Effect of Risk Message Framing on Tourists’ Travel Intention: Roles of Resilience and Impulsivity

Abstract

The effect of risk message framing on travel intention requires more empirical investigations in long-term high-risk situations like the current COVID-19 pandemic. Based on frame theory, this study employed an experimental design to examine how two contrasting approaches of COVID-19 risk message framing (amplifying vs. attenuating) affected post-pandemic travel intention via the mediation of perceived safety and travel fear, and how resilience and impulsivity as tourist traits moderate these relationships. Survey results based on 481 responses revealed that: (1) risk messages significantly predicted tourists’ perceived safety, travel fear, and travel intention; (2) tourists’ perceived safety and travel fear mediated the effects of risk messages on travel intention; (3) while resilience moderated the effects of message framing on perceived safety and travel intention, impulsivity only moderated the effect of message framing on travel fear. The study provides a theoretical basis and practical implications for destination risk communications.

Keywords

risk message perceived safety travel fear impulsivity frame theory

Introduction

Safety is a basic tourist need, and a destination’s prosperity is highly reliant on its ability to offer tourists safe and pleasant visit experiences (Chauhan 2007; Xie, Zhang, and Morrison 2021). Destinations’ effective communication of risk messages helps tourists form risk, safety, and image perceptions and evaluations appropriately and in a realistic way (Sano and Sano 2019; Xie et al. 2021), which fundamentally affect tourists’ travel decisions, experiences, and satisfaction. Especially in a crisis situation, risk messages related to crisis events tend to be important information affecting tourist perceptions of destination risk or safety (Oliveira and Huertas 2019; Sano and Sano 2019); these messages may even determine whether crisis events will lead to public opinion crises and enduring market crises (Luo and Zhai 2017; Xie et al. 2022). The COVID-19 pandemic is considered to be an unprecedented event in world history, creating a long-term high-risk situation and having greatly threatened the development of global tourism markets (Fong, Law, and Ye 2020). It is stridently important for all tourist destinations to seek strategies to sustain the tourism industry in a relatively high-risk situation caused by COVID-19 pandemic. Agenda-setting and media communication with risk messages seem to be a critical task for destinations in restoring safety images and confidence of travel markets.

Risk communication during public health crises is intended to improve the effectiveness of risk message dissemination and educate the public in adopting actions to cope with health issues (Freimuth, Linnan, and Potter 2000). Destination risk communication refers to the agenda-setting and media communication of information and messages about the nature, magnitude, significance or control of a risk, so as to ameliorate tourists’ threat perceptions, inform tourists about risks, and promote the safety of destination environment (Covello 1992; Xie et al. 2021). Generally speaking, tourists are highly sensitive to risk messages, and risk messages increase tourists’ perceived risk and reduce their perceived safety and travel intentions (Kapuściński and Richards 2016; Sano and Sano 2019; Xie et al. 2021). Meanwhile, tourists’ responses to risk messages and destination risk communication have been confirmed, especially in the health crisis context; furthermore, the moderation roles of travel experience, empathy, and perceived waiting time (the time that people predict that the crisis may last) in the effects of risk message communication have also been examined (Liu-Lastres, Schroeder, and Pennington-Gray 2019; Xie et al. 2021). In such a context, risk communication emerges as an important topic in tourism research, and destination risk message communication is a basic factor determining tourists’ perceptions, emotions, and behaviors. However, little research has investigated the mediation and moderation mechanisms in the relationship between risk message and travel intention from a risk communication perspective.

Three key gaps exist within this particular research area. First, the literature of travel decision-making in crisis situations shows inconsistent and inconclusive findings. Researchers generally believe that tourists avoid unsafe destinations, with crisis events reducing individuals’ confidence and willingness to travel (Liu-Lastres, Schroeder, and Pennington-Gray 2019; Pizam and Smith 2000; Sano and Sano 2019). However, a destination’s inherent risk may serve a drawcard to attract those tourists seeking adventure and a sense of excitement (George 2010; Page, Bentley, and Walker 2005; Wang et al. 2019). In effect, “terrorism and security” issues constitute only one dimension of tourists’ perceived risk in crisis situations (Fuchs et al. 2013), and some travelers perceive destinations to be safer after terrorist attacks (Wolff and Larsen 2014), and even adopt means of rationalization to justify their risky and seemingly irrational travel decisions (Fuchs et al. 2013). In addition, some tourists may visit post-disaster destinations as a form of “dark tourism” (Biran et al. 2014). Thus, travel intentions in crisis situations appear to be complex and for the time being require more empirical investigations from different theoretical perspectives.

Second, the underlying mediation mechanisms in the relationship between risk message and travel intention lack empirical investigation. Although the outcomes of destination risk messages are deemed negative (Kapuściński and Richards 2016; Sano and Sano 2019; Xie et al. 2021), little research has examined the mediation of individuals’ internal state in the relationship between risk message and travel intention. During a major crisis, destination risk messages trigger tourists’ cognitive evaluation and emotional responses in relation to safety and risk, which in turn may determine their travel decisions. Perceived safety and travel fear represent cognitive states and emotional experience commonly possessed by tourists in high-risk or threat situations (George 2010; Xie, Zhang, and Morrison 2021; Zhang et al. 2022; Zheng, Luo, and Ritchie 2021), which may mediate the effect of risk message on travel intention. Nevertheless, these possible mediation roles have not be empirically confirmed in a major crisis context.

Third, the moderation roles of pertinent tourist personal traits in crisis situations, such as resilience and impulsivity, in the effect of risk message on travel intention, remain to be unknown. Tourists’ interpretation and response to risk messages in high-risk situations are likely to be affected by their personal traits and characteristics (Kapuściński and Richards 2016; Liu-Lastres, Schroeder, and Pennington-Gray 2019; Wang et al. 2019; Xie et al. 2021). Resilience represents the capability and adaptability in effectively responding to crises and threat situations, reflecting the psychological mechanism for individual recovery and growth under high-risk situations (Dyer and McGuinness 1996; Gillespie, Chaboyer, and Wallis 2007; Wagnild and Young 1993). Accordingly, resilience is a positive personal trait that helps tourists cope with the negative impact of risk messages in a crisis situation (Zheng, Luo, and Ritchie 2021). In tourism field, impulsivity can be understood as spontaneous, unreflective, and hedonically-driven behavioral reaction and psychological tendencies of tourists in pleasant consumption situations (Karl, Chien, and Ong 2021; Meng et al. 2019). Since the concept of impulsivity is closely related to factors such as risk-taking, sensation seeking, adventuresomeness, and novelty seeking (Eysenck et al. 1985; Whiteside and Lynam 2001; Zuckerman et al. 1993), impulsive tourists may take risky decisions and impulse buying in crisis situations in order to seek adventurous, memorable, and novel experiences (Li 2007; Li et al. 2021; Meng et al. 2019). Thus, impulsivity is an important personal trait that may trigger tourists’ seemingly irrational cognitive evaluations and risk-seeking decisions in a crisis situation. Despite the relevance of both personal traits in tourists’ decision making under crisis situations, neither trait has been explicitly examined in tourists’ interpretation of risk messages and subsequent travel decisions.

To address the above-mentioned gaps, considering the long-term crisis situation of the COVID-19 pandemic, this study investigated the frame effect of pandemic risk messages on post-pandemic travel intention. It attempted to make three contributions. First, based on frame theory, the impact of risk message frame on perceived safety, travel fear, and travel intention was examined, providing an empirical case to clarify tourists’ travel decision-making in crisis situations. Second, based on the Stimuli-Organism-Response (SOR) framework, this study investigated the mediation effects of perceived safety and travel fear between risk messages and travel intention, which revealed the cognitive and emotional mediation mechanism between risk messages as stimuli and travel intention as a behavioral response. Third, based on the person-situation interaction perspective, this study introduced resilience and impulsivity as pertinent personal traits and confirmed the moderation effects of these personal traits on risk messages’ influences on tourists’ attitudes and behaviors during a long-term high-risk situation. In summary, this study revealed the cognitive and emotional mediation effects of risk message frames on travel intention as well as the moderation roles of resilience and impulsivity as personal traits, thus providing a theoretical basis and strategic guidance for destination risk communications during a major public health crisis such as the COVID-19 pandemic.

Literature Review and Hypotheses Development

Theoretical Foundation

Defining frames as selective constructions and presentations based on situation and reality, frame theory provides a valuable approach for understanding individuals’ interpretations and constructions of meaning relating to relevant events (Gamson and Modigliani 1989). Goffman (1974) first proposed the concept of frame, arguing that individuals adopt “schemata of interpretation” (p. 21) to organize new information and construct meanings. Frames enable individuals to “locate, perceive, identify, and label” the information around them. Following Goffman (1974), Gitlin (1980) treated frames as “principles of selection, emphasis and presentation composed of few tacit theories about what exists, what happens, and what matters” (p. 6). For Entman (1993), framing is a process to “select some aspects of a perceived reality and make them more salient in the communicating context, in such a way as to promote a particular problem definition, causal interpretation, moral evaluation and/or treatment recommendation for the item described” (p. 52). This definition indicates that frames exist in communication sources, texts, audiences, and social cultural contexts. Thus, frames can be understood as the selective presentation of events by communication sources as well as the subjective interpretation of audiences; the adoption of different frames produce different frame effects.

Framing effect as a concept is proposed initially by Kahneman and Tversky (1984), referring to the influence of differences in presentation of issues on individuals’ decision-making. Framing effect aims to reveal how the communicators can influence individuals’ attitudes and behaviors by presenting potentially relevant factors to draw individuals’ attention to specific aspects and making them more salient in issue statements (Sniderman and Theriault 2004). Based on message content, framing effects can be classified into two forms: equivalency and emphasis framing (Druckman 2001). The equivalency framing effect investigates how the use of different but logically equivalent words or sentences to highlight the positive or negative aspects of an issue causes different responses from people (e.g., 95% employment vs. 5% unemployment) (Kahneman and Tversky 1984). This type of framing research focuses on the wording effect on individuals’ decision-making. The emphasis framing effect is concerned with the influence of presenting different aspects of a complex issue without assuming that the contents are factually equivalent (e.g., emphasis on the advantages of technology over its disadvantages) (Druckman 2001; Tankard 2001). As the emphasis framing effect is more applicable to real news coverage and to complex communication situations, it has been suggested as the dominant framing approach in framing research in news coverage and risk communication fields (Kapuściński and Richards 2016; Nelson et al. 2012; Sniderman and Theriault 2004). In the field of tourism research, frame theory has been widely applied to investigate hotel booking intentions (Sparks and Browning 2011), destination image formation (Zhang et al. 2018), customers’ environmentally friendly activities (Kim and Kim 2014), and public holiday system and its reform (Wu et al. 2012). Since framing can help destinations emphasize certain aspects and agenda-set crisis events and related messages in a beneficial way, thereby reducing the potential negative impacts (Coombs 2007), the frame theory has received considerable attention in tourism crisis and risk communication research. For example, based on frame theory, Kapuściński and Richards (2016) proposed that the frames and portrayals of crises are crucial to tourists’ perceived risk associated with destinations. Xie et al. (2021) investigated the frame effect of risk message on tourists’ basic and destination travel intentions. Accordingly, this research adopted the emphasis framing approach in the COVID-19 pandemic context to examine the frame effect of pandemic risk messages on tourists’ post-pandemic travel intention.

Effects of Risk Message Framing on Perceived Safety, Travel Fear and Travel Intention

Risk refers to a possibility of danger, harm or loss, and a chance of hazard (Reisinger and Mavondo 2005). Risk message as a general term refers to an expressible set for various message and information elements related to risk (Xie et al. 2021). Individuals’ subjective assessment and comprehensive judgment of risk messages form their perceived risk (Rimal 2003). Based on individuals’ perceptions of the risk elements in the messages, risk messages can be classified into high-risk and low risk perception messages (Liu-Lastres, Schroeder, and Pennington-Gray 2019; Sano and Sano 2019). Accordingly, destinations can improve tourists’ perceived safety and travel intention by intentionally framing and constructing high-risk or low-risk messages. Risk message framing aims to control how communicators influence individuals’ attitudes and behaviors in risk message communication by selectively presenting risk elements and drawing individuals’ attention to specific risk aspects (Sniderman and Theriault 2004; Xie et al. 2021). Currently, destination risk communication studies have focused on exploring the effect of risk message based on risk amplifying and risk attenuating frames. For example, Kapuściński and Richards (2016) proposed that destinations’ portrayals of terrorism and political instability incidents could take two frames: risk amplifying and risk attenuating; they investigated the news media frame effects on tourists’ perceived safety based on these two frames. Xie et al. (2021) confirmed that both basic travel intentions and destination travel intentions in risk attenuating frame were greater than those in risk amplifying frame in the context of COVID-19. Following these studies, the current study aims to investigate the impacts of risk message on tourists’ perceived safety, travel fear, and travel intention based on risk-amplifying and risk-attenuating frames.

Perceived safety is a prime factor determining tourists’ travel intention (Sano and Sano 2019; Wang and Lopez 2020). Especially in high-risk or crisis situations, tourists tend to search and seek for more information in the pre-visit stage to reduce uncertainty and ensure safety (Aliperti and Cruz 2019; Law, Buhalis, and Cobanoglu 2014). Crisis events and risk messages from destinations fundamentally affect tourists’ perceived safety and may trigger travel fear, thereby reducing their travel intention. Specifically, Liu-Lastres, Schroeder, and Pennington-Gray (2019) confirmed that epidemic risk messages negatively affected tourists’ perceived safety and travel intention. According to Fennell (2017), travel fear refers to the negative emotional responses of tourists to crises, uncertainties, or other calamities; Zheng, Luo, and Ritchie (2021) founded that destination risk messages and threat perceptions triggered tourists’ pandemic travel fear, and then affected their travel decisions (e.g., travel avoidance, cautious travel) during COVID-19. Moreover, tourists’ cognitive assessment and behavioral decisions in risk situations involve two contrasting modes of risk information processing: heuristic and systematic (Aliperti and Cruz 2019; Ryu and Kim 2015; Zhang et al. 2022). Specifically, in heuristic processing, individuals tend to rely on external cues and accessible information when deciding to accept messaging, whist in systematic processing, individuals make efforts to comprehend and judge a message’s content and assess its validity in relation to its conclusion (Chaiken 1980; Ryu and Kim 2015; Zhang et al. 2022). Based on the dual-process model (i.e., heuristic and systematic), McCabe, Li, and Chen (2016) proposed a new general model to explain different types of choice strategies and the constructive nature of preference; this model emphasized the critical roles of individual and contextual factors in tourists’ behavioral decision-making. The above mentioned studies provide a theoretical basis for investigating tourists’ travel decisions in risk situations.

Frame theory posits that differences in presentation of issues may trigger individuals’ differentiated decision-making responses (Kahneman and Tversky 1984). Accordingly, tourists’ perceived safety, travel fear, and travel intention responses to different risk frames (i.e., risk amplifying and risk attenuating frames) may be different. Previous research confirmed that tourists’ perceived safety and travel intention in risk attenuating frames were greater than that in risk amplifying frames, and tourists’ perceived risk in risk amplifying frames was greater than that in risk attenuating frames (Kapuściński and Richards 2016; Xie et al. 2021). In addition, because of the social amplification effect of risk and risk frames (Renn et al. 1992), tourists’ cognition judgment and emotional response to macro tourism environments tend to be negative in crisis situations (Fennell 2017; Xie et al. 2021). Therefore, tourists’ perceived risk and travel fear are higher while perceived safety and travel intention are lower in risk amplifying frame compared with those in risk attenuating frame. Based on the above discussions, we propose the following hypotheses:

H1a: Tourists’ travel intention in responding to risk messages in risk attenuating frame is higher than that in risk amplifying frame; specifically, tourists exposed to low-risk COVID-19 messages tend to have higher travel intention than those exposed to high-risk COVID-19 messages.

H1b: Tourists’ perceived safety in responding to risk messages in risk attenuating frame is higher than that in risk amplifying frame; specifically, tourists exposed to low-risk COVID-19 messages tend to have higher perceived safety than those exposed to high-risk COVID-19 messages.

H1c: Tourists’ travel fear in responding to risk messages in risk attenuating frame is lower than that in risk amplifying frame; specifically, tourists exposed to low-risk COVID-19 messages tend to have lower travel fear than those exposed to high-risk COVID-19 messages.

Mediation of Perceived Safety and Travel Fear Between Risk Message Framing and Travel Intention

In a crisis situation, the purpose of destination risk and crisis communication is to diminish the effect of public perception of crisis through message agenda-setting and dissemination, as well as to alleviate threat and risk perception triggered by the crisis, thereby restoring tourists’ safety-related confidence and travel intention (Reynolds and Seeger 2005; Sano and Sano 2019; Xie et al. 2021). According to the Stimuli-Organism-Response framework (SOR), when receiving external stimuli, individuals’ internal states are activated, followed by behavioral responses of approaching or avoiding (Mehrabian and Russell 1974). In other words, individuals’ internal states such as cognition and affect, mediate the impacts of external stimuli on behavioral responses. Currently, SOR framework has presented good applicability and predictive power in many fields, and has been widely used in tourism field to reveal the mediation process of the internal state in which external stimuli trigger individual behavioral response. Specifically, SOR framework has been applied to study the impact of environmental stimuli on tourists’ mobile social tourism shopping (Hew et al. 2018), the impact of tourscapes and sensation seeking on tourists’ liminal experience (Zhang and Xu 2019), and that of destination social responsibility on tourists’ environmentally responsible behavior (Su and Swanson 2017). Moreover, SOR framework can well reveal the internal mediation mechanisms of destination risk stimuli (e.g., disease deterrence, COVID-19 knowledge, perceived threat, and perceived information overload) on tourists’ behavioral decisions (e.g., conspicuous cooking, information avoidance intention, and preventive behaviors) in crisis situations (Goyal and Verma 2021; Song, Yao, and Wen 2021). Accordingly, this study examined the internal mediation mechanism of risk message frame affecting tourists’ travel intention based on SOR framework.

In this study, when tourists receive destination risk communication messages, their cognitive evaluation and emotional response related to safety and risk will be triggered, which in turn influence their decisions on whether to travel. Importantly, tourists’ perceived safety is based on a comprehensive judgment and cognitive evaluation of the destination’s safety and risk messages (Xie, Zhang, and Morrison 2021). It is the primary effects formed after exposing to safety and risk messages (Sano and Sano 2019; Wang and Lopez 2020). On the other hand, travel fear is derived from tourists’ emotional responses to adverse situations such as crisis, uncertainty, trauma, calamity, manifested as a strong, depressive, and aversive emotional state when tourists attempt to escape adversity but are powerless (Fennell 2017; Sylvers, Lilienfeld, and LaPrairie 2011). Following Epstein (1972) and Barlow (2002), travel fear can be understood as an aversive emotional state during which tourists are motivated to escape a specific and imminent threat; travel fear is a defensive response to external threats and risk elements (Cisler, Olatunji, and Lohr 2009). Thus, based on the SOR framework, perceived safety and travel fear are respectively the cognitive and emotional mediating variables through which destination risk messages exert their effects on tourists’ travel intention. In the literature, Liu-Lastres, Schroeder, and Pennington-Gray (2019) confirmed that perceived safety mediated the impact of risk messages on travel intention, and Wang et al. (2019) investigated and confirmed the mediation role played by worry between perceived risk and self-protective behavior. Based on the above discussions, we propose the following hypotheses:

H2a: Perceived safety mediates the impact of risk message framing on tourists’ travel intention; specifically, COVID-19 messages trigger tourists’ perceived safety, which then affects their travel intention.

H2b: Travel fear mediates the impact of risk message framing on tourists’ travel intention; specifically, COVID-19 messages trigger tourists’ travel fear, which then affects their travel intention.

Moderation Roles of Resilience and Impulsivity

With the advances and development of positive psychology, resilience has received considerable attention in many fields, and thus has become a prominent psychological resource as well as a valuable personal character strength. It is widely believed that resilience denotes to the capability of adapting and responding to crises and negative events such as stress, trauma, frustration, stress, and challenge (Dai, Zhuang, and Huan 2019; Dyer and McGuinness 1996). Resilience is a positive and ideal personal trait that helps an individual cope with multiple adversities (Wagnild and Young 1993). Resilient individuals can effectively cope with difficulties, protect themselves against the impact of adversity, trauma, crisis and challenge, and thus achieve good adjustment and self-development from negative events (Kuntz, Malinen, and Näswall 2017; Nguyen et al. 2016; Zhu et al. 2019). Currently in the field of tourism, resilience has been applied to investigate how tourists respond to adversities, crises, and disasters in traveling (Prayag 2018). Resilience was also found to be able to help tourists resist and alleviate the psychological fear and threat perception triggered by crisis events, maintain strong adaptability and recovery in crisis situations, and then affect tourists’ travel decisions (Zheng, Luo, and Ritchie 2021). Nonetheless, empirical investigations of the impact of resilience on tourists’ decision-making in crisis situations are still limited.

Tourists’ travel intention in crisis situations depend on their processing and evaluation of risk messages, as well as their personal traits and psychological characteristics. From the person-situation interaction perspective, individual behavior is the result of the interaction between personal and situational factors (Bandura 1977; Palmer et al. 2017). This perspective has been adopted in tourism field in studying how the interaction between personal factors and situational factors eventually leads to individual behavior. For example, Dai, Zhuang, and Huan (2019) confirmed that turnover intention and job engagement (behavior) of travel agency employees are a result of the interaction between abusive supervision (situational factor) and resilience (personal factor). And Wang and Lopez (2020) examined the moderation effects of risk propensity and self-efficacy (personal factor) on the relationship between safety message (situational factor) and travel intention (behavior). A personal trait is a stable psychological characteristic of an individual, and its role of moderating the responses of tourists to risk and crisis situations has been largely confirmed in the literature (Kapuściński and Richards 2016; Wang and Lopez 2020; Wang et al. 2019). Specifically, the traits of resilience and impulsivity may potentially moderate the effects of risk message on perceived safety, travel fear, and travel intention during a major crisis like COVID-19. Resilient tourists can maintain high adaptability in long-term high-risk situations caused by COVID-19, have strong resistance and immunity to the risk messages and threat information associated with the destination, and can overcome the negative impacts of adverse conditions on their attitudes and behaviors through psychological adjustment (Zheng, Luo, and Ritchie 2021). Compared with tourists with low resilience, high-resilience tourists tend to have more positive cognitive evaluation, emotional reaction, and behavioral response to destination risk communication. On this basis, we propose:

H3a: Resilience moderates the relationship between risk message framing and travel intention; specifically, high resilience tourists’ travel intention in response to COVID-19 messages is higher than that of low resilience tourists.

H3b: Resilience moderates the relationship between risk message framing and perceived safety; specifically, high resilience tourists’ perceived safety in response to COVID-19 messages is higher than that of low resilience tourists.

H3c: Resilience moderates the relationship between risk message framing and travel fear; specifically, high resilience tourists’ travel fear in response to COVID-19 messages is lower than that of low resilience tourists.

Impulsivity is a stable personal trait which can be defined as “swift action without forethought or conscious judgment” (Moeller et al. 2001, 1783). Individuals with this trait are prone to make rapid and unplanned reactions to internal or external stimuli without due considerations of the possible negative consequences of these reactions to themselves or others. Patton, Stanford, and Barratt (1995) categorized impulsivity into three dimensions: (1) acting on the spur of the moment (motor impulsivity), (2) not focusing on the task at hand (attentional impulsivity), and (3) not planning and thinking carefully (non-planning impulsivity). In addition, Whiteside and Lynam (2001) confirmed that impulsivity was composed of four factors: sensation seeking (the tendency to seek novelty, excitement and adventure), urgency (the tendency to cope with immediate threats), perseverance (the tendency to resist interference and complete difficult tasks), and premeditation (the tendency to plan and consider before action). Highly impulsive individuals have a strong tendency for sensation seeking and are willing to take risks for exciting, adventure, and novel experiences (Eysenck et al. 1985; Zuckerman et al. 1993). Tourism is an experiential consumption activity of people temporarily traveling to places outside their usual environments (Govers, Van Hecke, and Cabus 2008), and some tourists, driven by hedonism and sensation seeking, tend to visit risky destinations to engage in adventure activities and seek thrilling experiences (Fuchs et al. 2013; Page, Bentley, and Walker 2005; Wang et al. 2019). When a tourist has a high tendency of seeking novelty, adventure, and excitement experience, his or her behavioral decisions and cognitive evaluations are often dominated by the impulsive system, leading to intuition-based heuristic processing to risk messages rather than rational-based systematic processing (Karl, Chien, and Ong 2021; McCabe, Li, and Chen 2016). Moreover, tourists’ impulse buying of local products may be easily aroused by factors such as time scarcity, repurchase costs, cultural differences, and purchase pressures (Li 2007; Li et al. 2021). However, there is a void of research on the impacts of impulsivity as a personal trait on tourists’ decision-making in major crisis situations.

In this study, the motor dimension of impulsivity, defined as the predisposition toward rapid and reckless reactions to internal or external stimuli and of being less sensitive to the negative consequences of these reactions (Patton, Stanford, and Barratt 1995), was adopted. Impulsive tourists’ decisions are often toward hedonism, sensation seeking, and novelty pursuing, and they tend to lack planning and sufficient consideration before making decisions (Karl, Chien, and Ong 2021; Whiteside and Lynam 2001; Zuckerman et al. 1993). Thus, to seek excitement, memorable, and novel experience, impulsive tourists tend to make rapid, unplanned, and seemingly irrational cognitive evaluations and impulsive decisions toward external risk messages and threat elements. In addition, facing long-term travel restrictions and being in the usual mundane environments would form individuals’ strong motivations and intentions of traveling to other places (Xie et al. 2021). As such, impulsivity may moderate the impacts of risk message on tourists’ travel intention. Following the cognitive dissonance viewpoint (Festinger 1957), a travel decision in a crisis situation may be well in conflict with the evaluation that “it is no safe to go outside.” Since it is relatively easier to change the belief and attitude toward the crisis situation than to resolve the crisis, individuals with high sensation and novelty seeking tendency tend to improve perceived safety and reduce travel fear to reduce cognitive dissonance and achieve consistency in cognition and behavior. Thus, highly impulsive tourists may use rationalized means (e.g., blaming the media for overexposure of the crisis) to reduce their perceived risk and travel fear, thereby justifying their travel decisions in crisis situations (Fuchs et al. 2013). Compared with tourists with low impulsivity, high-impulsivity tourists tend to have higher perceived safety, travel intention and lower travel fear in response to destination risk communication in long-term travel restriction situations. On such a basis, we propose:

H4a: Impulsivity moderates the relationship between risk message framing and travel intention; specifically, high impulsivity tourists’ travel intention in response to COVID-19 messages is higher than that of low impulsivity tourists.

H4b: Impulsivity moderates the relationship between risk message framing and perceived safety; specifically, high impulsivity tourists’ perceived safety in response to COVID-19 messages is higher than that of low impulsivity tourists.

H4c: Impulsivity moderates the relationship between risk message framing and travel fear; specifically, high impulsivity tourists’ travel fear in response to COVID-19 messages is lower than that of low impulsivity tourists.

The proposed conceptual model is presented in Figure 1.

Figure 1.

Conceptual model.

Methodology

Research Design

we employed a between-subject factorial experiment design to examine how tourists responded to risk messages related to the COVID-19 pandemic using two risk frames (risk amplifying vs. risk attenuating). To ensure that the experiment was realistic, controllable and convenient to manipulate, we followed the recommendations of Kapuściński and Richards (2016) and Xie et al. (2021) in designing the stimuli materials, based on latest news of COVID-19 from domestic and international media. This study used China as the destination context to investigate the response of tourists to destination risk communication. Thus, high-risk messages such as confirmed COVID-19 cases, medium- and high-risk infection areas, virus mutations, and its new propagation characteristics in China were selected for the risk amplifying frame, whilst low-risk messages such as the effectiveness of pandemic prevention, COVID-19 cure rate, vaccine development, and vaccination in China were selected for the risk attenuating frame. Stimuli materials in different message frames contained similar message elements and adopted same presentation format (e.g., font, size, color and linewidth) to avoid any influence or bias from the message content and language. The stimuli materials were designed in Chinese, and they are basically the same in Chinese style and Chinese sentence structure. Moreover, the risk messages were optimized with the help of two tourism professors and six PhD students to ensure its content validity following in-depth discussion. English translations of the two frames are provided in the Supplemental Appendix.

The measurement scales were all adopted from previously published research, and some items were slightly revised based on the research context. The English scales were translated into Chinese and then back-translated with the help of two tourism professors and six PhD students. Four items from Rimal (2003) were adapted to measure perceived risk, literally assessing the severity of the possible consequences of infection with COVID-19 virus. Four items based on Zheng, Luo, and Ritchie (2021) were used to measure travel fear, asking the participants to judge their fears of traveling during the COVID-19. The three-item scale for perceived safety was adapted from Liu-Lastres, Schroeder, and Pennington-Gray (2019), assessing the safety level of current tourism environment or engaging in travel activities. Three items for travel intention were adopted from Xie et al. (2021), recording participants’ intention and wish to travel after the pandemic. Six items based on Smith et al. (2008) and Zheng, Luo, and Ritchie (2021) were chosen to measure resilience by identifying participants’ resistance, adaptation, recovery, and growth to threat situations caused by the pandemic. Four items based on Patton, Stanford, and Barratt (1995) and Li et al. (2011) (Chinese version) were selected to measure impulsivity, reflecting participants’ tendency to make hurried, simple, sensation-seeking, and impulsive decisions to external stimuli. Each item were measured with a seven-point Likert scale ranging from 1 (“strongly disagree ”) to 7 (“strongly agree”). Demographic information such as gender, marital status, education, occupation, monthly income, and travel frequency was also collected.

Research Procedure

A pilot experiment was conducted to assess the validity of the stimuli materials and experimental design, as well as the reliability of the measurement scales. A total of 100 university students who traveled during the COVID-19 period were invited to participate in the pilot experiment; 50 of them received the risk amplifying frame and 50 received the risk attenuating frame. They were first asked to read their assigned risk message frames and then answer a series of questions related to perceived risk, perceived safety (after), travel fear (after), resilience, impulsivity, and travel intention (after). Unlike that in the main experiment, they were not asked to answer any questions for the interested variables before reading the stimulus material. The results showed that the Cronbach’s alpha values for perceived risk (0.930), perceived safety (0.963), travel fear (0.793), resilience (0.949), impulsivity (0.888), and travel intention (0.866) were over 0.8, indicating good internal consistency of the scales. Moreover, participants who were exposed to messages in the risk amplifying frame rated perceived risk (M _Amplifying = 6.30, M _Attenuating = 4.25, t = 12.117, p < .001) and travel fear (M _Amplifying = 5.49, M _Attenuating = 3.39, t = 9.637, p < .001) significantly higher than those exposed to messages in the risk attenuating frame. In contrast, participants who were exposed to the message in the risk amplifying frame rated perceived safety (M _Amplifying = 2.11, M _Attenuating = 4.47, t = −11.867, p < .001) and travel intention (M _Amplifying = 2.47, M _Attenuating = 4.62, t = −10.287, p < .001) significantly lower than those exposed to messages in the risk attenuating frame. Thus, the designed stimuli materials and measurement scales could be used for the formal experiment.

A quasi-experiment design involving a situational experiment and questionnaire survey was applied in the formal experiment. Since face-to-face interviews during COVID-19 would violate the social distancing mandate, a nationwide web-based survey was conducted through a leading market research website (www.wjx.cn) from January 7 to February 4, 2021. A hyperlink to the survey was posted on major social media platforms in China, and participants were invited through snowball and convenience sampling. Specifically, the research team initially forwarded this hyperlink to family, friends, acquaintances, and strangers through social media platforms (e.g., WeChat, QQ), and also posted the hyperlink in social media groups to invite eligible people to participate in the research on a voluntary basis. Participants who accepted the invitation went to the market survey website (www.wjx.cn) after clicking the hyperlink, and were randomly assigned to one of the two frames (risk amplifying and risk attenuating frame) to complete the survey. After completion, participants were instructed to forward the survey hyperlink to people they knew who met the requirements of our experiment. The research team stated in the survey instructions that the data obtained is only for academic research purposes, and that the participants cannot be rewarded after participating in the experiment. To ensure the data quality, the research team clearly informed the research purpose, ensured anonymity, and stressed that there is no right or wrong answer for each item. After being randomly assigned to one of the two frames, participants were asked to answer a series of questions related to perceived safety (before), travel fear (before), travel intention (before) without reading the assigned stimuli materials. After answering these questions, they were presented with the assigned stimuli materials to read, the participants were then asked to answer the questions related to perceived risk, perceived safety (after), travel fear (after), travel intention (after), resilience, and impulsivity, followed by questions on demographic information. Eventually, 600 questionnaires were collected; after eliminating invalid questionnaires due to random filling, pattern answers (e.g., 1234567, 7777777), fast and repeated responses, 481 of the collected questionnaires turned out to be valid cases, indicating a valid response rate of 80.2%. The sample profile is presented in Table 1. Specifically, 42.8% of the respondents were males and 57.2% of them were females; 59.9% were married. Almost 80% had a junior college or higher education level, and more than half of the participants had a monthly income of less than 5,000 CYN. The most frequently chosen travel frequency during COVID-19 was low (77.8%). 23.7% of the respondents were students, and 15.4% were corporate staff. The demographic characteristics of the sample can reflect the social structure of China to a certain extent.

Table 1.

Sample Profiles.

Message Frame		Risk Amplifying Frame (n = 238)		Risk Attenuating Frame (n = 243)		Total (n = 481)
Category		n	%	n	%	n	%
Gender	Male	98	41.2	108	44.4	206	42.8
	Female	140	58.8	135	55.6	275	57.2
Marital status	Married	147	61.8	141	58.0	288	59.9
	Unmarried	91	38.2	102	42.0	193	40.1
Education	Junior high college or below	16	6.7	21	8.6	37	7.7
	Senior high school	31	13.0	32	13.2	63	13.1
	Junior college	29	12.2	33	13.6	62	12.9
	Bachelor’s degree	129	54.2	121	49.8	250	52.0
	Master’s degree or above	33	13.9	36	14.8	69	14.3
Travel frequency	Low	179	75.2	195	80.2	374	77.8
	Medium	50	21.0	46	18.9	96	20.0
	High	9	3.8	2	0.8	11	2.3
Monthly income (CNY)	≤2,500	67	28.2	89	36.6	156	32.4
	2,501–5,000	52	21.8	53	21.8	105	21.8
	5,001–10,000	74	31.1	62	25.5	136	28.3
	10,001–20,000	24	10.1	29	11.9	53	11.0
	≥20,001	21	8.8	10	4.1	31	6.4
Occupation	Corporate staff	39	16.4	35	14.4	74	15.4
	Public servant	31	13.0	22	9.1	53	11.0
	Teacher and researcher	16	6.7	17	7.0	33	6.9
	Self-employed	14	5.9	23	9.5	37	7.7
	Student	45	18.9	69	28.4	114	23.7
	Professional and technical	31	13.0	24	9.9	55	11.4
	Freelance	16	6.7	14	5.8	30	6.2
	Retired	17	7.1	11	4.5	28	5.8
	Soldier	2	0.8	0	0.0	2	0.4
	Other	27	11.3	28	11.5	55	11.4

Results

Manipulation Check

Independent sample t-tests were performed for the manipulation check of perceived risk after exposing the risk messages frame. As expected, participants who were exposed to messages in the risk amplifying frame rated perceived risk significantly higher (M _Amplifying = 5.64, M _Attenuating = 5.23, t = 3.83, p < .001) than those exposed to messages in the risk attenuating frame. Therefore, the results confirmed that the two risk message frames induced different levels of perceived risk among the participants.

Model Validation

Confirmatory factor analysis (CFA) through AMOS 21.0 was performed to examine the convergent validity of each variable. As shown in Table 2, all the model fit indices suggest an acceptable model fit: χ2/df = 2.240 (1<, <3), RMSEA = 0.051 (<0.08), SRMR = 0.049 (<0.08), CFI = 0.966 (>0.9), NFI = 0.940 (>0.9), RFI = 0.930 (>0.9), IFI = 0.966 (>0.9), TLI = 0.960 (>0.9), PNFI = 0.812 (>0.5). The standardized factor loadings and the average variance extracted (AVE) values of all constructs were above 0.5, and the composite reliabilities (CRs) were higher than 0.8, indicating good convergent validity. Moreover, Cronbach’s alpha (CA) value for all variables were over 0.8, indicating good internal consistency reliability.

Table 2.

Confirmatory Factor Analysis Results.

Variables	Items	Mean	Factor Loadings	CA	AVE	CR
Perceived risk	COVID-19 is a serious disease that can kill	5.42	0.727	0.908	0.7004	0.9028
	COVID-19 is more deadly than most people realize	5.57	0.817
	If I were to get the COVID-19 infection, I would experience serious negative consequences	5.37	0.895
	If I were to get the COVID-19 infection, it would have a severe negative impact on my health	5.36	0.897
Perceived safety (before)	The current tourism environment is safe	3.78	0.854	0.937	0.8368	0.9388
	It is safe to travel now	3.61	0.961
	Now I can travel without worry	3.34	0.926
Travel fear (before)	Traveling during the pandemic, I feel anxious	4.38	0.955	0.957	0.8333	0.9523
	Traveling during the pandemic, I feel nervous	4.52	0.934
	Traveling during the pandemic, I feel depressed	4.10	0.887
	Traveling during the pandemic, I feel frightened	4.12	0.873
Travel intention (before)	I plan to travel after the pandemic	3.87	0.861	0.911	0.7803	0.9141
	I would travel with friends after the pandemic	3.81	0.932
	I would recommend to my family to travel after the pandemic	3.96	0.855
Perceived safety (after)	The current tourism environment is safe	3.57	0.910	0.956	0.8803	0.9566
	It is safe to travel now	3.48	0.960
	Now I can travel without worry	3.31	0.944
Travel fear (after)	Traveling during the pandemic, I feel anxious	4.32	0.939	0.973	0.8928	0.9708
	Traveling during the pandemic, I feel nervous	4.41	0.971
	Traveling during the pandemic, I feel depressed	4.21	0.937
	Traveling during the pandemic, I feel frightened	4.15	0.932
Travel intention (after)	I plan to travel after the pandemic	3.88	0.938	0.958	0.8869	0.9592
	I would travel with friends after the pandemic	3.80	0.958
	I would recommend to my family to travel after the pandemic	3.84	0.929
Resilience	I am able to adapt to change brought about by the pandemic	4.62	0.707	0.916	0.6225	0.9076
	I can deal with whatever comes during the pandemic	4.73	0.803
	I can overcome the difficulties caused by the pandemic	4.85	0.811
	I experienced a different experience during the pandemic	5.15	0.778
	I can handle negative feelings during the pandemic	4.95	0.901
	I became stronger during the COVID-19 pandemic	5.01	0.718
Impulsivity	I act “on impulse”	2.74	0.736	0.850	0.5888	0.8512
	I do things with simple consideration	3.32	0.757
	I cannot control my actions well	3.05	0.787
	I act on the spur of the moment	3.16	0.788

Direct Effect Tests

A series of paired sample t-tests was conducted to assess whether there were significant differences in perceived safety, travel fear, travel intention before and after the risk message exposure for each of the two groups. As presented in Table 3, significant differences were found. Specifically, in the case of the risk amplifying frame, participants scored perceived safety (M_Before = 3.51, M_After = 3.12, t = 4.981, p < .001) and travel intention (M_Before = 3.81, M_After = 3.57, t = 3.243, p < .001) before the risk message exposure significantly higher than that after the message exposure, and participants scored travel fear (M_Before = 4.35, M_After = 4.53, t = −2.644, p < .001) before the risk message exposure significantly lower than that after the exposure. These results indicated that the mean values of perceived safety and travel intention reduced significantly, and the mean value of travel fear increased significantly after participants’ exposure to the risk amplifying frame. In the case of the risk attenuating frame, the results were contrastingly different. As shown in Table 3, perceived safety (M_Before = 3.64, M_After = 3.78, t = −2.273, p < .05) and travel intention (M_Before = 3.94, M_After = 4.10, t = −2.405, p < .05) significantly increased, whilst travel fear (M_Before = 4.21, M_After = 4.01, t = 3.521, p < .001) significantly decreased after the risk messaging exposure.

Table 3.

Paired Sample t-Test.

		Before		After
Message Frame	Variable	Mean	SD	Mean	SD.	t Value
isk amplifying frame (N = 238)	Perceived safety	3.51	1.58	3.12	1.63	4.981***
	Travel fear	4.35	1.64	4.53	1.67	−2.644**
	Travel intention	3.81	1.71	3.57	1.72	3.243**
Risk attenuating frame (N = 243)	Perceived safety	3.64	1.48	3.78	1.55	−2.273*
	Travel fear	4.21	1.44	4.01	1.56	3.521***
	Travel intention	3.94	1.65	4.10	1.61	−2.405*

***

p < .001. **p < .01. *p < .05.

To investigate potential differences between the risk message frames in terms of the outcome variables, a series of t-tests and least significant difference (LSD) tests were conducted for post hoc comparisons. As provided in Table 4, there were significant differences between the risk amplifying and risk attenuating frames for perceived safety, travel fear, and travel intention. The post hoc results showed that participants who were exposed to messages in the risk attenuating frame rated perceived safety (M _Amplifying = 3.12, M _Attenuating = 3.78, t = −4.509, p < .001) and travel intention (M _Amplifying = 3.57, M _Attenuating = 4.10, t = −3.489, p < .001) significantly higher than those exposed to messages in the risk amplifying frame. On the other hand, participants exposed to the risk attenuating frame rated travel fear (M _Amplifying = 4.53, M _Attenuating = 4.01, t = 3.535, p < .001) significantly lower than those exposed to the risk amplifying frame. Due to the differences in the sample characteristics between these two groups, in order to avoid interference from participants’ demographics (e.g., monthly income, travel frequency) and improve the validity of the results, we also performed analysis of covariance (ANCOVA) to separate the influence of demographic variables on dependent variables from that of the independent variables. The results showed that risk message frame still significantly predicted perceived safety (F[1,480] = 28.097, p < .001), travel fear (F[1,480] = 12.615, p < .001), and travel intention (F[2,480] = 15.895, p < .001) after controlling for the influence of demographic variables. A paired comparative analysis demonstrated that participants who were exposed to messages in the risk attenuating frame rated perceived safety and travel intention significantly higher than those exposed to messages in the risk amplifying frame, and participants exposed to the risk attenuating frame rated travel fear significantly lower than those exposed to the risk amplifying frame. Therefore, hypotheses 1a, 1b and 1c were supported.

Table 4.

Independent Sample t-Test.

	Risk Amplifying Frame (N = 238)		Risk Attenuating Frame (N = 243)		Total (N = 481)
Outcome variables (after)	Mean	SD	Mean	SD	t value
Perceived safety	3.12	1.63	3.78	1.55	−4.509***
Travel fear	4.53	1.67	4.01	1.56	3.535***
Travel intention	3.57	1.72	4.10	1.61	−3.489***

***

p < .001.

Mediation Effect Tests

In line with suggestions from Hayes and Preacher (2014), the SPSS PROCESS macro (model 4) was employed to perform bootstrapping for the mediation effects of perceived safety and travel fear. This procedure returned mediation effect estimates for multiple mediators, their standard errors, and confidence intervals (CIs) derived from the bootstrap distribution with 5,000 estimates. In the model, the independent variable was risk message frame, the mediation variables were increase in perceived safety (∆perceived safety) and increase in travel fear (∆travel fear), respectively, and the dependent variable was increase in travel intention (∆travel intention). In this analysis, the risk amplifying frame group was taken as the reference group, and a dummy code were created for the risk attenuating frame group. In addition, the demographic variables such as gender, marital status, education, occupation, monthly income, and travel frequency were included as covariates in model estimation.

The results (Table 5) indicated that ∆perceived safety positively predicted ∆travel intention (β = 0.2359, p < .001), and ∆travel fear negatively predicted ∆travel intention (β = −0.1757, p < .001). In terms of between-group differences, the risk attenuating frame group tended to exhibit higher ∆perceived safety (β = 0.5281, p < .001) than the risk amplifying frame group, and tended to exhibit lower ∆travel fear (β = −0.3371, p < .001) than the risk amplifying frame. In addition, since the direct effect of risk message frame on ∆travel intention was not significant (β = 0.1397, p > .05), ∆perceived safety (β = 0.1246, 95% CI: 0.0509, 0.2049) and ∆travel fear (β = 0.0592, 95% CI: 0.0125, 0.1264) fully mediated the relationship between risk message frame and ∆travel intention. Therefore, hypotheses 2a and 2b were supported.

Table 5.

The Mediation Effect of Perceived Safety and Travel Fear.

		∆perceived safety		∆travel fear		∆travel intention
		Β (SE)	95% CI	Β (SE)	95% CI	Β (SE)	95% CI
Direct effect	Gender	0.0146 (0.0995)	−0.1809, 0.2100	0.0945 (0.0899)	−0.0822, 0.2712	−0.0286 (0.0974)	−0.2200, 0.1627
	Marital status	0.0377 (0.1136)	−0.1855, 0.2609	−0.1454 (0.1027)	−0.3472, 0.0563	−0.1008 (0.1113)	−0.3195, 0.1178
	Education	0.0509 (0.0472)	−0.0418, 0.1436	−0.0196 (0.0427)	−0.1034, 0.0643	0.0641 (0.0462)	−0.0267, 0.1549
	Occupation	0.0049 (0.0178)	−0.0300, 0.0398	−0.0203 (0.0161)	−0.0518, 0.0113	−0.0133 (0.0174)	−0.0475, 0.0209
	Monthly income	0.0432 (0.0476)	−0.0503, 0.1368	0.0715 (0.0430)	−0.0131, 0.1561	−0.0323 (0.0468)	−0.1242, 0.0596
	Travel frequency	0.0139 (0.1041)	−0.01907, 0.2186	0.20305281* (0.0941)	0.0180, 0.3880	−0.1877 (0.1024)	−0.3889, 0.0134
	Risk message frame	0.5281*** (0.0972)	0.3371, 0.7191	−0.3371*** (0.0879)	−0.5098, –0.1644	0.1937 (0.0989)	−0.0006, 0.3880
	∆perceived safety	-	-	-	-	0.2359*** (0.0455)	0.1465, 0.3253
	∆travel fear	-	-	-	-	−0.1757*** (0.0503)	−0.2745, –0.0768
	Observation	481		481		481
	R ²	0.0627		0.0709		0.1318
Mediation effect	Message frame→∆perceived safety→∆travel intention	0.1246 (0.0404)	0.0509, 0.2049	-	-	-	-
	Message frame→∆travel fear→∆travel intention	-	-	0.0592 (0.0293)	0.0125, 0.1264	-	-
Total effect	Message frame→∆travel intention	-	-	-	-	0.3775*** (0.0993)	0.1823, 0.5752

Note: Values in paratheses are standard errors.

***

p < .001. *p < .05.

Moderation Effect Tests

Following Wang and Lopez’s (2020) approach, multivariate analysis of variance (MANOVA) was conducted to examine the moderation effects of resilience and impulsivity. To determine the potential moderation effect of resilience, the K-means cluster method was conducted to classify participants into high- and low resilience groups. The mean composite scores for resilience differed significantly between the two groups (M _HRE = 5.649, M _LRE = 3.938, t = 24.447, p < .001). After including gender, marital status, education, occupation, monthly income, and travel frequency as covariates, the results (Table 6) showed that resilience had no significant moderation effect on the relationship between risk message frame and travel fear (F[1,480] = 0.000, p > .1), but had a marginally significant moderation effect on the relationship between risk message frame and perceived safety (F[1,480] = 2.785, p < .1) as well as that between risk message frame and travel intention (F[1,480] = 3.248, p < .1). As shown in Figure 2, regardless of risk message frame (i.e., risk amplifying or attenuating frame), high-resilience participants scored higher in perceived safety and travel intention than low-resilience participants. Therefore, hypothesis 3a and 3b were supported.

Table 6.

The Moderation Effect of Resilience.

Variables		Perceived Safety		Travel Fear		Travel Intention
Variables		F value		F value		F value
Covariates	Gender	19.108**	9.015	17.966**	7.431	4.620	1.862
	Marital status	7.828^†	3.693	42.829***	17.714	18.251**	7.358
	Education	20.953**	9.885	0.014	0.006	7.814^†	3.150
	Occupation	0.843	0.398	0.455	0.188	3.377	1.361
	Monthly income	0.154	0.072	0.810	0.335	0.351	0.142
	Travel frequency	67.596***	31.889	20.231**	8.368	50.828***	20.490
Direct effect
Independent variables	Message frame	59.127***	27.894	30.074***	12.439	38.653***	15.582
	Resilience	17.496**	8.254	0.050	0.021	19.610**	7.905
Moderation effect
Interaction term	Message frame × resilience	5.902^†	2.785	0.001	0.000	8.058^†	3.248

***

p < .001. **p < .01. ^†p < .1.

Figure 2.

The moderation effect of resilience.

To determine the potential moderation effect of impulsivity, the K-means cluster method was conducted to classify participants into high- and low impulsivity groups. The mean composite scores for impulsivity differed significantly between the two groups (M _HIM = 4.09, M _LIM = 2.08, t = 30.776, p < .001). After including gender, marital status, education, occupation, monthly income, and travel frequency as covariates, the results (Table 7) showed that impulsivity had no significant moderation effect on the relationship between risk message frame and perceived safety (F[1,480] = 1.130, p > .1) as well as that between risk message frame and travel intention (F[1,480] = 0.150, p > .1); however, impulsivity had a significant moderation effect on the relationship between risk message frame and travel fear (F[1,480] = 12.615, p < .001). As shown in Figure 3, low-impulsivity participants’ travel fear in response to high COVID-19 messages is higher than that of high-impulsivity participants, while low-impulsivity participants’ travel fear in response to low COVID-19 message is lower than that of high-impulsivity participants. Therefore, hypothesis 4c was supported.

Table 7.

The Moderation Effect of Impulsivity.

Variables		Perceived Safety		Travel Fear		Travel Intention
Variables		F Value		F Value		F Value
Covariates	Gender	29.075**	13.931	20.543**	8.736	9.827*	3.986
	Marital status	3.970	1.902	40.771***	17.338	12.15*	4.928
	Education	29.874***	14.313	0.044	0.019	13.543*	5.493
	Occupation	0.049	0.023	0.795	0.338	1.184	0.480
	Monthly income	0.865	0.415	0.394	0.168	1.137	0.461
	Travel frequency	67.885***	32.526	17.457**	7.423	52.197***	21.171
Direct effect
Independent variables	Message frame	54.478***	26.102	29.882***	12.707	35.523***	14.409
	Impulsivity	37.466***	17.951	1.174	0.499	35.32***	14.326
Moderation effect
Interaction term	Message frame × impulsivity	2.359	1.130	29.666***	12.615	0.370	0.150

***

p < .001. **p < .01. *p < .05.

Figure 3.

The moderation effect of impulsivity.

Conclusions and Discussion

Conclusions

Based on frame theory and in the context of the COVID-19 pandemic, this study investigated the impact of risk message framing on perceived safety, travel fear, and post-pandemic travel intention. The moderation effects of resilience and impulsivity as personal traits on the above relationships were also examined. The main conclusions are as follows.

Firstly, risk message framing significantly predicted tourists’ perceived safety, travel fear, and travel intention. The results showed that perceived safety and travel intention in the risk attenuating frame were greater than that in the risk amplifying frame, and travel fear in the risk attenuating frame was lower than that in the risk amplifying frame. These findings indicated that risk message framing is a critical factor influencing tourists’ perceived safety, travel fear, and travel intention during a major global health crisis such as the COVID-19 pandemic. Communications and presentation of risk messages in a risk amplifying or risk attenuating frame at destinations triggered differentiated responses in tourists’ perceptions, emotions, and behaviors. These results are consistent with those of Kapuściński and Richards (2016), Liu-Lastres, Schroeder, and Pennington-Gray (2019), and Xie et al. (2021), providing enriching and expanding evidence regarding the relationship between risk message framing and tourists’ perceptions, attitudes and behaviors.

Secondly, tourists’ perceived safety and travel fear mediated the impact of risk messages on travel intention. Specifically, perceived safety and travel fear fully mediated the impact of risk messages on travel intention, suggesting that they are respectively cognitive construct and emotional construct that mediate the behavioral impact of risk messages in crisis situations. Previous research has investigated the mediation of perceived safety between risk message and travel intention (Liu-Lastres, Schroeder, and Pennington-Gray 2019; Zhang et al. 2022), as well as the mediation of worry between perceived risk and self-protective behavior (Wang et al. 2019). On the basis of the previous research (Liu-Lastres, Schroeder, and Pennington-Gray 2019; Wang et al. 2019; Zhang et al. 2022), this study demonstrated both the cognitive and the emotional mediation processes of risk messages affecting travel intention in long-term high-risk situations, thus presenting a significant contribution to this line of research.

Thirdly, the trait of resilience had a marginally significant moderation effect on the relationship between risk messages and tourists’ perceived safety/travel intention. Specifically, regardless of message frames (risk amplifying or risk attenuating), high-resilience tourists scored higher on perceived safety and travel intention than low-resilience tourists. This indicates that resilient tourists have strong resistibility and adaptability to destination risk communication, and thereby maintaining strong perceived safety and travel intention in crisis situations. However, the moderation effect of resilience is marginally significant, and the effect is stronger in the case of the risk attenuating frame, suggesting that resilience has limited mitigation effect on the negative impact of high risk messages. Previous research has empirically confirmed the strengthening effect of resilience on tourists’ decision-making in crisis situations (Zheng, Luo, and Ritchie 2021), which is logically consistent with our result. In addition, fear is one of the basic emotions, and travel fear is tourists’ defensive and preparatory response to a present threat and high-risk situation (Cisler, Olatunji, and Lohr 2009; Fennell 2017). Our study found that travel fear as tourists’ defensive response to risks was not dependent on the individual trait of resilience. The reason may be that the effect of resilience is to help tourists stay calm and adaptable in threatening situations after the travel fear is triggered, and take positive actions to cope with, rather than weakening tourists’ basic emotional response to risk messages.

Finally, the trait of impulsivity had a significant moderation effect on the relationship between risk messages and travel fear. High-impulsivity tourists scored lower on travel fear than low-impulsivity tourists in the case of the risk amplifying frame, which is basically consistent with the viewpoint of cognitive dissonance. Specifically, to seek novel, adventurous, exciting, and unusual tourism experiences in high risk situations, high-impulsivity tourists tended to lower their travel fear through numerous means of rationalization, thereby reducing cognitive dissonance (Festinger 1957; Fuchs et al. 2013). However, high-impulsivity tourists scored higher on travel fear than low-impulsivity tourists in the risk attenuating frame. The possible reason may be that the evaluation of high-impulsivity tourists on risk message is often dominated by emotion-driven, intuition-based, and less effort-oriented heuristic processing (Karl, Chien, and Ong 2021; McCabe, Li, and Chen 2016). Accordingly, their reactions and decisions are lack of planning, and their ability to cope with emergency and negative events is insufficient (Whiteside and Lynam 2001). Thus, high-impulsivity tourists are concerned and worried about the unexpected situations and uncertainty during the journey, even in a low risk situation. In summary, the co-existence of sensation seeking and heuristic information processing leads to this “scissors difference” pattern in the moderation effect of impulsivity as a personal trait on the relationship between risk message and travel fear. In addition, impulsivity can be understood as a psychological tendency with strong emotional response and weak rational control (Moeller et al. 2001; Patton, Stanford, and Barratt 1995). On the other hand, perceived safety in most cases reflects a comprehensive judgment and rational evaluation of safety and risk information based on people’s past experiences and knowledge, serving as the basis for tourists to avoid risks, change attitudes, and adjust behaviors (Sano and Sano 2019; Wang and Lopez 2020; Xie et al. 2021). In this regard, the irrational cognitive filtering of impulsive tourists toward risk message may make little difference on the impact of risk messages on perceivedsafety.

Theoretical Implications

This study revealed the mediation mechanism between risk message frame and travel intention from the integrated perspective of cognition and emotion, which provides new insights for understanding the frame effect of risk message, thereby providing theoretical basis and empirical support for the agenda-setting of destination risk message communication. Currently, tourists’ travel decision-making in crisis situations has received considerable research attention; however, travel intention in crisis situations and how risk perceptions can influence travel intention remain to be a contested topic among scholars (Biran et al. 2014; Sano and Sano 2019; Wang et al. 2019; Wolff and Larsen 2014). Although the direct impacts of risk message framing on tourists’ perceived risk/safety and travel intention have been empirically examined (Kapuściński and Richards 2016; Xie et al. 2021), there is still little research on tourists’ emotional response to risk message frame. Therefore, the possible mediation mechanisms between risk messages and travel intention need more empirical investigation, so as to provide clarification over this contested topic. On this basis, this study confirmed the mediation roles of perceived safety and travel fear between risk message frame and travel intention, revealing both the cognitive and emotional mediation processes through these two constructs. As such, one theoretical contribution of this study is that it reveals the dual-mediation mechanism of risk message frame affecting travel intention, by identifying perceived safety as a cognitive mediator and travel fear as an emotional mediator in the relationship between risk message and travel intention.

Based on person-situation interaction perspective, this study also verified the differentiating moderation effects of resilience and impulsivity as relevant personal traits on the impacts of risk messages, thus providing new perspective for analyzing the effects of risk messages. Resilience has been widely applied to investigate how destinations respond to crises and disasters (Fountain and Cradock-Henry 2020), and it has also been applied to examine how tourists respond to adversities, crises, and disasters (Prayag 2018; Zheng, Luo, and Ritchie 2021). However, limited research has investigated how resilient tourists evaluate the risk messages and their subsequent decision-making; specifically, the moderation effect of resilience in tourists’ decision-making under crisis situations lacks empirical investigation. Though tourist impulse buying has received considerable attention in tourism research (Li 2007; Li et al. 2021), few studies have explored the moderation effect of impulsivity on tourists’ responses in crisis situations. Since the connotation of impulsivity is closely related to factors such as sensation seeking, risk-taking, novelty-pursuing, and adventure orientation (Eysenck et al. 1985; Whiteside and Lynam 2001; Zuckerman et al. 1993), the evaluations and decisions of impulsive tourists are often dominated by impulsive thinking, which adopts the intuition-based heuristic processing of risk messages (Karl, Chien, and Ong 2021; McCabe, Li, and Chen 2016). Investigating the moderation effects of resilience and impulsivity in tourists’ responses to risk message can help destinations to cater to tourists’ psychological characteristics in tourism recovery during and after the pandemic. In this regard, this study confirmed the beneficial effect of resilience on perceived safety and travel intention, as well as the “scissors difference” moderating effect of impulsivity on the relationship between risk messages and travel fear. These two personal traits seem to play critical roles in moderating the effects of risk messages in crises. Therefore, the study extended the research on the roles of resilience and impulsivity especially in destination risk communications, thereby providing a valuable case for future studies investigating tourists’ travel decisions in crisis situations.

Practical Implications

First, destination management organizations (DMOs) should enhance the agenda-setting and frame constructing of risk messages, and adopt risk attenuating frames to disseminate messages. Specifically, DMOs should provide sufficient positive messages in risk communications, focusing on the controllability of the crisis, the effectiveness of organizational crisis response, and the safety status of the destination. DMOs need to provide risk guidance, protective instruction, and travel advisory to the public, guiding them to form a reasonable level of perceived safety and travel fear. For example, during a major crisis such as COVID-19, DMOs should encourage tourists to maintain social distance, make online appointments, and avoid crowded attractions, and inform them of the safety and risk status in different areas of the destination. Moreover, DMOs should help the public to objectively understand risk messages, increase their knowledge and ability to process and respond to crisis and risk messages, and encourage them to take actions to deal with emerging safety and health issues. In the new media era, DMOs should adopt mass media and social media tools to establish a professional risk message release system to prevent event crises from turning into public opinion crisis, thereby promoting rapid tourism market recovery.

Second, DMOs may develop post-crisis marketing strategies and recovery plans based on tourists’ personal traits. Specifically, resilient tourists presented higher perceived safety and travel intention, and impulsive tourists showed lower risk concern and worry. Thus, DMOs should develop targeted marketing strategies for such tourists after crisis, so as to promote tourism recovery. For example, DMOs can foster the travel intention of high-resilience and high-impulsivity tourists by adopting marketing tools such as free tickets and product discounts. Moreover, DMOs can tailor their offers to match the needs of impulsive tourists by developing and providing new and novel tourism products. For example, DMOs can develop dark tourism products based on crisis events, and reproduce crises and disaster scenarios using technologies such as virtual reality (VR), augmented reality (AR), thereby bringing novel and exciting experiences to these tourists.

Limitations and Future Research

This study has several limitations. First, this study took Chinese tourists in the COVID-19 pandemic as research subjects and may be limited by the research context and subjects. Future research may validate and expand the conclusions in different crises context and/or with tourists from other cultural backgrounds. Second, this study used convenience sampling and snowball sampling in its data collection; the sample method may not be solid enough to generate a representative sample for the study. Also, although travel intention is a strong predictor of actual travel behavior, intention may not necessarily lead to actual travel. Future research may adopt a better sampling method and use multi-wave longitudinal survey design to investigate the impact of intra-pandemic risk message on the post-pandemic actual behavior of tourists (e.g., online booking, safety behavior) rather than travel intention. Third, although the study attempted to control bias from stimuli materials, the stimuli materials in different message frames may compromise the reliability of the findings. Future research should therefore either optimize the experimental design and stimuli materials or use alternative methods (e.g., machine learning, structural equation modeling) to test the conceptual model.

Supplemental Material

sj-docx-1-jtr-10.1177_00472875221095212 – Supplemental material for Effect of Risk Message Framing on Tourists’ Travel Intention: Roles of Resilience and Impulsivity

Supplemental material, sj-docx-1-jtr-10.1177_00472875221095212 for Effect of Risk Message Framing on Tourists’ Travel Intention: Roles of Resilience and Impulsivity by Chaowu Xie, Jiangchi Zhang and Songshan (Sam) Huang in Journal of Travel Research

Footnotes

Declaration of Conflicting Interests

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

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Funding supported by the National Natural Science Foundation of China (Grant No. 41971182) Humanities and Social Sciences Foundation of Ministry of Education of China (Grant No. 19YJAZH097) Innovation Strategy Research Project of Science and Technology Department of Fujian Province (2021R0058).

ORCID iDs

Chaowu Xie

Jiangchi Zhang

Songshan (Sam) Huang

Supplemental Material

Supplemental material for this article is available online.

Author Biographies

Chaowu Xie, is a Professor of college of Tourism at Huaqiao University (China). His research interests include safety and security in tourism, smart tourism, and hospitality management.

Jiangchi Zhang, Corresponding Author, is a PhD student at Huaqiao University (China). His research interests include safety and security in tourism and hospitality management.

Songshan (Sam) Huang, PhD, is a research professor in Tourism and Services Marketing in the School of Business and Law, Edith Cowan University (Australia). His research interests include tourist behavior, destination marketing, tour guiding and various China tourism and hospitality issues.

References

Aliperti

Cruz

A. M.

2019. “Investigating Tourists’ Risk Information Processing.” Annals of Tourism Research 79:1–18.

Bandura

1977. Social Learning Theory. Englewood Cliffs, NJ: Prentice Hall.

Barlow

D. H.

2002. Anxiety and Its Disorders: The Nature and Treatment of Anxiety and Panic. New York, NY: Guilford Press.

Biran

Liu

Eichhorn

2014. “Consuming Post-Disaster Destinations: The Case of Sichuan, China.” Annals of Tourism Research 47:1–17.

Chaiken

1980. “Heuristic Versus Systematic Information Processing and the Use of Source Versus Message Cues in Persuasion.” Journal of Personality and Social Psychology 39 (5): 752–66.

Chauhan

2007. “Safety and Security Perceptions of Tourists Visiting Kashmir, India.” Advances in Hospitality and Leisure 3:3–17.

Cisler

J. M.

Olatunji

B. O.

Lohr

J. M.

2009. “Disgust, Fear, and the Anxiety Disorders: A Critical Review.” Clinical Psychology Review 29:34–46.

Coombs

W. T.

2007. “Protecting Organization Reputations During a Crisis: The Development and Application of Situational Crisis Communication Theory.” Corporate Reputation Review 10 (3): 163–76.

Covello

V. T.

1992. “Risk Communication: An Emerging Area of Health Communication Research.” Annals of the International Communication Association 15 (1): 359–73.

10.

Dai

Y. D.

Zhuang

W. L.

Huan

T. C.

2019. “Engage or Quit? The Moderating Role of Abusive Supervision Between Resilience, Intention to Leave and Work Engagement.” Tourism Management 70:69–77.

11.

Druckman

J. N.

2001. “The Implications of Framing Effects for Citizen Competence.” Political Behavior 23 (3): 225–56.

12.

Dyer

J. G.

McGuinness

T. M.

1996. “Resilience: Analysis of the Concept.” Archives of Psychiatric Nursing 10 (5): 276–82.

13.

Entman

R. M.

1993. “Framing: Toward Clarification of a Fractured Paradigm.” Journal of Communication 43 (4): 51–8.

14.

Epstein

1972. “The Nature of Anxiety With Emphasis Upon Its Relationship to Expectancy.” In Anxiety: Current Trends in Theory and Research, vol. 2, edited by Spielberger

C. D.

, 291–337. New York, NY: Academic Press.

15.

Eysenck

S. B. G.

Pearson

P. R.

Easting

Allsopp

J. F.

1985. “Age Norms for Impulsiveness, Venturesomeness, and Empathy in Adults.” Personality and Individual Differences 6:613–9.

16.

Fennell

D. A.

2017. “Towards a Model of Travel Fear.” Annals of Tourism Research 66 (9): 140–50.

17.

Festinger

1957. A Theory of Cognitive Dissonance. Evanston, IL: Row Peterson.

18.

Fong

L. H. N.

Law

B. H.

2020. “Outlook of Tourism Recovery Amid an Epidemic: Importance of Outbreak Control by the Government.” Annals of Tourism Research 86:102951.

19.

Fountain

Cradock-Henry

N. A.

2020. “Recovery, Risk and Resilience: Post-Disaster Tourism Experiences in Kaikōura, New Zealand.” Tourism Management Perspectives 35:1–11.

20.

Freimuth

Linnan

H. W.

Potter

2000. “Communicating the Threat of Emerging Infections to the Public.” Emerging Infectious Diseases 6 (4): 337–47.

21.

Fuchs

Uriely

Reichel

Maoz

2013. “Vacationing in a Terror-Stricken Destination: Tourists’ Risk Perceptions and Rationalizations.” Journal of Travel Research 52 (2): 182–91.

22.

Gamson

W. A.

Modigliani

1989. “Media Discourse and Public Opinion on Nuclear Power: A Constructionist Approach.” American Journal of Sociology 95 (1): 1–37.

23.

George

2010. “Visitor Perceptions of Crime-Safety and Attitudes Towards Risk: The Case of Table Mountain National Park, Cape Town.” Tourism Management 31 (6): 806–15.

24.

Gillespie

B. M.

Chaboyer

Wallis

2007. “Development of a Theoretically Derived Model of Resilience Through Concept Analysis.” Contemporary Nurse 25 (1-2): 124–35.

25.

Gitlin

1980. The Wwhole Wworld is Wwatching: Mass Mmedia in the Mmaking and Uunmaking of the Nnew Lleft. Berkeley, CA: University of California Press.

26.

Goffman

1974. Frame Analysis: An Essay on the Organization of Experience. New York, NY: Harper & Row.

27.

Govers

Van Hecke

Cabus

2008. “Delineating Tourism: Defining the Usual Environment.” Annals of Tourism Research 35 (4): 1053–73.

28.

Goyal

Verma

2021. “Do It Yourself Effect on Restaurants – The Pandemic Effect: Driven by the Fear Appeal Theory.” Journal of Hospitality and Tourism Management 48:76–87.

29.

Hayes

A. F.

Preacher

K. J.

2014. “Statistical Mediation Analysis With a Multicategorical Independent Variable.” British Journal of Mathematical and Statistical Psychology 67 (3): 451–70.

30.

Hew

J. J.

Leong

L. Y.

Tan

G. W. H.

Lee

V. H.

Ooi

K. B.

2018. “Mobile Social Tourism Shopping: A Dual-Stage Analysis of a Multi-Mediation Model.” Tourism Management 66:121–39.

31.

Kahneman

Tversky

1984. “Choices, Values, and Frames.” American Psychologist 39:341–50.

32.

Kapuściński

Richards

2016. “News Framing Effects on Destination Risk Perception.” Tourism Management 57:234–44.

33.

Karl

Chien

P. M.

Ong

2021. “Impulse Buying Behaviour in Tourism: A New Perspective.” Annals of Tourism Research 90:1–4.

34.

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.

35.

Kuntz

J. R. C.

Malinen

Näswall

2017. “Employee Resilience: Directions for Resilience Development.” Consulting Psychology Journal Practice and Research 69 (3): 223–42.

36.

Law

Buhalis

Cobanoglu

2014. “Progress on Information and Communication Technologies in Hospitality and Tourism.” European Journal of Marketing 26 (5): 727–50.

37.

Wang

2021. “To Buy or Not to Buy? The Effect of Time Scarcity and Travel Experience on Tourists’ Impulse Buying.” Annals of Tourism Research 86:1–13.

38.

Liu-Lastres

Schroeder

Pennington-Gray

2019. “Cruise Line Customers’ Responses to Risk and Crisis Communication Messages: An Application of the Risk Perception Attitude Framework.” Journal of Travel Research 58 (5): 849–65.

39.

Fei

Zhang

Y. L.

Yang

S. J.

Tong

Y. S.

Wang

Z. Q.

Niu

Y. J.

2011. “Reliability and Validity of an Adapted Chinese Version of Barratt Impulsivity Scale.” Chinese Mental Health Journal 25 (8): 610–5.

40.

2007. “Impact of Nonresident on Impulse Buying: An Empirical Research on Tourists.” Nankai Business Review 10 (6): 11–8.

41.

Luo

Zhai

2017. “‘I Will Never Go to Hong Kong Again!’ How the Secondary Crisis Communication of ‘Occupy Central’ on Weibo Shifted to a Tourism Boycott.” Tourism Management 62:159–72.

42.

McCabe

Chen

2016. “Time for a Radical Reappraisal of Tourist Decision Making? Toward a New Conceptual Model.” Journal of Travel Research 55 (1): 3–15.

43.

Mehrabian

Russell

J. A.

1974. An Approach to Environmental Psychology. Cambridge, MA: The MIT Press.

44.

Meng

Zhang

K. K. F.

2019. “Modeling Precursors of Impulsive Tourist Shopping Behavior: Evidence From Long-Haul Chinese Outbound Tourists.” International Journal of Tourism Research 21 (3): 344–58.

45.

Moeller

F. G.

Barratt

E. S.

Dougherty

D. M.

Schmitz

J. M.

Swann

A. C.

2001. “Psychiatric Aspects of Impulsivity.” American Journal of Psychiatry 158 (11): 1783–93.

46.

Nelson

T. E.

Lecheler

Schuck

R. T.

De Vreese

C. H.

2012. Framing Effects on Values, Emotions, and Attitudes. Portland, OR: Western Political Science Association.

47.

Nguyen

Kuntz

J. R. C.

Naswall

Malinen

2016. “Employee Resilience and Leadership Styles: The Moderating Role of Proactive Personality and Optimism.” New Zealand Journal of Psychology 45 (2): 13–21.

48.

Oliveira

Huertas

2019. “How Do Destinations Use Twitter to Recover Their Images After a Terrorist Attack?” Journal of Destination Marketing & Management 12:46–54.

49.

Page

S. J.

Bentley

T. A.

Walker

2005. “Scoping the Nature and Extent of Adventure Tourism Operations in Scotland: How Safe Are They?” Tourism Management 26 (3): 381–97.

50.

Palmer

J. C.

Komarraju

Carter

M. Z.

Karau

S. J.

2017. “Angel on One Shoulder: Can Perceived Organizational Support Moderate the Relationship Between the Dark Triad Traits and Counterproductive Work Behavior?” Personality and Individual Differences 110:31–7.

51.

Patton

J. H.

Stanford

M. S.

Barratt

E. S.

1995. “Factor Structure of the Barratt Impulsiveness Scale.” Journal of Clinical Psychology 51 (6): 768–74.

52.

Pizam

Smith

2000. “Tourism and Terrorism: A Quantitative Analysis of Major Terrorist Acts and Their Impact on Tourism Destinations.” Tourism Economics 6 (2): 123–38.

53.

Prayag

2018. “Symbiotic Relationship or Not? Understanding Resilience and Crisis Management in Tourism.” Tourism Management Perspectives 25:133–5.

54.

Reisinger

Mavondo

2005. “Travel Anxiety and Intentions to Travel Internationally: Implications of Travel Risk Perception.” Journal of Travel Research 43 (3): 212–25.

55.

Renn

Burns

W. J.

Kasperson

J. X.

Kasperson

R. E.

Slovic

1992. “The Social Amplification of Risk: Theoretical Foundations and Empirical Applications.” Journal of Social Issues 48 (4): 137–60.

56.

Reynolds

Seeger

2005. “Crisis and Emergency Risk Communication as an Integrative Model.” Journal of Health Communication 10 (1): 43–55.

57.

Rimal

R. N.

2003. “Perceived Risk and Efficacy Beliefs as Motivators of Change: Use of the Risk Perception Attitude (RPA) Framework to Understand Health Behaviors.” Human Communication Research 29 (3): 370–99.

58.

Ryu

Kim

2015. “Testing the Heuristic/Systematic Information-Processing Model (HSM) on the Perception of Risk After the Fukushima Nuclear Accidents.” Journal of Risk Research 18 (7): 840–59.

59.

Sano

2019. “The Effect of Different Crisis Communication Channels.” Annals of Tourism Research 79:1–12.

60.

Smith

B. W.

Dalen

Wiggins

Tooley

Christopher

Bernard

2008. “The Brief Resilience Scale: Assessing the Ability to Bounce Back.” International Journal of Behavioral Medicine 15 (3): 194–200.

61.

Sniderman

P. M.

Theriault

S. M.

2004. “The Structure of Political Argument and the Logic of Issues Framing.” In Studies in Public Opinion: Attitudes, Non-attitudes, Measurement Error, and Change, edited by Saris

W. E.

Sniderman

P. M.

, 133–65. Princeton, NJ: Princeton University Press.

62.

Song

Yao

Wen

2021. “What Motivates Chinese Consumers to Avoid Information About the COVID-19 Pandemic? The Perspective of the Stimulus-Organism-Response Model.” Information Processing & Management 58 (1): 102407.

63.

Sparks

B. A.

Browning

2011. “The Impact of Online Reviews on Hotel Booking Intentions and Perception of Trust.” Tourism Management 32 (6): 1310–23.

64.

Swanson

S. R.

2017. “The Effect of Destination Social Responsibility on Tourist Environmentally Responsible Behavior: Compared Analysis of First-Time and Repeat Tourists.” Tourism Management 60:308–21.

65.

Sylvers

Lilienfeld

S. O.

LaPrairie

J. L.

2011. “Differences Between Trait Fear and Trait Anxiety: Implications for Psychopathology.” Clinical Psychology Review 31 (1): 122–37.

66.

Tankard

J. W.

2001. “The Empirical Approach to the Study of Media Framing.” In Framing Public Life: Perspectives on Media and Our Understanding of the Social World, edited by Reese

S. D.

Gandy

O. H.

, 111–21. Mahwah, NJ: Lawrence Erlbaum.

67.

Wagnild

G. M.

Young

H. M.

1993. “Development and Psychometric Evaluation of the Resilience Scale.” Journal of Nursing Measurement 1:165–78.

68.

Wang

Lopez

2020. “Does Communicating Safety Matter?” Annals of Tourism Research 80:1–12.

69.

Wang

Liu-Lastres

Ritchie

B. W.

Pan

D. Z.

2019. “Risk Reduction and Adventure Tourism Safety: An Extension of the Risk Perception Attitude Framework (RPAF).” Tourism Management 74:247–57.

70.

Whiteside

S. P.

Lynam

D. R.

2001. “The Five Factor Model and Impulsivity: Using a Structural Model of Personality to Understand Impulsivity.” Personality and Individual Differences 30 (4): 669–89.

71.

Wolff

Larsen

2014. “Can Terrorism Make Us Feel Safer? Risk Perceptions and Worries Before and After the July 22nd Attacks.” Annals of Tourism Research 44 (1): 200–9.

72.

Xue

Morrison

A. M.

Leung

X. Y.

2012. “Frame Analysis on Golden Week Policy Reform in China.” Annals of Tourism Research 39 (2): 842–62.

73.

Xie

Zhang

Huang

Chen

Morrison

A. M.

2022. “An Analysis of User-Generated Crisis Frames: Online Public Responses to a Tourism Crisis.” Tourism Management Perspectives 41:1–16.

74.

Xie

Zhang

Morrison

A. M.

2021. “Developing a Scale to Measure Tourist Perceived Safety.” Journal of Travel Research 60:1232–51.

75.

Xie

Zhang

Morrison

A. M.

CocaStefaniak

J. A.

2021. “The Effects of Risk Message Frames on Post-Pandemic Travel Intentions: The Moderation of Empathy and Perceived Waiting Time.” Current Issues in Tourism 24 (23): 3387–406.

76.

Zhang

2019. “A Structural Model of Liminal Experience in Tourism.” Tourism Management 71:84–98.

77.

Zhang

Xie

Chen

Dai

Y. D.

Yi-Jun

2022. “The Matching Effect of Destinations’ Crisis Communication.” Journal of Travel Research 1:1–26.

78.

Zhang

G. Y.

Gursoy

X. R.

2018. “Message Framing and Regulatory Focus Effects on Destination Image Formation.” Tourism Management 69:397–407.

79.

Zheng

Luo

Ritchie

B. W.

2021. “Afraid to Travel After COVID-19? Self-Protection, Coping and Resilience Against Pandemic ‘Travel Fear’.” Tourism Management 83:1–13.

80.

Zhu

Zhao

Zhou

2019. “Resilience in Organizations: Construction of Protective Resources From Psychological and Systematic Perspective.” Advances in Psychological Science 27 (2): 357–69.

81.

Zuckerman

Kuhlman

D. M.

Joireman

Teta

1993. “A Comparison of Three Structural Models for Personality: The Big Three, the Big Five, and the Alternative Five.” Journal of Personality and Social Psychology 65:757–68.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.01 MB