
3=quite inaccurate
4=slightly inaccurate
1
2
Extremely Very
Inaccurate

5=slightly accurate
6=quite accurate
7=very accurate
8=extremely accurate

3
Quite

4
Slightly

5
Slightly

6
Quite

7
8
Very Extremely
Accurate

Pleasant

1

2

3

4

5

6

7

8

Nice

1

2

3

4

5

6

7

8

Pleasing

1

2

3

4

5

6

7

8

Pretty

1

2

3

4

5

6

7

8

Beautiful

1

2

3

4

5

6

7

8

Dissatisfying

1

2

3

4

5

6

7

8

Displeasing

1

2

3

4

5

6

7

8

Repulsive

1

2

3

4

5

6

7

8

Unpleasant

1

2

3

4

5

6

7

8

Uncomfortable

1

2

3

4

5

6

7

8

Intense

1

2

3

4

5

6

7

8

Arousing

1

2

3

4

5

6

7

8

Active

1

2

3

4

5

6

7

8

2

3

4

5

6

7

8

1

Forceful
Alive

1

2

3

4

5

6

7

8

Inactive

1

2

3

4

5

6

7

8

Drowsy

1

2

3

4

5

6

7

8

Idle

1

2

3

4

5

6

7

8

Lazy

1

2

3

4

5

6

7

8

Slow

1

2

3

4

5

6

7

8

155

9. Please answer the following questions based on your knowledge after watching the
movie.
a. In the movie, “The Bride” was attacked in a small wedding chapel in California some
years ago and almost lost her life.
1 . False

2. True

b. The color of the truck “The Bride” drove to escape from the hospital after she woke
up from a coma was green.
1. False

2. True

c. The female assassin O-Ren Ishii’s parents were ruthlessly killed by a gangster boss
when she was still a little girl.
1. False

2. True

d. “The Bride” visited a samurai sword maker somewhere in Japan before she went to
Tokyo to look for O-Ren Ishii.
1. False

2. True

e. The baby of “The Bride” was killed after she was attacked by the assassins in the
small wedding chapel many years ago.
1. False

2. True

10. Please indicate to what degree you agree or disagree with the following statements
after watching the movie.
a. After watching the movie, it is very likely that I am going to travel to Tokyo.
Neutral
Strongly disagree
1
2
3
4
5
6
7
Strongly agree
b. After watching the movie, I would like to travel around Tokyo.
Neutral
Strongly disagree
1
2
3
4
5
6
7

Strongly agree

c. After watching the movie, I would like to travel to Tokyo for my next vacation.
Neutral
Strongly disagree
1
2
3
4
5
6
7
Strongly agree

156

11. Please indicate to what degree you agree or disagree with the following statements.
a. Revenge is a forest and one can get lost in the forest.
Neutral
Strongly disagree
1
2
3
4
5
6

7

Strongly agree

b. Sometimes people have to do something awfully terrible to truly get one’s revenge.
Neutral
Strongly disagree
1
2
3
4
5
6
7
Strongly agree
c. Revenge can make a person a murderer.
Neutral
Strongly disagree
1
2
3
4
5

6

7

Strongly agree

d. Without becoming a murderer, one will probably never be able to give back all the
pain that the enemy gave him or her.
Neutral
Strongly disagree
1
2
3
4
5
6
7
Strongly agree
12. What is your ethnic background?
1. American Indian
2. Black, non-Hispanic
3. White, non-Hispanic

4.
5.
6.

Asian or Pacific Islander
Hispanic
Other (Please specify)
_______________________

13. Which continent is your home country located? (e.g. France is located in Europe)
1. Asia
2. Africa
3. South America

4.
5.
6.

Europe
North America
Australia

14. While you were answering the questions from 1 to 13, did any images of the recent
earthquake in Japan appear in your mind? (Note: After you read this question, please
do not go back and change any of your answers to question 1 to 13. Simply keep your
existing answers as they are. )
1. No
2. Yes
15. While you were answering the questions from 1 to 13, did you think about the recent
earthquake happened in Japan? (Note: After you read this question, please do not go
back and change any of your answers to question 1 to 13. Simply keep your existing
answers as they are. )
1. No
2. Yes

157

16. Please indicate to what degree you agree or disagree with the following statements.
(Note: After you read this question, please do not go back and change any of your
answers to question 4, 6, 8, or 10. Simply keep your existing answers as they are. )
a. The recent earthquake happened in Japan has negatively influenced my answers to
question 4.
Neutral
Strongly disagree
1
2
3
4
5
6
7
Strongly agree

b. The recent earthquake happened in Japan has negatively influenced my answers to
question 6.
Neutral
Strongly disagree
1
2
3
4
5
6
7
Strongly agree

c. The recent earthquake happened in Japan has negatively influenced my answers to
question 8.
Neutral
Strongly disagree
1
2
3
4
5
6
7
Strongly agree

d. The recent earthquake happened in Japan has negatively influenced my answers to
question 10.
Neutral
Strongly disagree
1
2
3
4
5
6
7
Strongly agree

17. Please write down your full name (for extra credit and the chance to win gift card).
Name __________________________________________
18. Please indicate from which course you will receive course extra credit (e.g. ADV 205,
etc).
_______________________________________________
19. Please write down your MSU email address (for the 5-minute follow up online survey).
MSU Email __________________________________________
Thank you for your time and consideration.
The study is still going on and please do not share your answers with your friends.

158

Movie: Lost in Translation
Please read each question carefully before responding. Please answer to the best of your
ability. Simply circle your choice for each question and thank you very much for your
help.
1. Have you seen this movie before?
1. No

2. Yes

2. Please indicate to which degree each adjective reflects your perception of Bob
Harris.
Neutral
a. Unattractive

1

2

3

4

5

6

7

Attractive

b. Bad

1

2

3

4

5

6

7

Good

c. Irresponsible

1

2

3

4

5

6

7

Responsible

d. Unpleasant

1

2

3

4

5

6

7

Pleasant

3. Please indicate to which degree each adjective reflects your perception of
Charlotte.
Neutral
a. Unattractive

1

2

3

4

5

6

7

Attractive

b. Bad

1

2

3

4

5

6

7

Good

c. Irresponsible

1

2

3

4

5

6

7

Responsible

d. Unpleasant

1

2

3

4

5

6

7

Pleasant

4. Please indicate to which degree each adjective reflects your perception of Tokyo.
Neutral
a. Gloomy

1

2

3

4

5

6

7

Exciting

b. Distressing

1

2

3

4

5

6

7

Relaxing

c. Sleepy

1

2

3

4

5

6

7

Arousing

d. Unpleasant

1

2

3

4

5

6

7

Pleasant

159

5. Please indicate to what extent you have experienced the following conditions while
you were watching the movie.
a. While I was watching the movie, I could easily picture the events in it taking place.
Neutral
Not at all
1
2
3
4
5
6
7
Very much
b. While I was watching the movie, activity going on in the room around me was on my
mind.
Neutral
Not at all
1
2
3
4
5
6
7
Very much
c. I could picture myself in the scene of the events described in the movie.
Neutral
Not at all
1
2
3
4
5
6
7
Very much
d. I was mentally involved in the movie while watching it.
Neutral
Not at all
1
2
3
4
5
6

7

Very much

e. After watching the movie, I found it easy to put it out of my mind.
Neutral
Not at all
1
2
3
4
5
6
7

Very much

f. I wanted to learn how the movie ended.
Neutral
Not at all
1
2
3
4

5

6

7

Very much

5

6

7

Very much

g. The movie affected me emotionally.
Not at all

1

2

3

Neutral
4

h. I found myself thinking of ways the movie could have turned out differently.
Neutral
Not at all
1
2
3
4
5
6
7
Very much
i. I found my mind wandering while watching the movie.
Neutral
Not at all
1
2
3
4
5
6

7

Very much

j. The events in the movie are relevant to my everyday life.
Neutral
Not at all
1
2
3
4
5
6

7

Very much

k. The events in the movie have changed my life.
Neutral
Not at all
1
2
3
4

7

Very much

5

6

l. While watching the movie I had a vivid image of the leader character Bob Harris.
Neutral
Not at all
1
2
3
4
5
6
7
Very much
m. While watching the movie I had a vivid image of the leader character Charlotte.
Neutral
Not at all
1
2
3
4
5
6
7
Very much

160

6. Please indicate to what degree you agree or disagree with the following statements
about Tokyo after watching the movie.
a. Tokyo has interesting cultural attractions.
Neutral
Strongly disagree
1
2
3
4
5

6

7

Strongly agree

b. Tokyo has interesting historical attractions.
Neutral
Strongly disagree
1
2
3
4
5

6

7

Strongly agree

c. It seems to me that Tokyo does NOT have impressive beautiful natural sceneries.
Neutral
Strongly disagree
1
2
3
4
5
6
7
Strongly agree
d. Quality accommodations are NOT available in Tokyo.
Neutral
Strongly disagree
1
2
3
4
5
6

7

Strongly agree

e. Tokyo has appealing local food (cuisine).
Neutral
Strongly disagree
1
2
3
4
5

7

Strongly agree

6

f. It seems to me that Tokyo’s standards of cleanliness and hygiene are low.
Neutral
Strongly disagree
1
2
3
4
5
6
7
Strongly agree
g. Tokyo offers quality nighttime entertainment.
Neutral
Strongly disagree
1
2
3
4
5

6

7

Strongly agree

h. Reliable local transportation is available in Tokyo.
Neutral
Strongly disagree
1
2
3
4
5

6

7

Strongly agree

i. In general, Tokyo is a safe place to visit.
Neutral
Strongly disagree
1
2
3
4
5

6

7

Strongly agree

6

7

Strongly agree

6

7

Strongly agree

6

7

Strongly agree

6

7

Strongly agree

j. I think Tokyo’s people are friendly and hospitable.
Neutral
Strongly disagree
1
2
3
4
5
The climate in Tokyo is good.
Neutral
Strongly disagree
1
2
3
4
5
k. Tokyo has unpolluted/unspoiled environment.
Neutral
Strongly disagree
1
2
3
4
5
l. A trip to Tokyo is good value for the money.
Neutral
Strongly disagree
1
2
3
4
5
161

7. Please indicate to what degree you agree or disagree with the following
statements.
a. I tried to understand the characters in the movie by imagining how things look from
their perspective.
Neutral
Strongly disagree
1
2
3
4
5
6
7
Strongly agree
b. I really got involved with the feelings of the characters in the movie.
Neutral
Strongly disagree
1
2
3
4
5
6
7
Strongly agree
c. While watching the movie, I easily put myself in the place of one of the leading
characters.
Neutral
Strongly disagree
1
2
3
4
5
6
7
Strongly agree
d. While watching the movie, I felt as if the characters’ thoughts and feelings were my
own.
Neutral
Strongly disagree
1
2
3
4
5
6
7
Strongly agree
e. While watching the movie, I imagined how I would feel if the events in the story were
happening to me.
Neutral
Strongly disagree
1
2
3
4
5
6
7
Strongly agree
f. While watching the movie, I tried to imagine what the characters were thinking.
Neutral
Strongly disagree
1
2
3
4
5
6
7
Strongly agree
g. I became very involved in what the characters were experiencing throughout the story.
Neutral
Strongly disagree
1
2
3
4
5
6
7
Strongly agree
h. While watching the movie, I experienced many of the same feelings that the
characters portrayed.
Neutral
Strongly disagree
1
2
3
4
5
6
7
Strongly agree

162

8. We would like you to rate how accurately each word below describes Tokyo after
watching the movie. Please be sure that you have given an answer for each word.
1=extremely inaccurate
2= very inaccurate
3=quite inaccurate
4=slightly inaccurate
1
2
Extremely Very
Inaccurate

5=slightly accurate
6=quite accurate
7=very accurate
8=extremely accurate

3
Quite

4
Slightly

5
Slightly

6
Quite

7
8
Very Extremely
Accurate

Pleasant

1

2

3

4

5

6

7

8

Nice

1

2

3

4

5

6

7

8

Pleasing

1

2

3

4

5

6

7

8

Pretty

1

2

3

4

5

6

7

8

Beautiful

1

2

3

4

5

6

7

8

Dissatisfying

1

2

3

4

5

6

7

8

Displeasing

1

2

3

4

5

6

7

8

Repulsive

1

2

3

4

5

6

7

8

Unpleasant

1

2

3

4

5

6

7

8

Uncomfortable

1

2

3

4

5

6

7

8

Intense

1

2

3

4

5

6

7

8

Arousing

1

2

3

4

5

6

7

8

Active

1

2

3

4

5

6

7

8

2

3

4

5

6

7

8

1

Forceful
Alive

1

2

3

4

5

6

7

8

Inactive

1

2

3

4

5

6

7

8

Drowsy

1

2

3

4

5

6

7

8

Idle

1

2

3

4

5

6

7

8

Lazy

1

2

3

4

5

6

7

8

Slow

1

2

3

4

5

6

7

8

163

9. Please answer the following questions based on your knowledge after watching
the movie.
a. Bob Harris went to Tokyo to make a commercial for an ice tea brand.
1. False

2. True

b. Bob Harris appeared on a Japanese talk show when he was in Tokyo.
1. False

2. True

c. Charlotte invited Bob Harris to her friends’ party in Tokyo when her husband was
away for work.
1. False

2. True

d. Bob Harris kissed Charlotte in the street when he was on his way to the airport at the
end of the movie.
1. False

2. True

e. Charlotte and Bob Harris together visited a Japanese shrine in Kyoto.
1. False

2. True

10. Please indicate to what degree you agree or disagree with the following statements
after watching the movie.
a. After watching the movie, it is very likely that I am going to travel to Tokyo.
Neutral
Strongly disagree
1
2
3
4
5
6
7
Strongly agree
b. After watching the movie, I would like to travel around Tokyo.
Neutral
Strongly disagree
1
2
3
4
5
6
7

Strongly agree

c. After watching the movie, I would like to travel to Tokyo for my next vacation.
Neutral
Strongly disagree
1
2
3
4
5
6
7
Strongly agree

164

11. Please indicate to what degree you agree or disagree with the following
statements.
a. Marriage usually confuses a lot of people.
Neutral
Strongly disagree
1
2
3
4
5

6

7

Strongly agree

b. Marriage is an old concept and is no longer workable in today’s complex world.
Neutral
Strongly disagree
1
2
3
4
5
6
7
Strongly agree
c. The longevity of marriage in America today is much shorter than before.
Neutral
Strongly disagree
1
2
3
4
5
6
7
Strongly agree
d. Marriage should only be for people who are ready to spend the rest of their lives
together.
Neutral
Strongly disagree
1
2
3
4
5
6
7
Strongly agree
12. What is your ethnic background?
1. American Indian
2. Black, non-Hispanic
3. White, non-Hispanic

4.
5.
6.

13. Which continent is your home country located?
1. Asia
2. Africa
3. South America

4.
5.
6.

Asian or Pacific Islander
Hispanic
Other (Please specify)
_______________________
(e.g. France is located in Europe)

Europe
North America
Australia

14. While you were answering the questions from 1 to 13, did any images of the
recent earthquake in Japan appear in your mind? (Note: After you read this
question, please do not go back and change any of your answers to question 1 to 13.
Simply keep your existing answers as they are. )
a. No

b. Yes

15. While you were answering the questions from 1 to 13, did you think about the
recent earthquake happened in Japan? (Note: After you read this question, please
do not go back and change any of your answers to question 1 to 13. Simply keep your
existing answers as they are. )
a. No

b. Yes

165

16. Please indicate to what degree you agree or disagree with the following
statements.
(Note: After you read this question, please do not go back and change any of your
answers to question 4, 6, 8, or 10. Simply keep your existing answers as they are. )
a. The Recent earthquake happened in Japan has negatively influenced my answers to
question 4.
Neutral
Strongly disagree
1
2
3
4
5
6
7
Strongly agree
b. The recent earthquake happened in Japan has negatively influenced my answers to
question 6.
Neutral
Strongly disagree
1
2
3
4
5
6
7
Strongly agree
c. The recent earthquake happened in Japan has negatively influenced my answers to
question 8.
Neutral
Strongly disagree
1
2
3
4
5
6
7
Strongly agree
d. The recent earthquake happened in Japan has negatively influenced my answers to
question 10.
Neutral
Strongly disagree
1
2
3
4
5
6
7
Strongly agree

17. Please write down your full name (for extra credit and the chance to win gift card).
Name __________________________________________
18. Please indicate from which course you will receive course extra credit (e.g. ADV
205, etc).
________________________________________________
19. Please write down your MSU email address (for the 5-minute follow up online
survey).
MSU Email __________________________________________
Thank you for your time and consideration.
The study is still going on and please do not share your answers with your friends.

166

APPENDIX 3:
Control Group Questionnaire
Every year, the GfK Roper Public Affairs & Media will announce highlights from the
annual Anholt-GfK Roper City Reputation Index Report in early February. This report can
capture consumer’s perceptions of the reputation of 50 major cities worldwide. For example,
Paris ranks as the top city in terms of reputation in 2009, followed by Sydney and New York.
For this study, we would like to know your feelings about three cities, including Detroit,
Tokyo, and Boston. Please read the questions below carefully before responding and answer
to the best of your ability WITHOUT checking any reference or information online.
First, please tell us what you think about the city Detroit.
1. Please indicate to which degree each adjective reflects your perception of Detroit.
a. Gloomy
1 2 3 4 5 6 7
Exciting
b. Distressing
1 2 3 4 5 6 7
Relaxing
c. Sleepy
1 2 3 4 5 6 7
Arousing
d. Unpleasant
1 2 3 4 5 6 7
Pleasant
2. Please indicate to what degree you agree or disagree with the following statements about
Detroit.
a. Detroit has interesting cultural attractions.
Strongly disagree
1 2 3 4 5 6 7
Strongly agree
b. Detroit has interesting historical attractions.
Strongly disagree
1 2 3 4 5 6 7
Strongly agree
c. It seems to me that Detroit does NOT have impressive beautiful natural sceneries.
Strongly disagree
1 2 3 4 5 6 7
Strongly agree
d. Quality accommodations are NOT available in Detroit.
Strongly disagree
1 2 3 4 5 6 7
Strongly agree
e. Detroit has appealing local food (cuisine).
Strongly disagree
1 2 3 4 5 6 7
Strongly agree
f. It seems to me that Detroit’s standards of cleanliness and hygiene are low.
Strongly disagree
1 2 3 4 5 6 7
Strongly agree
g. Detroit offers quality nighttime entertainment.
Strongly disagree
1 2 3 4 5 6 7
Strongly agree
h. Reliable local transportation is available in Detroit.
Strongly disagree
1 2 3 4 5 6 7
Strongly agree
i. In general, Detroit is a safe place to visit.
Strongly disagree
1 2 3 4 5 6 7
Strongly agree
j. I think Detroit’s people are friendly and hospitable.
Strongly disagree
1 2 3 4 5 6 7
Strongly agree
k. The climate in Detroit is good.
Strongly disagree
1 2 3 4 5 6 7
Strongly agree
l. Detroit has unpolluted/unspoiled environment.
Strongly disagree
1 2 3 4 5 6 7
Strongly agree
m. A trip to Detroit is good value for the money.
Strongly disagree
1 2 3 4 5 6 7
Strongly agree
167

3. Please indicate to what degree you agree or disagree with the following statements.
a. It is very likely that I am going to travel to Detroit.
Strongly disagree
1 2 3 4 5 6 7
Strongly agree
b. I would like to travel around Detroit.
Strongly disagree
1 2 3 4 5 6 7
Strongly agree
c. I would like to travel to Detroit for my next vacation.
Strongly disagree
1 2 3 4 5 6 7
Strongly agree
Second, please tell us what you think about the city Tokyo.
4.Please indicate to which degree each adjective reflects your perception of Tokyo.
a. Gloomy
1 2 3 4 5 6 7
Exciting
b. Distressing
1 2 3 4 5 6 7
Relaxing
c. Sleepy
1 2 3 4 5 6 7
Arousing
d. Unpleasant
1 2 3 4 5 6 7
Pleasant
4. Please indicate to what degree you agree or disagree with the following statements
about Tokyo.
a. Tokyo has interesting cultural attractions.
Strongly disagree
1 2 3 4 5 6 7
Strongly agree
b. Tokyo has interesting historical attractions.
Strongly disagree
1 2 3 4 5 6 7
Strongly agree
c. It seems to me that Tokyo does NOT have impressive beautiful natural sceneries.
Strongly disagree
1 2 3 4 5 6 7
Strongly agree
d. Quality accommodations are NOT available in Tokyo.
Strongly disagree
1 2 3 4 5 6 7
Strongly agree
e. Tokyo has appealing local food (cuisine).
Strongly disagree
1 2 3 4 5 6 7
Strongly agree
f. It seems to me that Tokyo’s standards of cleanliness and hygiene are low.
Strongly disagree
1 2 3 4 5 6 7
Strongly agree
g. Tokyo offers quality nighttime entertainment.
Strongly disagree
1 2 3 4 5 6 7
Strongly agree
h. Reliable local transportation is available in Tokyo.
Strongly disagree
1 2 3 4 5 6 7
Strongly agree
i. In general, Tokyo is a safe place to visit.
Strongly disagree
1 2 3 4 5 6 7
Strongly agree
j. I think Tokyo’s people are friendly and hospitable.
Strongly disagree
1 2 3 4 5 6 7
Strongly agree
k. The climate in Tokyo is good.
Strongly disagree
1 2 3 4 5 6 7
Strongly agree
l. Tokyo has unpolluted/unspoiled environment.
Strongly disagree
1 2 3 4 5 6 7
Strongly agree
m. A trip to Tokyo is good value for the money.
Strongly disagree
1 2 3 4 5 6 7
Strongly agree
168

5. We would like you to rate how accurately each word below describes Tokyo. Please
be sure that you have given an answer for each word.
1=extremely inaccurate
5=slightly accurate
2= very inaccurate
6=quite accurate
3=quite inaccurate
7=very accurate
4=slightly inaccurate
8=extremely accurate
1
2
Extremely Very
Inaccurate

3
Quite

4
Slightly

5
Slightly

6
Quite

7
8
Very Extremely
Accurate

Pleasant

1

2

3

4

5

6

7

8

Nice

1

2

3

4

5

6

7

8

Pleasing

1

2

3

4

5

6

7

8

Pretty

1

2

3

4

5

6

7

8

Beautiful

1

2

3

4

5

6

7

8

Dissatisfying

1

2

3

4

5

6

7

8

Displeasing

1

2

3

4

5

6

7

8

Repulsive

1

2

3

4

5

6

7

8

Unpleasant

1

2

3

4

5

6

7

8

Uncomfortable

1

2

3

4

5

6

7

8

Intense

1

2

3

4

5

6

7

8

Arousing

1

2

3

4

5

6

7

8

Active

1

2

3

4

5

6

7

8

Forceful

1

2

3

4

5

6

7

8

Alive

1

2

3

4

5

6

7

8

Inactive

1

2

3

4

5

6

7

8

Drowsy

1

2

3

4

5

6

7

8

Idle

1

2

3

4

5

6

7

8

Lazy

1

2

3

4

5

6

7

8

Slow

1

2

3

4

5

6

7

8

169

6.Please indicate to what degree you agree or disagree with the following statements.
a. It is very likely that I am going to travel to Tokyo.
Strongly disagree
1 2 3 4 5 6 7
Strongly agree
b. I would like to travel around Tokyo.
Strongly disagree
1 2 3 4 5 6 7
Strongly agree
c. I would like to travel to Tokyo for my next vacation.
Strongly disagree
1 2 3 4 5 6 7
Strongly agree
Finally, please tell us what you think about the city Boston.
7. Please indicate to which degree each adjective reflects your perception of Boston.
a. Gloomy
1 2 3 4 5 6 7
Exciting
b. Distressing
1 2 3 4 5 6 7
Relaxing
c. Sleepy
1 2 3 4 5 6 7
Arousing
d. Unpleasant
1 2 3 4 5 6 7
Pleasant
8.Please indicate to what degree you agree or disagree with the following statements about
Boston.
a. Boston has interesting cultural attractions.
Strongly disagree
1 2 3 4 5 6 7
Strongly agree
b. Boston has interesting historical attractions.
Strongly disagree
1 2 3 4 5 6 7
Strongly agree
c. It seems to me that Boston does NOT have impressive beautiful natural sceneries.
Strongly disagree
1 2 3 4 5 6 7
Strongly agree
d. Quality accommodations are NOT available in Boston.
Strongly disagree
1 2 3 4 5 6 7
Strongly agree
e. Boston has appealing local food (cuisine).
Strongly disagree
1 2 3 4 5 6 7
Strongly agree
f. It seems to me that Boston’s standards of cleanliness and hygiene are low.
Strongly disagree
1 2 3 4 5 6 7
Strongly agree
g. Boston offers quality nighttime entertainment.
Strongly disagree
1 2 3 4 5 6 7
Strongly agree
h. Reliable local transportation is available in Boston.
Strongly disagree
1 2 3 4 5 6 7
Strongly agree
i. In general, Boston is a safe place to visit.
Strongly disagree
1 2 3 4 5 6 7
Strongly agree
j. I think Boston’s people are friendly and hospitable.
Strongly disagree
1 2 3 4 5 6 7
Strongly agree
k. The climate in Boston is good.
Strongly disagree
1 2 3 4 5 6 7
Strongly agree
l. Boston has unpolluted/unspoiled environment.
Strongly disagree
1 2 3 4 5 6 7
Strongly agree
m. A trip to Boston is good value for the money.
Strongly disagree
1 2 3 4 5 6 7
Strongly agree
170

9.Please indicate to what degree you agree or disagree with the following statements.
a. It is very likely that I am going to travel to Boston.
Strongly disagree
1 2 3 4 5 6 7
Strongly agree
b. I would like to travel around Boston.
Strongly disagree
1 2 3 4 5 6 7
Strongly agree
c. I would like to travel to Boston for my next vacation.
Strongly disagree
1 2 3 4 5 6 7
Strongly agree
Thanks for your response. Next please complete some questions about you and your
previous experience.
10.How many times have you been to Detroit, Tokyo, and Boston in the past 5 years
(2006-2011)? Please check a box that applies.
Detroit
Tokyo
Boston

None
None
None

1 time
1 time
1 time

2 times or more
2 times or more
2 times or more

11.What is your ethnic background?
a. American Indian
b. Black, non-Hispanic
c. White, non-Hispanic

d.
e.
f.

12.Which continent is your home country located?
a. Asia
b. Africa
c. South America

Asian or Pacific Islander
Hispanic
Other (Please specify)
_______________________
(e.g. France is located in Europe)
d.
e.
f.

Europe
North America
Australia

13.While you were answering the questions about your general impression of Tokyo (i.e.
questions 4 to 7), did any images of the recent earthquake in Japan appear in your mind?
(Note: After you read this question, please do not go back and change any of your answers to
question 4 to 7. Simply keep your existing answers as they are. )
b. No
b. Yes
14.While you were answering the questions about your general impression of Tokyo (i.e.
questions 4 to 7), did you think about the recent earthquake happened in Japan? (Note: After
you read this question, please do not go back and change any of your answers to question 4
to 7. Simply keep your existing answers as they are. )
b. No
b. Yes

171

15.Please indicate to what degree you agree or disagree with the following statements. (Note:
After you read this question, please do not go back and change any of your answers to
question 4 to 7. Simply keep your existing answers as they are. )
a). The recent earthquake happened in Japan has NEGATIVELY influenced my answers
to Question 4 (Question 4 asks about to what extent some adjectives can describe your
general feelings about Tokyo, including gloomy/exciting, distressing/relaxing,
sleepy/arousing, and unpleasant/pleasant).
Strongly disagree

1

2

3

4

5

6

7

Strongly agree

b). The recent earthquake happened in Japan has NEGATIVELY influenced my answers
to Question 5 (Question 5 is about your general impression of Tokyo’s cultural
attractions, landscapes, people, night life, etc.).
Strongly disagree

1

2

3

4

5

6

7

Strongly agree

c). The recent earthquake happened in Japan has NEGATIVELY influenced my answers
to Question 6 (Question 6 asks about how accurate some adjectives can describe Tokyo,
including nice, displeasing, intense, drowsy, interesting, boring, etc.).
Strongly disagree

1

2

3

4

5

6

7

Strongly agree

d). The recent earthquake happened in Japan has NEGATIVELY influenced my answers
to Question 7 (Question 7 asks about your general interest to visit Tokyo in person).
Strongly disagree

1

2

3

4

5

6

7

Strongly agree

16.Did you check any references or relevant information online while answering the
questions?
a. No
b. Yes
17.Please write down your full name (for extra credit and the chance to win gift card).
Name __________________________________________

18.Please indicate from which course you will receive course extra credit (e.g. ADV 205,
etc). ___________
19.Please write down your MSU email address (for the 5-minute follow up online survey).
MSU Email __________________________________________
Thank you for your time and consideration.
The study is still going on and please do not share your answers with your friends.

172

APPENDIX 4:
Posttest Questionnaire 2
Please read each question carefully before responding. Please answer to the best of your
ability. Thank you for your help.
1. Please indicate to what degree you agree or disagree with the following statements
about Tokyo.
a. Tokyo has interesting cultural attractions.
Strongly disagree
1 2 3 4 5 6 7
Strongly agree
b. Tokyo has interesting historical attractions.
Strongly disagree
1 2 3 4 5 6 7
Strongly agree
c. It seems to me that Tokyo does NOT have impressive beautiful natural sceneries.
Strongly disagree
1 2 3 4 5 6 7
Strongly agree
d. Quality accommodations are NOT available in Tokyo.
Strongly disagree
1 2 3 4 5 6 7
Strongly agree
e. Tokyo has appealing local food (cuisine).
Strongly disagree
1 2 3 4 5 6 7
Strongly agree
f. It seems to me that Tokyo’s standards of cleanliness and hygiene are low.
Strongly disagree
1 2 3 4 5 6 7
Strongly agree
g. Tokyo offers quality nighttime entertainment.
Strongly disagree
1 2 3 4 5 6 7
Strongly agree
h. Reliable local transportation is available in Tokyo.
Strongly disagree
1 2 3 4 5 6 7
Strongly agree
i. In general, Tokyo is a safe place to visit.
Strongly disagree
1 2 3 4 5 6 7
Strongly agree
j. I think Tokyo’s people are friendly and hospitable.
Strongly disagree
1 2 3 4 5 6 7
Strongly agree
k. The climate in Tokyo is good.
Strongly disagree
1 2 3 4 5 6 7
Strongly agree
l. Tokyo has unpolluted/unspoiled environment.
Strongly disagree
1 2 3 4 5 6 7
Strongly agree
m. A trip to city Tokyo is good value for the money.
Strongly disagree
1 2 3 4 5 6 7
Strongly agree

173

2. We would like you to rate how accurately each word below describes Tokyo. Please
be sure that you have given an answer for each word.
1=extremely inaccurate
5=slightly accurate
2= very inaccurate
6=quite accurate
3=quite inaccurate
7=very accurate
4=slightly inaccurate
8=extremely accurate
1
2
3
4
Extremely Very
Quite
Slightly
Inaccurate

5
Slightly

6
Quite

7
8
Very Extremely
Accurate

Pleasant

1

2

3

4

5

6

7

8

Nice

1

2

3

4

5

6

7

8

Pleasing

1

2

3

4

5

6

7

8

Pretty

1

2

3

4

5

6

7

8

Beautiful

1

2

3

4

5

6

7

8

Dissatisfying

1

2

3

4

5

6

7

8

Displeasing

1

2

3

4

5

6

7

8

Repulsive

1

2

3

4

5

6

7

8

Unpleasant

1

2

3

4

5

6

7

8

Uncomfortable

1

2

3

4

5

6

7

8

Intense

1

2

3

4

5

6

7

8

Arousing

1

2

3

4

5

6

7

8

Active

1

2

3

4

5

6

7

8

Forceful

1

2

3

4

5

6

7

8

Alive

1

2

3

4

5

6

7

8

Inactive

1

2

3

4

5

6

7

8

Drowsy

1

2

3

4

5

6

7

8

Idle

1

2

3

4

5

6

7

8

Lazy

1

2

3

4

5

6

7

8

Slow

1

2

3

4

5

6

7

8

174

3. Please indicate to what degree you agree or disagree with the following statements.
a. It is very likely that I am going to travel to Tokyo.
Strongly disagree
1 2 3 4 5 6 7
Strongly agree
b. I would like to travel around Tokyo.
Strongly disagree
1 2 3 4 5 6 7
Strongly agree
c. I would like to travel to Tokyo for my next vacation.
Strongly disagree
1 2 3 4 5 6 7
Strongly agree
4.While you were answering the questions about your general impression of Tokyo (i.e.
questions 1 to 3), did any images of the recent earthquake in Japan appear in your mind?
(Note: After you read this question, please do not go back and change any of your answers to
question 1 to 3. Simply keep your existing answers as they are. )
a. No
b. Yes

5. While you were answering the questions about your general impression of Tokyo (i.e.
questions 1 to 3), did you think about the recent earthquake happened in Japan? (Note: After
you read this question, please do not go back and change any of your answers to question 1 to
3. Simply keep your existing answers as they are. )
a. No
b. Yes
6.Please indicate to what degree you agree or disagree with the following statements. (Note:
After you read this question, please do not go back and change any of your answers to
question 1 to 3. Simply keep your existing answers as they are. )
a). The recent earthquake happened in Japan has NEGATIVELY influenced my answers
to Question 1 (Question 1 is about your general impression of Tokyo’s cultural
attractions, landscapes, people, night life, etc.).
Strongly disagree

1

2

3

4

5

6

7

Strongly agree

b). The recent earthquake happened in Japan has NEGATIVELY influenced my answers
to Question 2 (Question 2 asks about how accurate some adjectives can describe Tokyo,
including nice, displeasing, intense, drowsy, etc.).
Strongly disagree

1

2

3

4

5

6

7

Strongly agree

c). The recent earthquake happened in Japan has NEGATIVELY influenced my answers
to Question 3 (Question 3 asks about your general interest to visit Tokyo in person).
Strongly disagree

1

2

3

4

5

175

6

7

Strongly agree

7. Did you check any references or relevant information online while answering the
questions?
a. No
b. Yes
8. Please write down your full name (for extra credit and gift card).
Name __________________________________________
9.Please indicate from which course you will receive course extra credit (e.g. ADV 205, etc).
_______________________________________________
10.Please write down your MSU email address (for extra credit and gift card).
Email __________________________________________
Thank you for your time and consideration.

176

REFERENCES

177

REFERENCES

Alcaniz, E.B., Sanchez, I.S., & Blas, S.S. (2009). The functional-psychological continuum in
the cognitive image of a destination: A confirmatory analysis. Tourism Management,
30, 715-723.
Altman, R. (1996). Cinema and genre. In G. Nowell-Smith (Eds.), The Oxford history of
world cinema (pp.276-285). Oxford: Oxford University Press.
Anand, P., Holbrook, M. B., & Stephens, D. (1988). The formation of affective judgments:
The cognitive-affective model versus the independence hypothesis. Journal of
Consumer Research, 15 (3), 386-391.
Anholt, S. (2002). Nation branding: A continuing theme. Journal of Brand Management,
10 (1), 59-60.
Anholt, S. (2006). Anholt city brand index-How the world views its cities. Bellevue, WA:
Global Market Insight.
Anholt, S. (2007). Competitive identity: The new brand management for nations, cities and
regions. New York: Palgrave Macmilan.
Anholt, S. (2010). Places: Identiy, image and reputation. New York: Palgrave MacMillan.
Ashworth, G. & Kavaratzis, M. (2007). Beyond the logo: Brand management for cities.
Brand Management, 16 (8), 520-531.
Babin, L.A., & Carder, S.T. (1996). Viewers’ recognition of brands placed within a film.
International Journal of Advertising, 15, 140-151.
Baker, B. (2007). Destination branding for small cities. Portland, OR: Creative Leap Books.
Balasubramanian, S.K. (1994). Beyond advertising and publicity: Hybrid messages and
public policy issues. Journal of Advertising, 23 (4), 29-47.
Baloglu, S., & Brinberg, D. (1997). Affective images of tourism destinations. Journal of
Travel Research, 35(4), 11-15.
Baloglu, S. & McClearly, K. (1999). A model of destination image formation. Annals of
Tourism Research, 26, 868-897.
Beeton, S. (2005). Film-induced tourism. Clevedon, UK: Channel View Publications.

178

Bolan, P., & Williams, L. (2008). The role of image in service promotion: Focusing on the
influence of film on consumer choice within tourism. International Journal of
Consumer Studies, 32, 382-290.
Boller, G. W., & Olson, J.C. (1991). Experiencing ad meanings: crucial aspects of
narrative/drama processing. Advances in Consumer Research, 18, 172-175.
Buscombe, E. (1995). The idea of genre in the American cinema. In B.K. Grant (Eds.), Film
genre reader II (pp.3-10). Austin: University of Texas Press.
Campbell, D., & Stanley, J. (1963). Experimental and quasi-experimental designs for
research. Chicago: Houghton Mifflin.
Cardy, T. (2006, January 24). Tourists still lured by rings magic. The Dominion Post, p.4.
Celsi, R.L., & Olson, J.C. (1980). The role of involvement in attention and comprehensive
process. Journal of Consumer Research, 15, 210-224.
Chen, C. F. & Tsai, D.C. (2007). How destination image and evaluative factors affects
behavioral intentions? Tourism Management, 28, 1115-1122.
Crane, D. (1998, February 7). Canada should shout about its assets. The Toronto Star, p. C2.
Crompton, J. (1979). An assessment of the image of mexico as a vacation destination and
the influence of geographical location upon that image. Journal of Travel Research,
17, 18-24.
Dann, G. (1996). Tourists’ images of a destination: An alternative analysis. Tourism
Marketing Research, 5 (1/2), 41-55.
Dal Cin, S., Zanna, M.P., & Fong, G. T. (2004). Narrative persuasion and overcoming
resistance. In E.S. Knowles & J.A. Linn (Eds.), Resistance and persuasion (pp.
175-191). Mahwah, NJ: Erlbaum.
Deighton, J., Romer, D., & McQueen, J. (1989). Using drama to persuade. Journal of
Consumer Research, 16 (December), 335-343.
Echtner, C.M. & Ritchie, J.R. B. (1993). The measurement of destination image: An
empirical assessment. Journal of Travel Research, 31(4), 3-13.
Escalas, J.E. (2004). Imagine yourself in the product: Mental simulation, narrative
transportation, and persuasion. Journal of Advertising, 33(2), 37-48.
Florek, M., Insch, A., & Gnoth, J. (2006). City council websites as means of place brand
179

identity communication. Place Branding, 2(4), 276-296.
Fowler, A. (1982). An introduction to the theory of genres and modes. Cambridge: Harvard
University Press.
Gartner, W. (1993). Image formation process. Journal of Travel and Tourism Marketing, 2,
191-216.
Grant, B.K. (Eds.). (1995). Film genre reader II. Austin: University of Texas Press.
Green, M. C., & Brock, T. C. (2000). The role of transportation in the persuasiveness of
public narratives. Journal of Personality and Social Psychology, 79, 701–721.
Green, M.C., & Brock, T. C. (2002). In the mind's eye: Transportation-imagery model of
narrative persuasion. In M. C.Green, J. J. Strange, & T. C. Brock (Eds.), Narrative
impact: Social and cognitive foundations. Mahwah, NJ: Erlbaum, 315–341.
Green, M.C., Brock, T.C., & Kaufman, G.F. (2004). Understanding media enjoyment: The
role of transportation into narrative worlds. Communication Theory, 14(4), 311-327.
Green, M.C., Kass S., Carrey J., Feeney, R., Herzig, B., & Sabini, J. (2008). Transportation
across media: Repeated exposure to print and film. Media Psychology, 11(4),
512-539.
Gupta, P. B., & Lord, K.L. (1998). Product placement in movies: The effect of prominence
And mode on audience recall. Journal of Current Issues and Research in Advertising,
20 (1), 47-59.
Harris Interactive. (2002). College students spend $200 billion each year [online web page].
Retrieved from: http://
www.harrisinteractive.com/news/allnewsbydate.asp?NewsID=480.
Ham, P. (2001). The rise of the brand state: The postmodern politics of image and
reputation. Foreign Affairs, 80(5), 2-6.
Hampson, R. (2010, August 9). Seeking tourists, states try to recast their image; “pure
michigan”? think coastline, not urban decay. USA Today, p. 1A.
Holt, D. B. (2004). How brands become icons: The principles of cultural branding. U.S.:
Harvard Business School Publishing Corporation.
Hu, Y., & Ritchie, J.R.B. (1993). Measuring destination attractiveness: A contextual
approach. Journal of Travel Research, 32, 25-34.

180

Hudson, S., & Ritchie, J.R.B. (2006). Promoting destinations via film tourism: An empirical
identification of supporting marketing initiatives. Journal of Travel Research, 44,
387-396.
Hunt, J. D. (1975). Image as a factor in tourism development. Journal of Travel Research,
13(Winter), 1-7.
Iwashita, C. (2008). Roles of films and television dramas in international tourism: The case
of Japanese tourists to the UK. Journal of Travel and Tourism Marketing, 23(2/3),
139-151.
Johnson, B. T., & Eagly, A. H. (1989). Effects of involvement on persuasion: a meta-analysis.
Psychological Bulletin, 106(2), 290-314.
Jacobsen, B.P. (2009). Investor-based place brand equity: A theoretical framework. Journal
of Place Management and Development, 2(1), 70-84.
Jeffrey, J. (2005, November 12). Up the creek. Weekend Australian, p. 7.
Karrh, J.A., McKee, K.B., & Pardun, C. J. (2003). Practitioners’ evolving views on product
placement effectiveness. Journal of Advertising Research, 43(2), 138-149.
Kavaratzis, M. (2005). Place branding: A review of trends and conceptual models. The
Marketing Review, 5, 329-342.
Kim, H., & Richardson, S.L. (2003). Motion picture impacts on destination images. Annals of
Tourism Research, 30 (1), 216-237.
Kotler, P., Asplund, C., Rein, I., & Heider, D. (1999). Marketing places Europe: Attracting
investments, industries, residents and visitors to European cities, communities,
regions and nations. London: Pearson Education Ltd.
Krugman, H.E. (1965). The impact of television advertising: Learning without involvement.
Public Opinion Quarterly, 30, 349-356.
Langford, B. (2005). Film genre: Hollywood and beyond. Edinburgh: Edinburgh University
Press.
Lee, J. (2009, November 23). Four agencies still standing in brand Australia contest. Sydney
Morning Herald, p.7.
MacKay, K., & Fesenmaier, D. (1997). Pictorial element of destination in image formation.
Annals of Tourism Research, 24, 537-565.

181

Marco, D. D. (2003, January 9). A zeal for the rings. The Washington Times.
Morgan, N., & Pritchard, A. (1998). Tourism promotion and power: Creating images,
creating Identities. Chichester, UK: John Wiley & Sons.
Morgan, N., Pritchard, A., & Pride, R. (Eds.). (2002). Destination branding: Creating the
unique destination proposition. Oxford: Butterworth-Heinemann.
Olsen, J., McAlexander, J., & Roberts, S. (1986). The impact of the visual content of
advertisements upon the perceived vacation experience. In W. Joseph, L. Mautinho
& I. Vernon (Eds.) (pp.260-269), Tourism services marketing: Advances in theory
and practice. Cleveland: Cleveland State University.
Ong, B.S., & Mri, D. (1994). Should product placement in movies be banned? Journal of
Promotion Management, 2, 159-175.
Petty, R. E., & Cacioppo, J.T. (1986a). Communication and persuasion: Central and
peripheral routes to attitude change. New York: Springer-Verlag.
Petty, R. E., & Cacioppo, J.T. (1986b). The elaboration likelihood model of persuasion.
Advances in Experimental Social Psychology, 19, 123-205.
Pike, S. & Ryan, C. (2004). Destination positioning analysis through a comparison of
cognitive, affective, and conative perceptions. Journal of Travel Research, 42,
333-342.
Preston, C. L. (2000). Hanging on a star: The resurrection of the romance film in the
1990s. In W. W. Dixon (Eds.), Film genre 2000: New critical essays. Albany: State
University of New York Press.
Puto, C. P., & Wells, W. D. (1984). Informational and transformational advertising: The
differential effects of time. In T.C.Kinnear (Eds.), Advances in consumer research,
(Vol.11, pp.572-576). Provo, UT: Association for Consumer Research.
Putrevu, S., & Lord, K. R. (1994). Comparative and noncomparative advertising: Attitudinal
effects under cognitive and affective involvement conditions. Journal of Advertising,
23 (June), 77-90.
Rainisto, S. K. (2003). Success factors of place marketing: A study of place marketing
practices in northern Europe and the United States. (Unpublished doctoral
dissertation). Helsinki University of Technology, Institute of Strategy and
International Business.
Riley, R., Baker, D., & Van Doren, C.S. (1998). Movie induced tourism. Annals of Tourism
182

Research, 25 (4), 919-935.
Riley, R., & Van Doren, C.S. (1992). Movies as tourism promotion: A pull factor in a push
location. Tourism Management, 13 (3), 267-274.
Russell, C.A. (1998). Toward a framework of product placement: Theoretical propositions.
Advances in Consumer Research, 25, 357-362.
Russel, J., & Pratt, G. (1980). A description of affective quality attributed to environment.
Journal of Personality and Social Psychology, 38, 311-322.
Russell, C.A., & Puto, C. P. (1999). Rethinking television audience measures: An exploration
into the construct of audience connectedness. Marketing Letters, 10 (4), 387-401.
Russell, C.A., & Stern, B. B. (2006). Consumers, characters, and products: A balance model
of sitcom product placement effects. Journal of Advertising, 35 (1), 7-21.
Schofield, P. (1996). Cinematographic images of a city. Tourism Management, 17(5),
333-340.
Shani, A., Wang, Y., Hudson, S., & Gil., S. M. (2009). Impacts of a historical film on the
destination image of South America. Journal of Vacation Marketing. 15(3), 229-242.
Slater, M.D. (2002). Involvement as goal-directed strategic processing: Extended the
elaboration likelihood model. In J.P. Dillard & M. Pfau (Eds.), The persuasion
handbook: Development in theory and practice. London, Sage: 195-211.
Staats, Arthur W. (1996). Behavior and personality: Psychological behaviorism. New York:
Springer Publishing Company, Inc.
Tabachnick, B. G., & Fidell, L.S. (2007). Using multivariate statistics (5th ed.). Boston:
Pearson Educaiton, Inc.
Tang, L. (2009). Destination websites as advertising: An application of elaboration
likelihood model. (Unpublished doctoral dissertation), Purdue University.
Tasci, A. D. A. (2009). Social distance: The missing link in the loop of movies, destination
image, and tourist behavior? Journal of Travel Research, 47 (4), 494-507.
Tetley, S. J. (1997). Visitor attitudes to authenticity at literary and television-related
destinations. On Worldwide hospitality and tourism trends [CD], WHATT, HCIMA.
Tooke, N., & Baker, M. (1996). Seeing is believing: The effect of film on visitor numbers to
screened locations. Tourism Management, 17(2), 87-94.
183

Tudor, A. (1995). Genre. In B.K. Grant (Eds.), Film genre reader II (pp.3-10). Austin:
University of Texas Press.
Urry, J. (1990). The tourist gaze: Leisure and travel in contemporary societies. London: Sage
Publications.
Verrier, R. (2010, February 24). Global movie ticket receipts rise in 2010; growth in Latin
America and the Asia pacific region help push sales up 8%. Lost Angeles Times, p. 3.
Vollmers, S., & Mizerski, R. (1994). A review and investigation into the effectiveness of
product placements in films. In K.W. King (Eds.), Proceedings of the 1994
conference of the American Academy of Advertising (pp.97-102). Athens, GA:
American Academy of Advertising.
Wang, J., & Calder, B.J. (2006). Media transportation and advertising. Journal of Consumer
Research, 33, 151-162.
Wellek, R. & Warren, A. (1956). Theory of literature (3rd ed.). New York: Harcourt, Brace
and World, p. 260.
Wilson, R. (2000). The left-handed form of human endeavor: Crime films during the 1990s.
In W. W. Dixon (Eds.), Film genre 2000: New critical essays. Albany: State
University of New York Press.
Yueksel, A. & Akguel, O. (2007). Postcards as affective image makers: An idle agent in
destination marketing, Tourism Management, 28(3), 714-725.
Zaichkowsky, J. L. (1985). Measuring the involvement construct. Journal of Consumer
Research, 12, 341-352.
Zaichkowsky, J. L. (1986). Conceptualizing involvement. Journal of Advertising, 15 (2),
4-34.
Zenker, S. & Braun, E. (2010). Branding a city: A conceptual approach for place branding
and place brand management. Paper presented at the 39th European Marketing
Academy Conference: 1-4 June. Copenhagen. Denmark.

184