Scholarly article on topic 'News framing effects on destination risk perception'

News framing effects on destination risk perception Academic research paper on "Economics and business"

CC BY
0
0
Share paper
Academic journal
Tourism Management
OECD Field of science
Keywords
{"Perceived risk" / "Media effects" / "Emphasis framing" / Terrorism / "Political instability"}

Abstract of research paper on Economics and business, author of scientific article — Grzegorz Kapuściński, Barry Richards

Abstract News coverage of hazards is often commented to be of critical importance to individuals' perceived risk associated with tourist destinations. Despite the significance of this issue to the global tourism industry, the link between portrayals of hazards and audience reception is rarely studied in this context. This study adopted the framing theory to evaluate media effect on tourists' perceived risk of portrayals of terrorism and political instability incidents. This involved a survey-embedded experiment which manipulated potential elements of a news report concerning a hazard. The content of fictitious articles used in the experiment was created on the basis of extant risk perception theories. Results revealed that the use of risk amplifying frame and risk attenuating frame result in higher and lower ratings of risk respectively. Moreover, tourist psychographic characteristics were found to moderate the influence of news frames on perceived risk. Implications for tourism destination managers and marketers were discussed.

Academic research paper on topic "News framing effects on destination risk perception"

Contents lists available at ScienceDirect

Tourism Management

journal homepage: www.elsevier.com/locate/tourman

News framing effects on destination risk perception

Grzegorz Kapuscinski a *, Barry Richards b

a Faculty of Management, Bournemouth University, Fern Barrow, Poole, Dorset, BH12 5BB England, United Kingdom b Faculty of Media and Communication, Bournemouth University, Fern Barrow, Poole, Dorset, BH12 5BB England, United Kingdom

CrossMark

HIGHLIGHTS

• Application of media framing theory to study effects on perceived risk.

• Exposure to different risk frames results in different levels of perceived risk.

• Influence of risk frames depends on tourists' psychographic characteristics.

ARTICLE INFO

ABSTRACT

Article history: Received 10 December 2015 Received in revised form 1 June 2016 Accepted 20 June 2016 Available online 30 June 2016

Keywords: Perceived risk Media effects Emphasis framing Terrorism Political instability

News coverage of hazards is often commented to be of critical importance to individuals' perceived risk associated with tourist destinations. Despite the significance of this issue to the global tourism industry, the link between portrayals of hazards and audience reception is rarely studied in this context. This study adopted the framing theory to evaluate media effect on tourists' perceived risk of portrayals of terrorism and political instability incidents. This involved a survey-embedded experiment which manipulated potential elements of a news report concerning a hazard. The content of fictitious articles used in the experiment was created on the basis of extant risk perception theories. Results revealed that the use of risk amplifying frame and risk attenuating frame result in higher and lower ratings of risk respectively. Moreover, tourist psychographic characteristics were found to moderate the influence of news frames on perceived risk. Implications for tourism destination managers and marketers were discussed.

© 2016 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY license

(http://creativecommons.org/licenses/by/4.0/).

1. Introduction

The media coverage of hazards is fundamental as to the way in which some hazards are perceived as risky and some are trivialised (Flynn, Slovic, & Kunreuther, 2001; Kasperson, 2005; Petts, Horlick-Jones, & Murdock, 2001 ). Lack of personal experience with hazards such as terrorism or political instability (hereafter 'PI') increases audiences' reliance on secondary sources of information. Among these, the news media are often commented to be a particularly important source of risk perceptions of hazards and the tourist destinations in which they take place. While providing people with crucial information in a timely manner, news media coverage of hazards is commonly believed to be associated with producing distorted understanding of safety levels at destinations (Chew & Jahari, 2014; Larsen, Brun, Torvald, & Selstad, 2011; L'Etang, Falkheimer, & Lugo, 2007). Consequently, this is of importance to

* Corresponding author. E-mail addresses: GKapuscinski@Bournemouth.ac.uk (G. Kapuscinski), BRichards@Bournemouth.ac.uk (B. Richards).

Destination Management Organisations (DMOs) who wish to minimise negative impacts of such coverage by understanding how potential tourists make sense and act upon this information and devising their own communication strategies to convey a more balanced view of the situation (Baxter & Bowen, 2004; Ritchie, Dorrell, Miller, & Miller, 2004).

Media scholars note that key to understanding the potential media effect on perception of risk is the way in which hazardous events are framed (Hove, Paek, Y., &Jwa, 2015; Hughes, Kitzinger, & Murdock, 2006; Marks, Kalaitzandonakes, Wilkins, & Zakharova, 2007). A number of studies demonstrate cases of mass media coverage of hazards which puts emphasis on some aspects of hazards to the exclusion of others (Daye, 2014; Spencer & Triche, 1994; Woods, 2007). It is widely agreed that the aspects of an issue highlighted by the speaker can influence the way audiences understand and interpret the issues covered (Price & Tewksbury, 1997). Tourism and hospitality studies that employ framing theory demonstrate that message frames can influence hotel booking intentions and trust (Sparks & Browning, 2011) and destination attractiveness (Min, Martin, & Jung, 2013). However, despite the

http://dx.doi.org/10.1016/j.tourman.2016.06.017

0261-5177/© 2016 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/40/).

agreement among tourism scholars that news framing can cause destinations to appear safer or more dangerous (Hall, 2002), there is a paucity of empirical studies on tourists' responses to different portrayals of risk (risk frames). The need for a study in this area is reinforced by a recent call on framing effects research in tourism by Liu and Pennington-Gray (2015).

Beyond considering the influence of different ways of portraying hazards, framing effects scholars note that it is equally important to consider characteristics of the audiences (Lecheler, de Vreese, & Slothuus, 2009). This is especially relevant in the context of perceived risk, which is subjective and depends upon a broad range of tourist characteristics. Among these, the degree of allocentricity has seldom been used to understand variance in perceived risk. Plog (1974) model of allocentricity identifies tourists' personality along a continuum from allocentrics to psychocentrics. The allo-centric type is described by Plog (2001; 2002) as more outgoing, adventuresome and exploring, and self-confident than the psy-chocentric traveller.

Guided by framing effects theory and existing knowledge of perceived risk in tourism, this study sought to understand whether different media frames concerning hazards influence tourists' judgment of risk. Specifically, this study employed an experimental design to explore the impact of different portrayals of terrorism and PI hazards. Moreover, tourist characteristics of age, gender, and degree of allocentricity were considered as potential moderators of the media effect. The findings should provide needed empirical support for the often implied yet rarely tested influence of media messages on tourists' perception of risk. This issue is of significance from a theoretical standpoint and for the practice of communication strategies that help to recover problematic destination images.

2. Literature review

2.1. Risk perception

The concept of perceived risk, or subjective risk, in social scientific literature is commonly agreed to mean 'the processing of physical signals and/or information about potentially harmful events or activities, and the formation of a judgement about seriousness, likelihood and acceptability of the respective event or activity' (Slovic, Fischhoff, & Lichtenstein, 1982 cited by Grobe et al. 2008, p. 16; Brehmer, 1987; Renn, 2004). Proponents of the psychometric paradigm of risk (Fischhoff, Slovic, Lichtenstein, Read, & Combs, 1978) demonstrate that when evaluating risks, people are influenced by a host of qualitative risk features (see Table 1), or simplifying heuristics, which produce subjective biases and misjudgements of formal (quantitative) risk assessments (Fischhoff et al., 1978; Slovic et al., 1982). For example, the public is more concerned about involuntary risks (Slovic, 1987) and risk in affect-rich contexts such as terrorist attacks, than about less spectacular but much more probable ones such as heart disease (Lowenstein,

Table 1

Qualitative aspects attenuating or amplifying perceived risk.

Attenuate perceived risk Amplify perceived risk

Familiar Exotic/New

Controllable Uncontrollable

Limited effects Catastrophic effects

Natural Man-made

Fair impact distribution Unfair impact distribution

Clear benefits No clear benefits

Voluntary Imposed

Positive affect Negative affect

Consequences not-fatal Consequences fatal

Adapted from: Renn (2008).

Hsee, Weber, & Welch, 2001; Sjoberg, 2007).

2.2. Risk perception in tourism: terrorism and political instability

In tourism, risk perception is a function of uncertainty and consequences (Moutinho, 2000), with some consequences being more desirable to tourists than others. Although risk may entail positive and negative outcomes, it is often studied as undesirable outcomes, such as loss of time or money, that may arise from consuming tourism products, for example, trekking in unfamiliar environments. Risks identified include threats to health (e.g. food poisoning), and terrorism or PI, e.g. in hazards such as coups, kidnappings, bombings, and street disturbances. Terrorism and PI are often commented to be particularly intimidating risks due to the uncontrollable, involuntary and random nature of the potential harm involved in visiting destinations struck by such incidents (Cavlek, 2002; Heng, 2006). This is supported by Gray and Wilson (2009) who find that political hazards such as terrorism are perceived as riskier than other physical threats (e.g. weather) and social hazards (e.g. hostile local people). This may be partly attributed to the emotional charge carried by such events, which is further amplified by the man-made nature of harm involved, as opposed to acts of nature. One of the main consequences of man-made disasters is that, apart from the physical damage, the biggest impact is often felt on the psychological level (Jenkin, 2006; Schmid, 2005).

While related, PI and terrorism have a separate conceptual identity, which is supported by a number of authors who examine the phenomena from this perspective (e.g. Enders & Sandler, 1998; Richter & Waugh, 1986; Saha & Yap, 2013; Sonmez, 1998). Numerous and diverse definitions have been proposed over the years. In an attempt to address the call from scholars for terminological unity and conceptual clarity (Gupta, 1990; Seddighi, Theocharous, & Nuttall, 2002), Tcheocharous (2010) reviews and evaluates existing approaches to studying PI and proposes a definition which encompasses the key features of the phenomenon. He describes PI as "a situation where a political system is subjected to challenges or changes in the form of internal conflict, internal change and external conflict. The extent/level of instability is determined by the deviation of any given political event (or a combination of events) from the specific normal pattern of the system in which it occurs." (Tcheocharous, 2010, p. 358). PI is related to terrorism in the sense that the latter can be an indicator and an expression of the former (Sonmez, 1998). This said, it is not uncommon for terrorist attacks to take place in politically stable destinations (e.g. Bali, 2005; London 2005; Marrakech, 2010), which is one of the differences between these phenomena. Others may be identified on the basis of a definition of terrorism proposed by Schmid and Jongman (1988, p. 28) who suggest that "Terrorism is an anxiety-inspiring method of repeated violent action ... whereby the direct targets of violence are not the main targets. The immediate human victims of violence are generally chosen randomly (targets of opportunity) or selectively (representative or symbolic targets) from a target population, and serve as message generators. Threat and violence-based communication processes between terrorists (organisations), (imperilled) victims, and main targets are used to manipulate the main target (audience)." This definition highlights that terrorism is concerned with the intentional use of violence against carefully selected targets to communicate a message and satisfy specific aims, for example, targeting civilians to generate media attention, moral repugnance in the target population, and behavioural changes undertaken by the affected (e.g. avoiding travel, or certain places) (Rubin, Brewin, Greenberg, Simpson, & Wessely, 2007; Silver, Holman, McIntosh, Poulin, & Gil-Rivas, 2002) as a means to satisfy political or

religiously motivated goals. While media attention may aid the goals of actors such as protestors, for example by generating sympathy amongst international audiences to their goals, its involvement is not the key to achievement of these goals. The dependence of terrorism upon media attention is well captured by the comparison by Ted Koppel, ABC's host, who said: "without television, terrorism becomes rather like the philosopher's hypothetical tree falling in the forest: no one hears it fall and therefore it has no reason for being" (Koppel, quoted in Farnen, 1990, p. 104).

While some researchers note that terrorism and PI are not always a major concern for tourists (Fuchs, Uriely, Reichel, & Maoz, 2013; Hunter-Jones, Jeffs, & Smith, 2007; Rittichainuwat & Chak-raborty, 2009) the literature is rich in cases where demand for tourism products falls as an outcome of these hazards (see Table 2). Whether or not the magnitude of risk involved in visiting a destination is high, people tend to avoid places they perceive as unsafe because in tourism perception is reality when it comes to decision-making (Santana, 2001 ). These studies highlight that understanding how people perceive and respond to these hazards remains a serious concern for the tourism industry.

2.3. Tourist destination, risk management and media effects

In the context of numerous crises that happen around the world, risk management has become a vital aspect of the way businesses operate within the tourism system. Unmanaged, hazardous events can lead to problematic safety images of tourist destinations, which is a major constraint to a thriving tourism industry (Arana & Lenn, 2008; Avraham & Ketter, 2008; George, 2003; Law, 2006; Tarlow, 2014). Therefore, it is key that destinations have comprehensive management strategies in place that assess and respond to risks in a manner that protects the safety and security of visitors and staff, and maximises the potential for continuity of business (APEC, 2006). To address this issue, a number of frameworks and guidelines for crisis and disaster management in tourism have been created (APEC, 2006; Henderson, 2004; Ritchie, 2004; Tourism Victoria, 2010).

Media are key to any risk management strategy. Given the potential to misrepresent the magnitude of events and influence the image of destinations, Faulkner (2001) notes that monitoring of media reporting to assess the nature of information that is sent regarding the event is crucial. Beyond this, scholars stress the need for a proactive relationship with the news media such as provision of information about safety and the security situation at the destination and how it is being managed (Mansfeld, 2006; Ritchie et al., 2004; Wang & Ritchie, 2010). To this end, the literature on crisis and disaster management proposes tourism crisis communication strategies to engage the media in the process of mitigation of negative perceptions and recovery (Avraham & Ketter, 2008,

2016; Beirmann, 2003; Ritchie, 2009). According to Avraham and Ketter (2008), three groups of media strategies should be considered in designing a campaign; these are source, audience, and message strategies. The latter focuses on the message itself and may include tactics such as contradicting the negative messages, perceptions, stereotypes, etc. in reports upon the destination. In essence, such strategy is concerned with a focus on the development ofmessages that correct and neutralise unbalanced portrayals of destinations. Important to these management strategies is the issue of framing and the influence that different representations of hazards may have on the perceived reality of tourist destination safety among audiences in the tourist-generating countries.

2.4. Media coverage — framing effects

In essence, research on framing effects is concerned with the influence of differences in presentation of issues on the attitudes, emotions and decisions of media users (Iyengar, 1991; Pan & Kosicki, 1993; de Vreese, 2005). Framing studies can be separated into two distinct types depending on the kind of framing effect studied (Druckman, 2001b). The first type, concerning 'equivalency' or 'valence' framing effect, investigates how the use of different but logically equivalent words that highlight positive or negative aspects of an issue (e.g. 5% unemployment versus 95% employment) elicits different responses from individuals (e.g. Kahneman & Tversky, 1984). This type of framing research into wording effects (Druckman, 2001a) is an approach not easily applicable to more complex communicative situations (Sniderman & Theriault, 2004) such as terrorism or PI.

Research on framing in relation to tourism effects typically employs equivalence framing, for example studies of tourists' attitudes to environmental protection (Kim & Kim, 2014), hotel linen reuse (Blose, Mack, & Pitts, 2015), and hotel booking intentions in response to online reviews (Sparks & Browning, 2011).

The second type, emphasis framing, is 'closer to "real" journalistic news coverage' (Nelson, Lecheler, Schuck, & Vreese, 2012, p. 3) and is concerned with the influence on individuals of presenting alternative perspectives that stress different aspects of complex topics (e.g. emphasis on advantages over disadvantages of technology) without the assumption that the information is factually equivalent (Tankard, 2001). The focus of this study is to examine tourists' perceptions and attitudes in relation to coverage of the complex hazards of terrorism and PI. This suggests the use of an 'emphasis' framing approach to framing effects.

de Vreese (2005) notes that from the perspective of emphasis framing, frames are generally seen as coherent packages of information containing a 'central organising idea or story line that provides meaning to an unfolding strip of events' (Gamson & Modigliani, 1989: 143). In short, a news frame is an 'emphasis in

Table 2

The effects of terrorism and political instability on tourism.

Destination Effect(s) of incidents

Thailand As an effect of PI the number of tourist arrivals in Thailand went down from 1.7 million in December 2009 to 1.1 million in April 2010.

With security concerns hotel occupancy fell from 60.2% in January 2010 to 46.6% in April 2010 (Euromonitor, 2010)

Egypt Ongoing disturbances and demonstrations throughout Egypt which erupted on January 2011 resulted in a 37% fall in visitor numbers to

reach 9 million compared to over 14 million in 2010 (Euromonitor, 2012)

Nepal The war in Afghanistan initiated in October 2001 had an adverse impact on visitor arrivals to Nepal in 2002,

which fell at an average of 40% in the peak season (JAN-JUN) harming the total economy of the country (Thapa, 2003, p .129, p .129)

India In the five month period following the Mumbai 2008 attacks Foreign Tourist Arrivals to India declined by 10,000 (Bhattacharya & Basu, 2010)

Bali Following the 2005 terrorist attack, tourist arrivals decreased by 5% in 2005 and by 9% in 2006 (Aschauer, 2014, p. 168).

Kenya The bombings of US targets in Nairobi 1998 resulted in 90% of inbound international flights being cancelled, which caused serious harm to

the Kenyan economy (Kuto & Groves, 2004)

USA Following the September 11th 2001 terrorist attacks North American International Air Transport Association recorded a 33% decline in

passenger traffic from European, Far Eastern, and Central and South American carriers (IATA cited in Lee, Oh, & O'Leary, 2005)

salience of different aspects of a topic' (de Vreese, 2005: 53). Through the selection and emphasis on some aspects of a topic above others, journalists present a story to the public within a particular frame of reference (Entman, 2004; Van Gorp, 2007). In this way, certain attributes, judgments and decisions are suggested (Entman, 1993) which helps people classify information to handle it efficiently (Scheufele, 2006). A framing effect is said to occur when, in the course of describing an issue or event, a speaker's emphasis on a subset of potentially relevant considerations causes receivers to focus on these considerations when constructing their opinions or judgements (Sniderman & Theriault, 2004).

Numerous framing effects studies have manipulated news frames through experimentally prepared news stories to determine their influence upon individual responses. These studies demonstrate how variations in the make-up of messages elicit noticeable variations in audiences' perceived risk associated with complex issues such as earth quality (Durfee, 2006), nanotechnology (Schütz & Wiedemann, 2008), climate change (Otieno et al., 2014), and invasive species (Otieno, Spada, & Renkl, 2013). Woods (2011) investigated perceived risk in response to fictitious news stories about terror threat which contained different degrees of emphasis on the level of control over hazard and catastrophic potential (Slovic, 1987). He found that his study subjects perceived more risk of terrorism when the danger was associated with 'radical Islamic groups' (suggestive of perceived lack of control over potential exposure to an attack), and 'nuclear' technology (suggestive of perceived high catastrophic potential of an attack).

2.5. Factors influencing perceived risk in tourism

Tourism literature suggests that perceived risk depends on a range of tourist characteristics. Previous studies found that perceived risk ratings can differ based on factors such as gender (Carr, 2001), age (Floyd & Pennington-Gray, 2004; Kozak, Crotts, & Law, 2007), travel experience (Sonmez & Graefe, 1998a), nationality (Seddighi, Nuttall, & Theocharous, 2001) and personality (Lepp & Gibson, 2003). More recent studies suggest the significance for subjective risk of religion (Adam, 2015; Mansfeld, Jonas, & Cahaner, 2016), knowledge (Sharifpour, Walters, Ritchie, & Winter, 2014), risk tolerance and risk related competences (Williams & Balaz, 2013). Among these, personality traits are often considered to be of particular relevance (Fuchs, 2011; Pizam et al., 2004; Roehl & Fesenmaier, 1992). Perception of risk has been found to depend on degree of novelty seeking (Correia, Pimpao, & Crouch, 2008), self-confidence (Valencia & Crouch, 2008) sensation-seeking (Fuchs, 2011; Lepp & Gibson, 2008; Sharifpour, Walters, & Ritchie, 2013) and multi-trait factors (e.g. extroversion, venturesomeness) (Reisinger & Mavondo, 2005). A variable that has received less attention in this context is the psychographic typology of Plog (1974; 1991), who divided tourism consumers into two broad groups of allocentrics and psychocentrics (with additional near-psychocentric, mid-centric, and near-allocentric groups). According to Plog, psychocentrics are characterised by being anxious within their daily lives, risk averse, and preferring to travel on package tours to familiar and commonplace destinations. Moreover, they tend to visit sun-and-fun destinations which are consistent with their preference for low activity levels (Plog, 2001). In contrast, allocentrics prefer unstructured trips to unusual places and more contact with local cultures. They are more confident, less anxious, motivated by novelty, discovery, and seek active vacations that allow them to explore the physical and cultural worlds around them (Plog, 2002). In reality the situation is a lot more complicated, as most people are mid-centric, that is, they combine elements of both types, and therefore distinct types are difficult to identify.

A number of studies used Plog's system to investigate tourist

preferences, which resulted in some support for this personality factor (Griffith & Albanese, 1996; Nickerson & Ellis, 1991; Weaver, 2012). Others found no association between Plog's personality types and destination preferences (Litvin, 2006; Smith, 1990). While the ability of the instrument to predict destination choices has received much attention, its use in understanding variances in tourists' risk perception in responses to hazards deserves more research.

In light of the outlined gaps in the literature on the relationship between media framing, terrorism, PI and tourism risk perception, the study set out to address the following research questions:

RQ1: Does leisure tourists' perceived risk depend on media frames concerning the magnitude of terrorism and political instability risk (high or low)?

RQ2 Does leisure tourists' perceived risk depend on media frames concerning event type (terrorism or PI)?

RQ3 Does leisure tourists' perceived risk depend on the degree of allocentrism (allocentric, midcentric, or psychocentric).

3. Methods

An online survey-based experiment was employed with a focus on the causal link between news media frames of terrorism and PI, and leisure tourists' perceived risk. A mixed factorial 2 (hazard type: terrorism or PI) x 2 (risk framing: risk amplifying or risk attenuating) between-subject design was employed (see Table 3). The sample of experiment participants was obtained from a questionnaire-survey (N = 475) that preceded the experiment. All respondents identified themselves as leisure tourists. A total of 160 respondents (N = 160) expressed an interest in participating in a follow-up study by providing an e-mail address. Apart from providing a sample for the experiment, the first questionnaire survey, which measured UK leisure tourists' risk perceptions, holiday choices and a range of demographic and psychographic factors, identified tourist characteristics (i.e. degree of allocentricity, age, gender) relevant to the dependent variable (i.e. perceived risk). The personality variable was measured with a scale comprised of 8 items adapted from Jackson and Inbakaran (2006). Each of the items represented a different aspect of the tourist personality, i.e the need for structure, familiarity/novelty, off-the-beaten-track, reliance on the tourism industry, venturesomeness, intellectual curiosity, activity, and openness to other cultures, which was phrased as a statement referring to tourists' holiday preferences. The data were then used to control for relevant variables in the experiment. Following the procedure used in media effects studies (Wimmer & Dominick, 2011), participants were matched on relevant variables (degree of allocentricity, age, gender) to create four groups before being assigned at random to different treatments.

On the 6th of March 2013 the participants (N = 160) were sent an e-mail containing a link to an online survey questionnaire. Each group received a different scenario, i.e. one of Terrorism A, Terrorism B, PI A or PI B. After the initial campaign, 7 e-mails were returned as non-deliverable and identified as non-existing. The

Table 3

Experiment design.

Levels Factor 1 — hazard type Terrorism Political instability (PI)

Factor 2 -Risk (A) Perceived risk Treatment 1 Treatment 3

framing amplifying Terrorism A PIA

(B) Perceived risk Treatment 2 Treatment 4

attenuating Terrorism B PIB

final number of confirmed e-mail deliveries was reduced to 153 (N = 153).

3.1. Stimulus material and measures

The respondents were placed in a scenario of considering a holiday in a non-specific country. Role playing is a commonly adopted method in experimental designs (Petty et al. 1991; cited by Jun & Vogt, 2013). The reason for the de-contextualised scenario was to exclude the influence on dependent variables of the potentially confounding effects of respondents' attitudes and feelings towards the country and prior visit experiences. After reading the scenario the respondents were invited to read a short news article about an incident in this destination country and rate their perceived risk associated with visiting the destination. Perceived risk was measured only post-test to avoid cueing the respondent to the specific goal of the media stimulus. Similar experimental designs have been used in past research on framing effects to avoid sensitising participants (e.g. Lecheler & de Vreese, 2010; Maoz, 2012). No control group was used in the experiment. Each of the four treatments acted as a control group for the other three treatments, as it manipulated message elements that the other treatments did not.

With respect to the stimulus material, the news article was produced in four alternative versions. Following common practice employed in experimental framing studies to separate content from frames (e.g. Iyengar, 1991; Price et al. 1997; Valkatenburg et al. 1997; cited in; Schuck & de Vreese, 2006), versions about terrorism and PI were designed in a fashion that holds the factual base of events constant while manipulating aspects of risk.

With respect to factor 1 (event type), terrorism and PI have been identified in the tourism literature as similar but also distinct phenomena. Consequently, different characteristics associated with these distinct hazardous events were incorporated into two pairs of fictitious articles. Within each pair, the articles differed in the hypothesised direction of their influence on dependent variables of interest. Event descriptions should not be treated as representative of any case of PI or terrorism as these may involve different targets, tactics, locations etc. The goal of manipulation was to increase the likelihood of activation in audiences of different thoughts and feelings associated with these events, and potentially different judgments of risk. To achieve this the event descriptions were designed in a manner that resembles hazardous events that received attention in UK national newspapers and were identified by reporters as either PI or terrorism. With respect to PI, given that data was collected in March 2013, arguably, the most prominent and recent event preceding this date were the January 2011 protests in Egypt. A search on the Lexis Nexis UK database between (28.01.2011 and 28.03.2011) with keywords ('Egypt'; 'Political

Instability', 'Unrest') resulted in 481 hits. On the basis of this content, PI was described as an incident involving protests in a city centre location, human casualties, some potential spread of protests to other locations and no advice from the FCO against travel to the country.

With respect to terrorism, two prominent events preceding the data collection period were the April 2011 bombing in Marrakech and July 2011 triple attack in Mumbai. The following searches were performed to source articles about the events in 1) Marrakech: Keywords ('Marrakech'; 'Terror!') between (28.04.2011 and 28.06.2011) resulted in 22 hits, and 2) Mumbai: Keywords ('Mumbai'; 'Terror!') between (13.7.11 and 13.9.11) resulted in 41 hits. Next, terrorist attack was conceptualised as an incident involving a bomb explosion in a city centre location, with human casualties, unknown perpetrators and no advice from the Foreign and Commonwealth Office (FCO) against travel to the country.

Manipulations of factor 2 (risk framing) were guided by a theoretical assumption that different portrayals of qualitative features of terrorism and PI will have different effects on tourists' risk perception. In this sense, factor 2 represents the message framing with two levels, i.e. frame intended to amplify perceived risk (A), or frame intended to attenuate perceived risk (frame B). Each frame embedded within an article involved manipulation of message elements concerning four qualitative features of terrorism and PI (see Tables 4 and 5). The qualitative dimensions selected for this exploratory experiment that may indicate different levels of magnitude of risk (e.g. level of control) were guided by research on risk perception. What needs to be noted is that the message elements selected are not an exhaustive list of risk indicators that may be used in reporting on terrorism and PI. Rather, they are a set of potential elements of news reports that may influence the level of audiences' perceived risk associated with visiting an affected area. The specificity of the message elements employed within versions A and B, and their expected direction of influence, are explained and justified in the following sections.

In the case of terrorism, the four qualitative dimensions of the hazard that involved different levels of emphasis on information potentially relevant to the judgment of magnitude of risk were (see Table 4): 1) targets of attack, 2) suspected perpetrators, 3) location of explosion and threat of further attacks, and 4) event atmosphere and confidence level.

With respect to targets of attack, the extent to which a hazard receives media attention and possibly affects audiences' perceived risk is partly dependent on who the violence, or its threat, is directed at. In other words, it isn't only about 'body counts' but also about 'whose body counts' (Kitzinger, 2009). Media coverage which puts emphasis on victims that are particularly relevant to audiences, for example, tourists of certain nationality, religion etc., as opposed to 'others' such as foreign military units, may be

Table 4

Message elements used in constructing the articles about a terrorist attack.

Dimension of terrorism risk Scenario A Scenario B

Perceived risk amplifying Perceived risk attenuating

1) Targets of attack 2) Suspected Perpetrators 3) Location of explosion and threat of further attacks 4) VoxPopuli- Event atmosphere and confidence level • "Including British tourists" • "al-Qaeda and associated radical Islamic groups" • "Police vehicles parked in city square situated on the edge of a district full of restaurants, cafes and shops" • "City centre locations" • Security "Airports, train station and markets" • "Further indiscriminate attacks in areas popular with tourists cannot be ruled out" • "I have never seen anything like this and I cannot believe it happened right here. Now people will not have peace of mind" • "Mainly police officers" • "Security Forces" • "Domestic rebel separatist group" • "Police vehicles parked in city square" • Security "Across the country" • "Yes it was a terrorist attack but we refuse to be terrorised. Life here goes on as usual".

Table 5

Message elements used in constructing the articles about an event of political instability.

Dimensions of PI risk

Scenario A

Scenario B

Perceived risk amplifying

Perceived risk attenuating

1) Violence

2) Commentary on degree of socio-political tension VoxPopuli commentary on socio-political tension and consequences (confidence)

3) Geographical spread and consequences

4) Disruptions to transport network

"Violent clashes" "Violent protests"

"Threatening atmosphere of high tension" "I have never seen anything like this, it was complete chaos. We all feel nervous because the problem will not just go away overnight" "There is a possibility that further violent protests could spread"

"which would likely have serious consequences for public safety and order" "in the event of conflict escalation, delays and cancellations cannot be ruled out"

• "Clashes"

• "Protests"

• "Isolated acts of frustration"

• "It was loud at the square but outside life went on as usual. I do not think there

will much trouble, people are just venting anger"

• "Any further protests are likely to be confined to city squares"

• Outside "predicted to remain calm and not affected in any way"

Absence

particularly relevant to judgments of vulnerability and greater personal risk.

Another frame dimension employed in the article is the information on suspected perpetrators. Given that attribution of responsibility is one of the most basic heuristics people employ to make sense of an issue or event (Iyengar, 1991), emphasis on a particular group may activate a schematic representation concerning, for example, typical tactics, targets, or other memorable events and the magnitude of threat. In Britain, as well as globally, it is not uncommon for the media to frame the contemporary threat of terrorism as a problem of Islamic extremism (Alouche & Lind, 2010). Examples include several high profile events of 'new' terrorism such as Bali (2002, 2005), London (2005), or Madrid (2004), where responsibility has been attributed to, or claimed by, al-Qaeda and associated networks. Moreover, according to Pape (2005, cited by Woods, 2011), terrorism framed as motivated by religious extremism, may appear irrational, beyond compromise, and uncontrollable. As a consequence, as postulated by the psychometric risk paradigm (Fischhoff et al., 1978), a lack of control in association with a hazard may lead to higher levels of perceived risk. While there are no strong theoretical bases for an assumption that separatist-nationalist groups such as the Basque Fatherland and Freedom (ETA) would appear less threatening to people, it is possible that the connection between violence and civilian victims (or western victims) in this case is less prominent in peoples' minds. A search on the Global Terrorism Database (GTD) (START, 2012) of incidents between 1980 and 2011 carried out by the IRA and ETA on non-civilian targets (i.e. Police, Military, Government, Utilities) produced 1825 instances, compared to 391 attacks against civilians (including tourists). Using this logic, the article was designed in a way as to suggest the linkage between particular groups and typical targets of attacks. Beyond this, differences in perception may exist mainly on the basis of activation of beliefs about al-Qaeda as an extreme expression of otherness and intolerance to the western way of life, e.g. the freedoms and wealth displayed by tourists.

With regards to location of explosion and threat of further attacks, it is logical to expect that a report of an attack on a 'paradise' beach island resort carries different implications for interpretation of risk than an attack on a police vehicle on the outskirts of a city. The location of an attack may be indicative of motives and tactics that lead to certain conclusions e.g. 'targeting tourists'.

The fourth dimension of risk amplification/attenuation concerns how commentary from the general public (vox populi, or 'vox pops') influences perceptions of event atmosphere and confidence level. Vox pops (or any other news source) can be used to frame an issue by supplying background information or story

suggestions (Matthews, 2010). Drawing on findings from the psychometric tradition of perceived risk, new risks (e.g. unknown technologies) are perceived as involving more risk than ones that audiences are familiar with. Using this logic, version A of the article emphasises the 'newness' of the problem as perceived by the local public, and its negative consequences on individuals' confidence in response to the event. In contrast, version B presents the information from a balanced point of view with an emphasis on the habituation of the members of the public to the problem, and their resilience. As such, the frames represent two distinct ways of reacting to terrorism, i.e. with fear or determination, which may suggest to readers which course of action is most suitable.

In the case of PI, the four dimensions were: 1) violence, 2) commentary on degree of socio-political tensions, 3) geographical spread and consequences, 4) disruptions to transport network (see Table 5).

The first dimension of the frame employed in the article concerned the presence or absence of violence. While the challenge to a political system can be sought through peaceful means such as protests or strikes (Scarbrough, 1998) PI is often associated with violence, whether civil, wars, military coups or riots (Hall & O'Sullivan, 1996). Exposure of audiences to reports of violence may result in risk perceptions disproportionate to the actual probability of harm, a phenomenon Sunstein (2003) refers to as probability neglect. Therefore, in describing the nature of the confrontations, the word "violent" is employed in version A of the article, expecting to add to the concern of the audience, as opposed to its absence in version B.

Next, socio-political tension, seen by Siermann (1998) as a common thread to different expressions of PI, was used as another dimension of PI that can be used by individuals to arrive at risk judgments. However, it can be argued that the ability of tourists to judge personal risk on the basis of the extent of underlying sociopolitical tensions and the ability of the government to maintain social order is likely to be limited due to the complexity of such issues. In this context, a news frame may simplify the process and enable conclusions to be drawn. Therefore, portraying the event as being characterised by a "threatening atmosphere of high tension" in version A versus "isolated acts" in version B may promote different degrees of perceived risk. As such, the degree of social tension as captured in vox pops pertains to the level of control and ability to avoid the potential problems a tourist may have in an affected destination.

In a similar vein, the frame dimension concerning the geographical spread and consequences of the event pertains to the level of control and so has implications for the magnitude of risk perception. Therefore, version A of the article stresses the possibility of unrest spreading to areas across the country, and its

"serious consequences for public safety and order". Conversely, in version 'B', emphasis is placed on a prediction that events will remain "contained to city squares" and other areas will remain "calm and not affected", and is expected to produce lower ratings of risk.

The fourth dimension of the frame employed concerns disruptions to the transport network. It is proposed that an emphasis placed on the possibility of transport "delays and cancellations" in version A of the article would be expected to heighten tourists' risk perception, as opposed to an absence of such emphasis inversion B.

Perceived risk was measured only after participants' exposure to the article treatment on a 5-point Likert scale: from 1 = Very Worried, to 5 = Not at all worried. No pre-test of perceived risk was taken due to a concern over sensitising the participants to the objective of the experiment. Moreover, since the country context was non-specific, i.e. respondents had no perceptions of risk present in the country (or other feelings and attitudes which may have influenced the judgment), one measure was sufficient to observe the impact of the article.

3.2. Participants

A sample of UK leisure tourists was used in the experiment. The final groups that responded to each of four news report scenarios were as follows: Terrorism A = 30, Terrorism B = 34, PI A = 28, PI B = 32. The sample comprised of 60 (48.4%) males and 64 (51.6%) females. The ages ranged from 18 to 65 years old and over, with '45—54' being the largest age category (26.6%), followed by '35—44' (25%), and '25—34' and '55—64' (both 17.7%). '65+' and '18—24' were the smallest age categories with 7.3% and 5.6% respectively. The data reflect the UK population of outbound holidaymakers measured between 2005 and 2009 (ONS, 2011). Using psycho-graphic profiles established on the basis of items suggested by Weaver (2012) and Jackson and Inbakaran (2006), the sample included 21 (16.9%) Psychocentrics, 74 (76.6%) Midcentrics, and 29 (23.4%) Allocentrics. The data were analysed using Kruskal-Wallis and Mann Whitney U tests.

4. Results

4.1. RQ1

A non-parametric Kruskal-Wallis test was run to examine the differences in perceived risk between the groups of respondents exposed to different article versions. The result of the test was significant at p < 0.000 level. Therefore, to test for differences between article pairs A (risk amplifying) and B (risk attenuating), a post-hoc Mann-Whitney test was employed with the confidence level set at p < 0.025. It was hypothesised that the tourists exposed to article version A would perceive more risk than those who had read version B.

The results (Table 6) were significant in both the terrorism and PI pairs, that is, as hypothesised, the risk amplifying version of articles caused significantly higher risk concerns (lower scores) than risk attenuating version B. Therefore, the null hypothesis can be rejected. The variations in the presentation of the different aspects, or indicators, of risk produced an effect on the tourists' responses. The difference in tourists' risk perception between the articles in the PI pair was larger (r = -0.55) than in the terrorism pair (r = -0.31).

4.2. RQ2

A Mann-Whitney test was employed to examine the difference in risk perception between the frames concerning the event type,

Table 6

Results of the Mann-Whitney post-hoc test for perceived risk and magnitude of risk frames.

Perceived risk mean ranking MUW z Asympt. Sig.

Article Terrorism A Terrorism B

groups N = 30 N = 34

26.90 37.44 342.000 -2.463 0.014

PI A PI B

N = 28 N = 32

20.75 39.03 175.000 -4.267 0.000

i.e. PI versus terrorism. The results (see Table 7) indicate a statistically significant difference of low to medium strength (r = 0.337) between terrorism and PI in the B pair (risk attenuating frame). A reverse relationship was expected in the A version of the articles, where a severe version of PI, such as the recent events in Egypt or Syria, might be expected to be more intimidating than a terrorist attack which specifically targets tourists or popular tourist areas. While this may also be complicated by other factors such as, for instance, weapon type, in general a terrorist attack, such as a bombing, shooting or kidnapping, represents a different degree of threat to a large scale event of PI, both in terms of the potential for physical harm, the geographical spread of vulnerable zones and the length of time in which a destination may be vulnerable to such actions. This, however, was not supported by the findings of this study.

4.3. RQ3

A significant result (p < 0.000) was derived from application of a Kruskal-Wallis test comparing the three personality types across all four versions of the article. To follow up on this result, a series of Mann-Whitney tests were performed between the extreme groups of the psychocentrics and allocentrics in each article condition expecting the latter group to score higher (i.e. lower perceived risk). The results (see Table 8) point to differences in the expected direction, but the only significant result was noted with respect to the Terrorism A article (p < 0.005). Interestingly, when the perceived risk of the allocentrics and psychocentrics was compared across all the article groups, a highly significant result (p < 0.001) was obtained. That is, regardless of the article read, the allocentrics perceived less risk (r = -0.48) than their psychocentric counterparts. This suggests that psychocentrics are more sensitive than allocentrics to message elements which suggest a deviation from an acceptable level of risk. This finding is in line with a "cognitive-transactional" model of media effects (Perse, 2001, p. 51) which suggests that the effect of media content may be moderated by audience variables such as schema make-up and specific beliefs and attitudes.

Table 7

Results of the Mann-Whitney post-hoc test for perceived risk and event type frames.

Perceived risk mean ranking MUW z Asympt. Sig.

Article Terrorism A PI A

groups N = 30 N = 28

31.57 27.57 358.000 -1.066 0.286

Terrorism B PIB

N = 34 N = 32

27.90 39.45 353.500 -2.612 0.009

Terrorism PI

N = 58 N = 66

59.43 65.78 1723.500 -1.044 0.297

Table 8

Results of the Mann-Whitney post-hoc test for perceived risk and allo/psychocentric audience types.

Article groups Perceived risk mean ranking MUW Z Asympt sig

Terrorism A Alio N = 7 Psycho N = 5

8.79 3.30 1.500 -2.715 0.005

Terrorism B N = 7 N = 6

8.00 5.83 14.00 -1.096 0.366

PI A N = 5 N = 6

7.25 4.50 7.500 -1.535 0.177

PI B N = 9 N = 5

8.83 5.10 10.500 -1.893 0.058

All articles N = 29 N = 21

904.00 371.00 140.00 -3.413 0.001

5. Discussion

An effect on leisure tourists' risk perception was observed as a result of the exposure of audiences to different frames concerning the magnitude of terrorism/political instability risk. Readers of article versions A (risk amplifying) perceived more risk than those who read versions B (risk attenuating), regarding both terrorism and PI, albeit the differences were larger in the PI group. Specifically, the result demonstrates that the emphasis on some aspects of a source of risk, to the exclusion of others, may result in different risk perceptions associated with visiting a destination among the message recipients. This finding supports the results of framing effects studies carried out outside the tourism consumer behaviour context (e.g. Schuck & de Vreese, 2006; Woods, 2011), and addresses the calls for research on framing effects in tourism (Liu & Pennington-Gray, 2015), and specifically on the relationship between the media coverage of hazards and destination perceived risk (e.g. Schroeder, Pennington-Gray, Kaplanidou, & Zhan, 2013; Sonmez & Graefe, 1998a).

Interestingly, the differences in perceived risk between readers of risk amplifying and attenuating article versions were smaller in the terrorism pair. This suggests that in judging the risk associated with visiting foreign countries for leisure purposes, tourists may be particularly sensitive to any information about terrorism due to the discretionary nature of holiday activity and a limited knowledge of the country. In this context, they may be less motivated to consider incident characteristics, such as the perpetrators and the targets, and instead make a risk judgment on the basis of the fact that an attack has occurred and another one is possible. This explanation is contrary to the findings of the experimental research on terrorism of Woods (2011), which show that differences in the information concerning the perpetrators (i.e. 'Islamic extremists' versus other terrorists) significantly influence perceived risk. However, notably, this research was conducted in a different context, namely the risk of terrorism to US citizens, which may be completely different to assessing the risk to oneself in association with visiting a foreign country. In the case of PI, the larger difference may be due to the fact that while information concerning large-scale unrest likely brings to mind a range of dramatic images (e.g. violent unrest in Egypt, Syria etc.) and scares people, the portrayal of an event as localised and under control reassures people by conveying a picture of reasonably safe conditions.

The risk-attenuating frame had more impact on perceived risk in relation to PI than in relation to terrorism. This suggests that despite variations in event characteristics, a terrorist attack may be a much more deterring factor, at least in the short term, than a case of PI which appears to be limited in scope. No difference in risk perception between terrorism and PI was observed in version A of the article, where the risk-amplifying frame might be expected to have produced higher levels of anxiety in relation to PI. This is

contrary to the findings of Saha and Yap (2013) who find that political events have a more deterring effect on the tourism industry than terrorist attacks. The data of this study may be explained by the perceived random character of terrorist attacks which implies low ability to control the severe consequences, and hence leads to judgements of greater personal risk.

Notably, the study indicates that the risk perceived by respondents post-reading of the article about a hazard depends on their psychographic characteristics. In their study of the influence of Uganda's tourism website on tourists' perceived risk, Lepp, Gibson, and Lane (2011) found no relationship to Cohen's tourist types, a construct related to allocentricity. Here, however, the allocentric recipients were less concerned about terrorism and PI post reading of the article versions than the psychocentrics. This supports the notion that the impact of risk communication depends not only upon the content of the message but also the characteristics of the audience (e.g. Lecheler et al. 2009), in this case, the tourist personality profile.

6. Conclusion

The findings enhance the understanding of the relationship between perceived risk, the media, and tourist consumer behaviour by empirically supporting the validity of the framing theory of media effects (Scheufele, 1999). Drawing on theories of the psychometric paradigm of risk (Slovic, Fischhoff, & Lichtenstein, 1986), the exploratory experiment established that the effects were not uniform across recipients and were contingent upon audience characteristics, most notably tourist personality type. Importantly, this aspect of the media and perceived risk interaction points toward a two-directional relationship, which recognises both the power of the media to influence message recipients and the power of audiences to oppose and negotiate the messages. This research proposes that marketers can influence the way tourists attend to risk messages and evaluate tolerability of risk involved in holidays.

Methodologically, the experimental design employed in this study contributes to the growing body of experimental work in the study of consumer behaviour in tourism, and to our best knowledge is the first experimental study of framing effects on perceived risk in tourism. Jun and Vogt (2013) and Cohen, Prayag, and Moital (2014) note that experimental research designs are seldom used in travel and tourism research, which is dominated by the cross-sectional approach. Despite calls for their use in tourism perceived risk research (e.g. Chew & Jahari, 2014; Roehl & Fesenmaier, 1992), examples of studies that employ such designs are still rare (e.g. Lepp et al. 2011). Future studies that employ an experimental design could investigate the potential effects of media frames in a more dynamic context, i.e. with the use of images and audio. A further area for research would be to adopt a longitudinal approach to studying media effects, rather than a one-shot media exposure approach.

In terms of practical implications, on the one hand, the media effect found in this study poses challenges to tourism marketers who wish to minimise the negative effect of media coverage of hazards such as terrorism and PI. On the other, it stresses the need to further research the formats of presenting the information. Findings indicate that when judging the risk involved in visiting a country subject to a terrorist attack, tourists may be less sensitive to information about the characteristics of the event and draw conclusions mainly on the fact that the event took place. Secondly, the clear disparity in perceived risk between the readers of the two PI articles demonstrates that while tourists are relatively unconcerned about unrest portrayed as contained to small or non-tourist areas, its extreme expression is a source of grave concern. These findings are of importance to communication practices in a post-

disaster context, especially the aftermath of terrorist attacks and PI. Moreover, in recognition of tourist characteristics that determine differences in perceived risk, catering for experienced tourists with allocentric tendencies may be particularly useful for destinations in the post security crisis phase. To this end, destination marketers can tailor their offers to match the needs of this segment by providing novel, active, culturally stimulating and exciting experiences that increase the propensity to rationalise risk.

There are limitations in the research design used. To control for the potential confounding effects in the experiment that tourists' ideas and feelings about a destination may have on the perceived risk, the relationship between the news reports and perceived risk was investigated in a scenario of a hypothetical destination. Whereas a destination that tourists recognise would have been ideal, arguably the choice made by the researcher represents a common scenario where tourists make holiday decisions with imperfect knowledge of destinations. Furthermore, while allowing for the studying of responses of a diverse sample of respondents, the online survey-experiment was also associated with limited control over the participants, which is typically overcome by a laboratory approach. Following procedures employed by Jun and Vogt (2013), to control for the issue of respondents attending to other information during participation, responses from those who spent more than 20 min on the study were excluded from analysis.

With respect to the sample, the study focused on British tourists, which may not be representative of potential responses of other nationalities to similar fictitious stories. Future studies could focus on samples of different nationalities or those that comprise of two or more nationalities. Other questions require more attention. Due to the design employed in this study it is not possible to isolate the influence on perceived risk of, for example, nationality of perpetrators. Future studies could seek to manipulate this aspect of the message to seek to understand whether attacks and/or threat of further attacks on tourists of the same nationality have a stronger effect on perceived risk than those directed at 'others'.

Moreover, given the sample of the study and the fact that a number of recent attacks and/or warnings in major European destinations (e.g. France, Belgium, Germany) were perpetrated by the Islamic State (IS), rather than al Qaeda, it can be argued that the focus in this paper on the latter actor constitutes another limitation of this research. This said, at the time of data collection (March 2013) IS was an unknown entity in Europe, hence why al Qaeda was chosen as a group likely to be recognised by respondents and taken into account in making risk judgments. Future studies could focus on stories concerning other perpetrators and the relative impact of such information on tourists' judgments.

Finally, beyond the message content, the stories presented to respondents were not attributed to any specific news source or presented in different news outlets e.g. newspapers, social media, etc. Future studies could seek to address this by manipulating these variables.

References

Adam, I. (2015). Backpackers' risk perceptions and risk reduction strategies in

Ghana. Tourism Management, 49, 99—108. Alouche, J., & Lind, J. (2010). Public attitudes to global uncertainties :A research synthesis exploring the trends and gaps in knowledge. Institute of Development Studies.

APEC. (2006). Tourism risk management.

Araña, J. E., & León, C. J. (2008). The impact of terrorism on tourism demand. Annals

of Tourism Research, 35(2), 299—315. Aschauer, W. (2014). New approached in the research on terrorist attacks affecting

tourism demand. In H. Andrews (Ed.), Tourism and violence. Routledge. Avraham, E., & Ketter, E. (2008). Media strategies for marketing places in crisis. Improving the image of cities, countries and tourist destinations. Great Britain: Elsevier Inc.

Avraham, E., & Ketter, E. (2016). Tourism marketing for developing countries. Battling

stereotypes and crises in asia, africa and the Middle East, palgrave MacMillan.

Baxter, E., & Bowen, D. (2004). Anatomy of tourism Crisis: Explaining the effects on tourism of the Uk foot and mouth disease epidemics of 1967—68 and 2001 with special reference to media portrayal. International Journal of Tourism Research, 6(4), 263—273.

Beirmann, D. (2003). Restoring tourism destinations in crisis: A strategic marketing approach. United Kingdom: CABI Publishing.

Bhattacharya, M., & Basu, K. (2010). Impact of terror incidents on the foreign tourist arrivals in India: An econometric exploration. International Journal of Tourism Policy, 3(3), 213—222.

Blose, J. E., Mack, R. W., & Pitts, R. E. (2015). The influence of message framing on hotel guests' linen-reuse intentions. Cornell Hospitality Quarterly, 56(2), 145—154.

Brehmer, B. (1987). The Psychology of risk. In W. T. Singleton, & J. Howden (Eds.), Risk and decisions. New York: Wiley.

Carr, N. (2001). An exploratory study of gendered differences in young tourists' perception of danger within London. Tourism Management, 22, 565—570.

Cavlek, N. (2002). Tour operators and destination safety. Annals ofTourism Research, 29(2), 478—496.

Chew, E. Y. T., & Jahari, S. A. (2014). Destination image as a mediator between perceived risks and revisit intention: A case of post-disaster Japan. Tourism Management, 40, 382—393.

Cohen, S. A., Prayag, G., & Moital, M. (2014). Consumer behaviour in tourism: Concepts, influences and opportunities. Current Issues in Tourism, 17(10), 872—909.

Correia, A., Pimpao, A., & Crouch, G. I. (2008). Perceived risk and novelty-seeking behaviour: The case of tourists on low-cost travel in Algarve (Portugal). In A. Woodside (Ed.), Advances in culture, tourism, and hospitality research. Sage.

Daye, M. (2014). Framing tourist risk in Uk press accounts of Hurricane Ivan. Place Branding and Public Diplomacy, 10,186—198.

Druckman, J. N. (2001a). On the limits of framing effects: Who can frame? The Journal of Politics, 63(4), 1041—1066.

Druckman, J. N. (2001b). The implications of framing effects for citizen competence. Political Behavior, 23(3), 225—255.

Durfee, J. L. (2006). 'Social change' and 'status Quo' framing effects on risk perception : An exploratory experiment. Science Communication, 27, 459—495.

Enders, W., & Sandler, T. (1998). Transnational terrorism in the Post-Cold war era. Iowa State University, USA: Department of Economics.

Entman, R. (1993). Framing: Toward clarification of a fractured paradigm. Journal of Communication, 43(4), 51—58.

Entman, R. (2004). Projections of power: Framing news, public opinion, and U.S. Foreign policy. University of Chicago Press.

Euromonitor. (2010). Political instability in Thailand affects ASEAN countries euromonitor international [online]. Available from: http://blog.euromonitor.com/ 2010/07/political-instability-in-thailand-affects-asean-countries.html [Accessed 25 May 2012].

Euromonitor. (2012). Travel and tourism in Egypt. Available from: http://www. euromonitor.com/travel-and-tourism-in-egypt/report [Accessed 25 May 2012].

Farnen, R. F. (1990). Decoding the mass media and terrorism Connection: Militatnt extremism As systemic and symbiotic processes. In C. R. Sanders (Ed.), Marginal Conventions: Popular culture, mass media and social deviance. Ohio: Bowling Green State University Press.

Faulkner, B. (2001). Towards a framework for tourism disaster management. Tourism Management, 22(2), 135—147.

Fischhoff, B., Slovic, P., Lichtenstein, S., Read, S., & Combs, B. (1978). How safe is safe enough? A psychometric study of attitudes towards technological risks and benefits. Policy Sciences, 9,127—152.

Floyd, M. F., & Pennington-Gray, L. (2004). Profiling risk: Perception of tourists. Annals of Tourism Research, 31(4), 1051—1054.

Flynn, J., Slovic, P., & Kunreuther, H. (2001). Risk, media and Stigma: Understanding public challenges to modern science and technology. London: Earthscan.

Fuchs, G. (2011). Low versus high sensation-seeking tourists: A study of backpackers' experience risk perception. International Journal of Tourism Research, 15(1), 81—92.

Fuchs, G., Uriely, N., Reichel, A., & Maoz, D. (2013). Vacationing in a terror-stricken Destination: Tourists' risk perceptions and rationalizations. Journal of Travel Research, 52(2), 182—191.

Gamson, W. A., & Modigliani, A. (1989). Media discourse and public opinion of nuclear power: A constructionist approach. American Journal of Sociology, 95, 1 —37.

George, R. (2003). Tourist's perceptions of safety and security while visiting Cape Town. Tourism Management, 24(5), 575—585.

Gray, J., & Wilson, M. A. (2009). The relative risk perceptiion fo travel hazards. Environment and behaviour, 41(2), 185—204.

Griffith, D. A., & Albanese, P. A. (1996). An examination of Plog's psychographic travel model within a student population. Journal of Travel Research, 34, 47—51.

Gupta, D. (1990). The Economics of Political Instability: The Effect of Political Instability on Economic Growth. New York: Paeger.

Hall, C. M. (2002). Travel safety, terrorism and the media: The significance of the issue-attention cycle. Current Issues in Tourism, 5(5), 458—466.

Hall, C. M., & O'Sullivan, V. (1996). Tourism, political instability and violence. In A. Pizam, & Y. Mansfeld (Eds.), Tourism, crime and international security issues. New York: John Wiley.

Henderson, J. (2004). Tourism and crisis. London: Thompson Learning.

Heng, Y. (2006). War as risk management: Strategy and conflict in an age of globalised

risk. London: Routledge.

Hove, T., Paek, H.-J., Y., M., & Jwa, B. (2015). How newspapers represent environmental risk: The case of carcinogenic hazards in South Korea. Journal of Risk Research, ¡8(10), 1320—1336.

Hughes, E., Kitzinger, J., & Murdock, G. (2006). The media and risk. In P. Taylor-Gooby, & J. Zinn (Eds.), Risk in social science. Oxford: Oxford University Press.

Hunter-Jones, P., Jeffs, A., & Smith, D. (2007). Backpacking your way into crisis: An exploratory study into perceived risk and tourist behavior amongst young people. Journal of Travel & Tourism Marketing, 23, 237—248.

Iyengar, S. (1991). Is anyone responsible? How television frames political issues. Chicago, IL: University of Chicago Press.

Jackson, M. S., & Inbakaran, R. (2006). Development of a tourist personality inventory. In CAUTHE 2006 conference.

Jenkin, C. (2006). Risk perception and Terrorism: Applying the psychometric paradigm. Homeland security affairs, 2(2).

Jun, S. H., & Vogt, C. (2013). Travel information processing applying a dual-process model. Annals od Tourism Research, 40,191—212.

Kahneman, D., & Tversky, A. (1984). Choices, values, and frames. American Psychologist, 39, 341—350.

Kasperson, R. E. (2005). Social contours of risk, volume 1 : Publics, risk communication and the social amplification of risk. London,, GBR: Earthscan.

Kim, S. B., & Kim, D. Y. (2014). The effects of message framing and source credibility on green messages in hotels. Cornell Hospitality Quarterly, 55(1), 64—75.

Kitzinger, J. (2009). Risk, news coverage and public responses. Managing the social impacts of change from a risk perspective (Beijing).

Kozak, M., Crotts, J. C., & Law, R. (2007). The impact of the perception of risk on international travellers. International Journal of Tourism Research, 9(4), 233—242.

Kuto, B., & Groves, J. (2004). The effects of terrorism: Evaluating Kenya's tourism crisis. E-review of tourism Research, 2(4), 88—95.

Larsen, S., Brun, W., Torvald, O., & Selstad, L (2011). Effects of sudden and dramatic events on travel desire and risk judgments. Scandinavian Journal of Hospitality and Tourism, 11(3), 268—285.

Law, R. (2006). The perceived impact of risks on travel decisions. International Journal of Tourism Research, 8, 289—300.

Lecheler, S., & de Vreese, C. H. (2010). Framing Serbia: The effects of news framing on public support for EU enlargement. European Political Science Review, 2(1), 73—93.

Lecheler, S., de Vreese, C. H., & Slothuus, R. (2009). Issue importance as a moderator of framing effects. Communication Research, 36(3), 400—425.

Lee, S. K., Oh, C., & O'Leary, J. (2005). Estimating the impact of the september 11 terrorist attacks on the US air transport demand using Intervention analysis. Tourism Analysis, 9(4), 355—36 .

Lepp, A., & Gibson, H. (2003). Tourist roles, perceived risk and international tourism. Annals of Tourism Research, 30(3), 606—624.

Lepp, A., & Gibson, H. (2008). Sensation seeking and tourism: Tourist role, perception of risk and destination choice. Tourism Management, 29, 740—750.

Lepp, A., Gibson, H., & Lane, C. (2011). Image and perceived risk: A study of Uganda and its official tourism website. Tourism Management, 32, 675—684.

Litvin, S. W. (2006). Revisiting Plog's model of allocentricity and psychocentricity... One more time. Cornell Hotel and Restaurant Administration Quarterly, 47(3), 245—253.

Liu, B., & Pennington-Gray, L (2015). Bed bugs bite the hospitality industry? A framing analysis of bed bug news coverage. Tourism Management, 48, 33—42.

Lowenstein, G., Hsee, C., Weber, E., & Welch, N. (2001). Risk as feelings. Psychological bulletin, ¡27(2), 267—286.

L'Etang, J., Falkheimer, J., & Lugo, J. (2007). Public relations and tourism: Critical reflections and a research agenda. Public Relations Review, 33, 68—76.

Mansfeld, Y. (2006). The Role of Security Information in Tourism Crisis Management: The Missing Link. In Y. Masfeld, & A. Pizam (Eds.), Tourism, security and safety: From the theory to practice. Butterworth-Heinemann.

Mansfeld, Y., Jonas, A., & Cahaner, L (2016). Between tourists' faith and perceptions of travel Risk: An exploratory study of the Israeli Haredi community. Journal of travel research, 55(3), 395—413.

Maoz, I. (2012). Threats versus promises: How the framing of concession appeals in news coverage affects support for compromise. International Journal of Communication, 6, 2280—2300.

Marks, L. A., Kalaitzandonakes, N., Wilkins, L., & Zakharova, L. (2007). Mass media framing of biotechnology news. Public Understanding of Science, ¡6(2), 183—203.

Matthews, J. (2010). News sources and perceptual effects: An analysis of source attribution within news coverage of alleged terrorist plots. PhD, Bournemouth University.

Min, K. S., Martin, D., & Jung, J. M. (2013). Designing advertising campaigns for destinations with mixed images: Using visitor campaign goal messages to motivate visitors. Journal of Business Research, 66(6), 759—764.

Moutinho, J. (2000). Strategic management in tourism. Wallingford: CABI Publishing.

Nelson, T. E., Lecheler, S., Schuck, R. T., & De Vreese, C. H. (2012). Framing effects on values, emotions, and attitudes. Portland, Oregon: Western Political Science Association.

Nickerson, N., & Ellis, G. D. (1991). Traveler types and activation theory: A comparison of two models. Journal of Travel Research, 29, 26—31.

ONS. (2011). In J. Beaumont (Ed.), Social trends: Lifestyles and social participation.

Otieno, C., Renkl, A., Liebler, K., Deil, U., Ludemann, T., & Spada, H. (2014). Informing about climate change and invasive species: How the presentation of information affects perception of risk, emotions, and learning. Environmental Education Research, 20(5), 612—638.

Otieno, C., Spada, H., & Renkl, A. (2013). Effects of news frames on perceived risk, emotions, and learning. PLoS ONE, 8(11).

Pan, Z., & Kosicki, G. M. (1993). Framing Analysis: An approach to news discourse. Political Communication, ¡0, 55—75.

Pape, R. A. (2005). Dying to Win: The Strategic Logic of Suicide Terrorism. New York, NY: Random House.

Perse, E. (2001). Media effects and society. Mahwah, NJ: Lawrence Erlbaum Associates, Inc., Publishers.

Petts, J., Horlick-Jones, T., & Murdock, G. (2001). Social amplification of risk: The media and the public. Health and Safety Executive.

Pizam, A., Jeong, G., Reichel, A., Van Boemmel, H., Lusson, J., Steynberg, L., et al. (2004). The relationship between risk taking, sensation seeking and the tourist behavior of young adults: A cross cultural study. Journal of Travel Research, 42, 251—260.

Plog, S. C. (1974). Why destination areas rise and fall in popularity. Cornell Hotel and restaurant quarterly, ¡4(4), 55—58.

Plog, S. C. (1991). Leisure travel: Making it a growth market... again! New York: Wiley.

Plog, S. C. (2001). Why destination areas rise and fall in popularity. Cornell Hotel and Restaurant Administration Quarterly, 42(3), 13—24.

Plog, S. C. (2002). The power of psychographics and the concept of venturesome-ness. Journal of travel research, 40(3), 244—251.

Price, V., & Tewksbury, D. (1997). News values and public opinion: A theoretical account of media priming and framing. In G. A. Barett, & F. J. Boster (Eds.), Progress in communication sciences: Advances in persuasion. Greenwich, CT: Ablex.

Reisinger, Y., & Mavondo, F. (2005). Travel anxiety and intentions to travel Internationally: Implications of travel risk perception. Journal of travel research, 43, 212—225.

Renn, O. (2004). Perception of risks. The Geneva Papers on Risk and Insurance, 29(1), 102—114.

Renn, O. (2008). Risk Governance: Coping with uncretainty in a complex world (London).

Richter, K. R., & Waugh, W. L (1986). Terrorism and tourism as logical companions. Tourism Management, 7(4), 230—238.

Ritchie, B. W. (2004). Chaos, crises and disasters: A strategic approach to crisis management in the tourism industry. Tourism Management, 25, 669—683.

Ritchie, B. (2009). Crisis and disaster management for tourism. Bristol, UK: Channel View Publications.

Ritchie, B., Dorrell, H., Miller, D., & Miller, G. A. (2004). Crisis communication and recovery for the tourism industry. Journal of Travel & Tourism Marketing, ¡5(2—3), 199—216.

Rittichainuwat, B. N., & Chakraborty, G. (2009). Perceived travel risks regarding terrorism and disease: The case of Thailand. Tourism Management, 30,410—418.

Roehl, W. S., & Fesenmaier, D. R. (1992). Risk perception and pleasure travel: An exploratory analysis. Journal of Travel Research, 30(4).

Rubin, G. J., Brewin, C. R., Greenberg, N., Simpson, J., & Wessely, S. (2007). Enduring consequences of terrorism: 7-Month follow-up survey of reactions to the bombings in London on 7 july 2005. The British Journal of Psychiatry, ¡90, 350—356.

Saha, S., & Yap, G. (2013). The moderation effects of political instability and terrorism on tourism development: A cross-country panel analysis. Journal of Travel Research, 53(4), 509—52 .

Santana, G. (2001). Globalisation, safety and national security. In S. Wahab, & C. Cooper (Eds.), Tourism in the age of globalisation. London: Routtledge.

Scarbrough, G. I. (1998). An expert system for assessing vulnerabity to instability. In J. L. Davies, & T. R. Gurr (Eds.), Preventive measures: Building risk assessment and crisis early warning systems. Rowman & Littlefield.

Scheufele, D. A. (1999). Framing as a theory of media effects. Journal of Communication, 49(1), 103—122.

Scheufele, B. T. (2006). Frames, schemata, and news reporting. Communications, 3¡, 65—83.

Schmid, A. (2005). Terrorism as psychological warfare. Democracy and Security, ¡, 137—146.

Schmid, & Jongman. (1988). Political terrorism: A new guide to actors, authors, concepts, data bases, theories, and literature. Amsterdam: Trnasaction Books.

Schroeder, A., Pennington-Gray, L., Kaplanidou, K., & Zhan, F. (2013). Destination risk perceptions among U.S. residents for London as the host city of the 2012 Summer Olympic Games. Tourism Management, 38,107—119.

Schuck, R. T., & de Vreese, C. H. (2006). Between risk and opportunity : News framing and its effects on public support for EU enlargement. European Journal of Communication, 2¡ (5), 5—32.

Schütz, H., & Wiedemann, P. M. (2008). Framing effects on risk perception of nanotechnology. Public Understanding of Science, ¡7, 369—379.

Seddighi, H. R., Nuttall, M. W., & Theocharous, A. L. (2001). Does cultural background of tourists infuence the destination choice? an empirical study with special reference to political instability. Tourism Management, 22,181—191.

Seddighi, H. R., Theocharous, A. L., & Nuttall, M. W. (2002). Political instability and Tourism: An empirical study with special reference to the microstate of Cyprus. International Journal of Hospitality & Tourism Administration, 3(1).

Sharifpour, M., Walters, G., & Ritchie, B. W. (2013). The mediating role of sensation seeking on the relationship between risk perceptions and travel behavior. Tourism Analysis, ¡8(5), 543—55 .

Sharifpour, M., Walters, G., Ritchie, B. W., & Winter, C. (2014). Investigating the role of prior knowledge in tourist decision making: A structural equation model of risk perceptions and information search. Journal of Travel Research, 53(3),

307-322.

Siermann, C. L. J. (1998). Politics, institutions and the economic performance of nations. Northampton: Edward Elgar Publishing, Inc.

Silver, R. C., Holman, E. A., Mcintosh, D. N., Poulin, M., & Gil-Rivas, V. (2002). Nationwide longitudinal study of psychological responses to September 11. JAMA, 288,1235—1244.

Sjoberg, L. (2007). Emotions and risk perception. Risk management, 9, 222—237.

Slovic, P. (1987). Perception of risk. Science, 236, 280-285.

Slovic, P., Fischhoff, B., & Lichtenstein, S. (1982). Why study risk perception? Risk Analysis, 2, 83-94.

Slovic, P., Fischhoff, B., & Lichtenstein, S. (1986). The psychometric study of risk perception. in V. Covello, J. Menkos, & J. Mumpower (Eds.), Risk evaluation and management. New York: Pienum Publishing Corporation.

Smith, S. L. J. (1990). A test of Plog's allocentric/psychocentric model: Evidence from seven nations. Journal of Travel Research, 28, 40—43.

Sniderman, P. M., & Theriault, S. M. (2004). The structure of political argument and the logic of issues framing. in W. E. Saris, & P. M. Sniderman (Eds.), Studies in public Opinion: Attitudes, non-attitudes, measurement error, and change. Princeton: Princeton Univrsity Press.

Sonmez, S. (1998). Tourism, terrorism and political instability. Annals od Tourism Research, 25(2), 416—456.

Sonmez, S. F., & Graefe, A. R. (1998a). infuence of terrorism risk on foreign tourism decisions. Annals of Tourism Research, 25(1), 112—144.

Sparks, B., & Browning, V. (2011). The impact of online reviews on hotel booking intentions and perception of trust. Tourism Management, 32(6), 1310—1323.

Spencer, J. W., & Triche, E. (1994). Media constructions of risk and Safety: Differential framings of hazard events. Sociological Inquiry, 64(2), 199—213.

START. (2012). National consortium for the study of terrorism and responses to terrorism. Global terrorism database.

Sunstein, C. R. (2003). Terrorism and probability neglect. The Journal of Risk and Uncertainty, 26,121—136.

Tankard, J. W. (2001). The empirical approach to the study of media framing. in S. D. Reese, & O. H. Gandy (Eds.), Framing public Life: Perspectives on media and our understanding of the social world. Mahwah, NJ: Lawrence Erlbaum.

Tarlow, P. (2014). Tourism Security: Strategies for effectively managing travel risk and safety. Oxford: Butterworth-Heinemann.

Tcheocharous, A. (2010). A contextual typology for the study of the relationship between political instability and tourism. International Journal of Tourism Policy, 3(4), 354—363.

Thapa, B. (2003). Tourism in Nepal: Shangri-La's troubled times. in C. M. Hall, D. J. Timothy, & D. T. Duval (Eds.), Safety and security in Tourism: Relationships, management, and marketing. Routledge.

Tourism Victoria. (2010). Crisis essentials. Crisis management for tourism businesses. Online.

Valencia, J., & Crouch, J. (2008). Travel Behavior in Troubled Times: The Role of Consumer Self-Confidence. Journal of Travel and Tourism Marketing, 25(1), 25—42.

Van Gorp, B. (2007). The constructionist approach to Framing: Bringing culture back

in. Journal of Communication, 57(1), 60—78.

de Vreese, C. H. (2005). News Framing: Theory and typology. Information Design Journal+Document Design, 13(1), 51—62.

Wang, J., & Ritchie, B. W. (2010). A theoretical model for strategic crisis planning: Factors influencing crisis planning in the hotel industry. International Journal of Tourism Policy, 3(4), 297—317.

Weaver, D. B. (2012). Psychographic insights from a South Carolina protected area. Tourism Management, 33, 371—379.

Williams, A. M., & Balóz, V. (2013). Tourism, risk tolerance and competences: Travel organization and tourism hazards. Tourism Management, 35, 209—221.

Wimmer, R. D., & Dominick, J. R. (2011). Mass media Research: An introduction. Cengage Learning.

Woods, J. (2007). What we talk about when we talk about terrorism: Elite press coverage of terrorism risk from 1997 to 2005. The Harvard International Journal of Press/Politics, 12(3).

Woods, J. (2011). Framing Terror: An experimental framing effects study of the perceived threat of terrorism. Critical Studies on Terrorism, 4(2), 199—217.

Grzegorz Kapuscinski is a lecturer of Marketing at Bournemouth University. He holds a Ph.D in Travel Behaviour and Media Studies (Bournemouth University, 2014). His research interests include consumer behaviour; media effects; place branding; crisis communications; risk perception and travel behaviour.

Barry Richards is a Professor of Political Psychology at Bournemouth University, before which he was Professor and Head of the Department of Human Relations at the University of East London. He is the author of Emotional Governance: Politics, Media and Terror (Palgrave, 2007), and articles and book chapters on psychological aspects of the threat of violent extremism. He is a founding co-editor of the Sage journal Media, War and Conflict. He has written widely on popular culture and politics; his major interests are in the emotional dynamics of conflict and extremism, and in psychosocial dimensions of cultural change. His latest book, What Is Holding Us Together? is to be published by Karnac Books in 2015.