Impact of risk aversion,reactance proneness and risk appraisal on travel destination risk perception

Abstract

The aim of the study is to investigate the motivational effects of tourist traits and risk appraisal on tourist destination risk perception. Risk appraisal involves subjective estimates of vulnerability to a threat and the threat’s consequential severity. Fear levels influence both of these elements of risk appraisal. Individual differences in reactance proneness and risk aversion are introduced into the study model to more fully account for differences in travel destination risk perceptions. The study design involves US adults, who have used their passports for international travel in the past 5 years. Travel risk assessments were studied for four destination sites: London, Tokyo, Kuala Lumpur and Istanbul. A general structural model is developed to test hypotheses about antecedents and consequents of risk appraisal and destination risk perception.

Keywords

Destination risk reactance proneness risk appraisal risk aversion

Introduction

The worldwide threat of the COVID-19 pandemic has resulted in a shift of tourist travel perspectives. Destinations that may have once been considered safe may now pose potentially serious perceived risks for tourists. In the age of COVID-19, updated tourist information about the threat of destination risk is blatantly inconsistent with past destination tourist risk impressions (Sheeran et al., 2014; Weber et al., 2002).

The current study is an exposition of the general process and prediction of travel destination risk perception. The focal conceptual framework involves protection motivation theory. Protection Motivation Theory (PMT) was developed by Rogers (1975, 1983). The initial PMT model was applied to disease prevention and health promotion. The theory has been used to motivate people to engage in preventive health-related behaviors, and in early detection examinations. Much PMT research has involved individual responses to various threats. Selected factors from this framework are used to generate central study concepts. In PMT, factors are classified as threat appraisal or coping appraisal. In this study, threat appraisal is used to formulate hypotheses involving variations in destination risk perceptions. Coping appraisals (self-efficacy and protective efficacy) are not included since they explain the adoption of protective behaviors.

PMT has been used in a variety of tourism and travel articles. Previous studies have involved travel intentions as a key dependent variable. These studies have measured the impact of perceived risk on the avoidance of host cities, nations or regions (Law, 2006; Qi et al., 2009; Schroeder et al., 2013; Sonmez and Graefe, 1998). These studies include only a single factor involving PMT threat appraisal. For example, some address only perceived vulnerability, while others restrict analyses to perceived severity. The present study is a unique contribution in that it breaks new ground by explaining perceived travel destination risk as a dependent variable on the basis of all elements of threat appraisal, and antecedent conditions.

An important qualifying factor in determining the relation between risk appraisal and actual risk taking is the denial or acceptance of personal vulnerability assessment, and levels of risk avoidance (Ferrer et al., 2018; Mills et al., 2008). In the area of tourist risk appraisal, there is the concept of a critical risk value, beyond which travel intentions will diminish. Different tourists will exhibit different critical risk values (Cui et al., 2016).

A study by Caber et al. (2020) among German tourists visiting Turkey, investigated risk perceptions of travel to Spain or Greece. They confirmed the negative effects of risk perception on destination attitudes. Destination risks were assessed at the country level, rather than localized cities or communities. However, no antecedent psychological traits were introduced in their research. This omission limited their study in that different tourist traits predictive of travel risk assessments were not addressed in their research.

In the following sections, the conceptual and literature summary is available for each study variable. This includes the study design for testing hypotheses about factors affecting perceived destination risk. A general model of destination perceived risk for varied destinations is advanced and tested. Theoretical and managerial implications of the study’s results are disclosed. We conclude with limitations and suggested future research.

Conceptual background and hypotheses

Threat appraisal

PMT is invoked to measure health destination threat appraisals of international destination travelers. Threat appraisal, or risk under PMT, involves health-related susceptibility and severity expectations relating to destination travel. There is a widespread belief that anyone may be vulnerable, to some degree, for contracting infectious disease (Norman et al., 2005; Wang et al., 2019). Other studies involving risk-related tourist vulnerability and severity have included only one of these risk appraisal variables. Studies of travel risk may completely bypass threat appraisal phenomena and deal exclusively with coping appraisal (Wang et al., 2019).

Threat perceptions tend to generate adaptive behaviors. Perceived susceptibility and severity levels change for individuals as new information is absorbed. In the case of pandemics, beliefs about adverse outcomes enter into threat appraisal, or riskiness, associated with international travel.

H1: Expectations of vulnerability to COVID-19 illness have a positive influence on destination risk perceptions.

H2: Expectations of dire symptoms and outcomes if COVID-19 was developed will have a positive influence on destination risk perceptions.

Fear arousal

Sunstein (2003) asserts that affective risk perceptions are antecedents of cognitive risk perceptions. Emotions such as fear are influential in reactions to risky situations (Lowenstein et al., 2001). The risk-as-feelings hypothesis states that both affective and cognitive risk perceptions influence decision-making.

Extreme uncertainty about the possibility of escaping an unpleasant outcome generates fear. The appraisal dimension of adverse external circumstances is driven by fear. Agency, or external forces, when combined with self-assessments of powerlessness plays an important role in fear elicitation.

Fear arousal activates motivations to act in a self-protective manner (Tanner et al., 1991). This is a feedback mechanism that accelerates the processing of threat information. Fear-generated threat appraisal results in heightened attention and understanding of threat-related information. In essence, when fear insistently stokes threat appraisal, the result is the activation of self-protective motivation.

H3: Fear associated with the COVID-19 pandemic has a positive influence on elements of threat appraisal—vulnerability and severity.

Reactance proneness

Reactance proneness is a psychological trait capable of intensifying anger and counter-argumentation. Trait reactance more generally reflects an individual’s desire for autonomy. This trait has a positive association with denial, self-sufficiency and lack of conformity. Trait reactance involves extreme sensitivity to central and peripheral threats to freedom, especially for salient issues (Quick and Stephenson, 2008; Quick et al., 2011). High trait reactance individuals are more likely to associate to COVID-19 contagion as a potential threat to freedom.

The PMT model specifies personality variables that may be operative, but does not reference reactance proneness in particular. Under PMT, threat appraisal is likely to be interpreted as linked to the probability of restrictive autonomy and limited expressive behavior. Reactance prone individuals will equate perceived severity and perceived vulnerability with the threat to personal freedom. Thus, overall threat appraisal will become more salient.

This threat involves the possibility of contagion during air travel, with an enforced quarantine when arriving at the destination. Additional threats to freedom depend on the estimated severity of the contagion if contracted. The threat includes the possibility of prolonged hospitalization and permanent disability.

H4: Reactance proneness has a positive influence on threat appraisals—vulnerability and severity.

Risk aversion

Risk aversion involves a basic predisposition toward risk across a variety of situations (Mandrik and Bao, 2005). Risk aversion is the escape from harsh conditions, and especially under uncertainty. A high-risk averse person will have a keen vigilance for risky situations. Factors affecting risk would come under intense scrutiny. In principle, this would result in an exaggerated view of adverse probabilities.

PMT involves affective risk perceptions which extend the cognitive threat appraisal facet of the model. Affective risk perceptions in the context of travel destination involves worry and anxiety. These fear-laden emotions intensify cognitively derived risk levels.

In sum, the risk aversion trait is viewed as an individual’s degree of negative attitude toward engagement with potentially fear-arousing situations. Such attitudes spill over into states of fear. Risk aversion is associated with a self-efficacy image (Verma and Sharma, 2013). Self-efficacy is a determinant of anticipated fears and calamities (Bandura, 1983)

H5: Risk aversion has a positive influence on fear.

The hypotheses are shown pictorially in Figure 1.

Figure 1.

Conceptual model.

Gender as a moderator

It has been well established that for a variety of risk-taking tasks males generally assume greater risks than females (Gowen et al., 2019; Panno et al., 2018). No study has investigated gender differences in tourist destination risk perceptions. This is in spite of the fact that the gender gap in risk perceptions exists in many contexts (Byrnes et al., 1999). Patently, one might expect males to perceive risk differently from females. A gender effect as a moderating variable in the present study may be expected, but with some uncertainty.

H6: There is a significant categorical moderating effect of gender on relationships among model constructs.

Research model

The model that integrates study hypotheses is comprehensive in that it not only links risk appraisal to travel risk attitudes. In this model, risk appraisal levels measurably surface through personality traits. These traits include reactance proneness and risk aversion. The emotional response of fear also influences cognitive risk appraisal. Important model development criteria are met. These include: usability, practical implications, prediction, suitability for reliability and validity testing, and developing understanding (Vincent-Wayne, 1999).

Methodology

Data collection procedure and sampling

A survey was conducted among M-Turk panelists in two waves, spanning the month of July, 2020. Panelists consisted of qualified respondents were people were born in the United States, with at least one first-generation (immigrant) parent. MTurk participants were the sampling frame in exchange for a small payment. Cross checks in the questionnaire validated respondent authenticity.

Demographic sample comparisons between the two waves showed no significant differences with respect to age, gender, education, income, marital status and ethnicity (see Appendix 2). Means for latent variables in the model showed no significant differences between waves. These variables included destination risk, traveler vulnerability to contagious disease, fear of exposure to disease, and expectations of severity if illness was contracted, and reactance proneness to the consequences of the threat of illness (see Appendix 3).

MTurk sampling has been found to be reliable for consumer research studies; it is more representative of the general population parameters than traditional convenience samples found in published articles (Goodman et al., 2013). To further enhance the quality of our data in this study only MTurk workers with intellectual capabilities (i.e., a 90–100% Human Intelligence Task [HIT] approval rating and 100 or more approved HITs) were recruited to participate in this research.

Respondents were informed that researchers were interested in talking with people who had used their passports in the past 5 years. They understood that they would be evaluating tourist destination city advertising, and then asked to provide personal information. Afterward, each respondent was exposed to an advertising stimulus to one of four travel destination cities: Tokyo, London, Istanbul and Kuala Lumpur.

Advertising for destination cities varied by advertising puffery levels i.e. the degree to which the advertising contained exaggerated or boastful statements about the travel destination experience. Type A was low puffery and type B was high puffery (see Figure 2). We consulted communications literature for advertising puffery language (see Carrera et al., 2014; Hernandez et al., 2015; Lermer et al., 2015; Liberman and Trope, 1998; Massara et al., 2020; Tan, 2017; Trope and Liberman, 2010).

Figure 2.

Destination ad stimuli. (a) Low risk city (Tokyo)/type A, (b) low risk city (Tokyo)/type B, (c) low risk city (London)/type A, (d) low risk city (London)/type B, (e) high risk city (Istanbul)/type A, (f) high risk city (Istanbul)/type B, (g) high risk city (Kuala Lumpur)/type A, (h) high risk city (Kuala Lumpur)/type B.

The selection of the four destination cities was based on the 10 most popular contemporary tourist destinations. Use of popular destination cities enabled data collection from segments of respondents who were largely aware of the advertised city. Additionally, the four cities selected for the study varied in tourists’ subjective risk levels. A pilot study that employed a Q-sort methodology demonstrated the heterogeneity of these cities with respect to subjective risk levels. The pilot study was conducted using the same qualifiers as in the main study. The Q-sort task involved respondent classification of city travel destinations by travel risk expectations. This led to the selection of destination cities for the main study.

Measurement

After exposure to an initial travel destination ad, participants were given a fixed amount of time to assimilate ad content before they were allowed to move forward. They continued to review the same advertisement for a second exposure, to maximize message comprehension. Respondents were asked to evaluate the advertising message and to reveal their feelings and cognitions.

All constructs and items stemmed from the existing literature. Three latent variables were used to measure travel threat appraisal under the worldwide COVID-19 pandemic (Milne et al., 2002). These included illness severity (two items), vulnerability (two items), and fear (four items). Three items for reactance proneness were from Hong and Faedda (1996), whereas Mandrik and Bao (2005) provided four items for risk aversion. Finally, several studies produced five selected items for destination risk perception (Boshoff, 2002; Roehl and Fesenmaier, 1992; Um and Crompton, 1992). Refer to Appendix 1 for the items used in the survey.

Data analysis and findings

In order to test hypotheses, a total of 362 adults 18 and over were recruited via Amazon Mechanical Turk (MTurk) in exchange for a small payment. 64.9% of the participants were male and their ages ranged from 18 to 65, with a mean age of 36. About three-fifths had percent of participants had a college education and the median household income was $66,500.

Assessment of measurement model

We studied all variables hypothetically related in the study model. A principal components analysis disclosed that model variables were not explained by a single factor. Thus, common method variance is of little concern. PLS-SEM is a multivariate analysis approach used to estimate path models with latent variables. The study used a structural equation model (SEM) with the Consistent PLS (PLSc-SEM) approach and applied the SMART PLS 3.2.6 data analysis tool for model estimation and multivariate analysis. Traditional PLS algorithms may overestimate loadings and underestimate correlations between latent variables. R² may be underestimated for endogenous latent variables (Dijkstra, 2010; Dijkspra and Schermelleh-Engel, 2014). PLSc is a corrective to these potential shortcomings. In addition, the choice of PLSc for model testing in the present study is based on using constructs to represent factors. Table 1 shows that all the associations between the factors were below .600. Thus, each factor was statistically distinct from the other.

Table 1.

Correlation matrix among each construct.

Variable	Reactance Proneness	Risk Aversion	Vulnerability	Fear	Severity
Reactance Proneness
Risk Aversion	0.550**
Vulnerability	0.399**	0.417**
Fear	0.280**	0.416**	0.467**
Severity	0.522**	0.429**	0.574**	0.495**
Destination Risk Perception	0.439**	0.472**	0.494**	0.275**	0.475**

** p < .01.

All study constructs achieved satisfactory convergent validity and internal consistency reliability. A variety of criteria were satisfied including tests by Cronbach, rho_alpha, composite reliability and average variance extracted (see Table 2).

Table 2.

Construct reliability.

Variable	Cronbach’s Alpha	rho_A	Composite Reliability	Average Variance Extracted (AVE)
Fear	0.835	0.843	0.836	0.562
Reactance Proneness	0.784	0.785	0.785	0.646
Risk Aversion	0.796	0.799	0.795	0.494
Destination Risk Perception	0.685	0.685	0.685	0.521
Severity	0.810	0.810	0.810	0.681
Vulnerability	0.794	0.802	0.797	0.663

Study constructs were tested for discriminant validity. This established that each construct is unique and captures phenomena other than those captured by other model constructs. Constructs were measured by Heterotrait-monotrait ratios (HTMT). HTMT values above 0.90 indicate a lack of discriminant validity. Table 3 shows that all study constructs are independent of each other.

Table 3.

Discriminant validity.

Variable	Fear	Reactance Proneness	Risk Aversion	Destination Risk Perception	Severity
Fear
Reactance Proneness	0.317
Risk Aversion	0.498	0.602
Destination Risk Perception	0.390	0.577	0.569
Severity	0.604	0.64	0.539	0.612
Vulnerability	0.574	0.526	0.517	0.692	0.718

Structural model results

The overall fit of the structural model was good. Table 4 shows results for the standardized mean square residual SRMR. This measure describes the discrepancy between observed correlations and model implied correlations. Results obtained from the analysis show minimal discrepancies, which ranged from 0.066 to 0.080 for saturated and estimated models. Other fit statistics confirmed the adequacy of the model fit.

Table 4.

Model fit.

	Saturated Model	Estimated Model
SRMR	0.039	0.058
d_ULS	0.211	0.46
d_G	0.152	0.183
Chi-Square	294.273	344.382
NFI	0.89	0.871

Table 5 reveals that path coefficients from the model are all significant with probabilities less than P < .01. The magnitudes of the paths are relatively strong, as well. Path coefficients are displayed in the configuration diagram of structural relationships (see Figure 3).

Table 5.

Path coefficients.

	Original Sample (O)	Sample Mean (M)	Standard Deviation (STDEV)	T Statistics (O/STDEV)	P Values
Vulnerability → Destination Risk Perception (H1)	0.371	0.373	0.075	4.977	0.000
Severity → Destination Risk Perception (H2)	0.241	0.243	0.07	3.426	0.001
Fear → Vulnerability (H3a)	0.387	0.389	0.053	7.268	0.000
Fear → Severity (H3b)	0.396	0.394	0.053	7.526	0.000
Reactance Proneness → Vulnerability (H4a)	0.316	0.319	0.047	6.788	0.000
Reactance Proneness → Severity (H4b)	0.405	0.407	0.049	8.206	0.000
Risk Aversion → Fear (H5)	0.410	0.411	0.053	7.743	0.000

Figure 3.

A model of key factors influencing travelers’ destination risk perceptions.

All hypotheses were confirmed with path coefficients as shown:

H1: Expectations of vulnerability to COVID-19 illness has a positive influence on destination risk perceptions (path coefficient = .516).

H2: Expectations of dire symptoms and outcomes if COVID-19 was developed has a positive influence on destination risk perceptions (path coefficient = .241).

H3: Fear associated with the COVID-19 pandemic has a positive influence on threat appraisals—vulnerability and severity (path coefficients = .448, .449).

H4: Reactance proneness has a positive influence on threat appraisals—vulnerability and severity (path coefficients = .383, .494).

H5: Risk aversion has a positive influence on fear (path coefficient = .496).

Expectations of vulnerability to COVID-19 has a strong path relationships on risk perception. The path from expected severity of COVID-19 illness, while significant, is moderate in its strength.

Measurement and structural invariance: Males and females

Thus far, we have explained the process of travel risk perception among international destination travelers. However, there is the potential problem of diverse data structures arising from possible differences between male and female segments. An analysis may provide assurance that the general model previously described is robust for these key market segments. Path coefficients for males and females are shown in Table 6.

Table 6.

Paths by gender

Path	Males	Females	Difference	P Values
Vulnerability → Destination Risk Perception (H1)	0.393	0.322	0.071	0.652
Severity → Destination Risk Perception (H2)	0.259	0.209	0.05	0.752
Fear → Vulnerability (H3a)	0.303	0.540	−0.237	0.035
Fear → Severity (H3b)	0.372	0.445	−0.073	0.496
Reactance Proneness → Vulnerability (H4a)	0.406	0.167	0.240	0.015
Reactance Proneness → Severity (H4b)	0.476	0.295	0.180	0.093
Risk Aversion → Fear (H5)	0.358	0.511	−0.153	0.177

Paths for males and females were estimated and analyzed. All paths for the separate models of males and females were significant. Thus, H6: There is a significant categorical moderating effect of gender on relationships among model constructs is rejected. However, between models two path coefficients differed significantly by gender. These involved the paths from fear to vulnerability (males .303 vs. females .540), and reactance proneness to vulnerability (males .406 vs females .167).

Discussion and practical implications

The major contribution of the current study consists of its unravelling a set of unexplored factors affecting travel risk perceptions for a variety of destination cities. In so doing, the study contributes a general model of destination risk. The model focuses on destination cities, rather than countries. This is because cities are at the heart of tourists’ plans. The work is contemporaneous and reflects a radically different tourist destination marketing environment, given risks associated with a worldwide pandemic.

The study contributes to the body of knowledge by combining threat appraisal and fear of COVID-19 to predict tourists’ risk perceptions of travel destinations. Additionally, antecedent personality factors—reactance proneness and risk aversion—were revealed as important factors driving threat appraisal. The general model of relationships that drive travel risk perceptions is applicable to male and female tourist segments. However, for males relative females reactance proneness is a stronger antecedent to vulnerability to the pandemic. For females relative to males, fear is a stronger antecedent to vulnerability to the pandemic.

The study extends the theoretical scope and application of risk appraisal. It includes the joint impact of both subjective vulnerability and severity of exposure to COVID-19 on travel destination perceived risk. This effect is shown to be magnified by the emotion of fear. Another contribution is the introduction of tourist personality traits that drive the entire process of travel destination risk generation.

Underlying travel threat appraisal are motivational dispositions that generate variations in action readiness. Motivational dispositions may culminate in a search for threat cues, driven by fear dispositions or self-protection tendencies (Frijda and Zeelenberg, 2001). This is how emotions influence threat appraisal processes. These processes filter available information which may be salient or inconspicuous, depending on system sensitivities.

From a theoretical perspective, the study generates insights into the relative impact of vulnerability and severity expectations in the context of tourism destination risk appraisal. PMT has been significantly advanced by introducing innovative personality variables to extend the theory. The equivalent bi-directional impact of fear on both of the other elements of threat appraisal is another theoretical contribution. Fear aversion as a primary factor in fear inducement provides a new direction for PMT theory development on the inner dynamics of threat appraisal. Progress in theoretical understanding of PMT is enabled by the introduction of reactance proneness as an antecedent of consequence to threat appraisal.

This study was conducted at a time of precipitate disengagement from tourism destination travel. Consequently, tourist variability in threat appraisal and their destination risk traits may be unlike parameters that have ever been seen in the past. This represents a unique and critical research opportunity that has been addressed. The value of risk perception analysis has been underscored by Vincent-Wayne (1999) and by Law (2006). It may strengthen or undermine plans for international travel. In general, risk perception analysis plays a role in brand-image development, targeting, positioning and segmentation. For example, consumers may be segmented based on their risk-reducing strategies usage.

The implications of the study for tourism marketing under pandemic and post-pandemic conditions are striking. Marketing practice should take into account both health threats and fear levels of tourists to understand their travel dispositions. Travel destination marketers can profit from this study by bypassing conventional gender-based segmentation strategies in addressing tourists’ destination risk levels. On the positive side, marketers can use study insights to employ psychographic strategies, targeting travelers who are not overly reactive prone by nature, and are more risk tolerant. Creative communications strategies can be developed to strengthen destination travel motivations among prospects who are less risk averse. For such consumers, destination travel rewards should receive high salience.

Sensation seeking has been defined as the need for varied, novel, and complex sensations and experiences, with a willingness to take risks for these experiences. Sensation seeking tourists are prime tourism prospects under pandemic conditions, since they are less risk averse and more adventurous. These individuals have a need for stimulation and a strong desire for interaction with the tourist environment. Primary motives are thrill and exploration. Sensation seeking is more prevalent among independent tourists, in distinction from mass tourists (Li et al., 2015). In PMT, risk appraisal is based on the calculus of rewards minus risks. Risks become increasingly acceptable in the light of desirable destination travel benefits, which may be emphasized in travel communications.

Limitations and future research

In this study the primary focus was on one type of perceived destination travel risk i.e. perceived health/life risk. Research findings showed that this type of risk, by itself, may be strongly associated with overall destination perceived risk. Risk perceptions are mediated by perceptions of benefits. The latter alter judgments about risk (Slevitch and Sharma, 2008). Future research can systematically test varieties of persuasive information designed to reduce risk perceptions, in the context of perceived benefits. What could be learned will have immense practical value. An increased sense of risk reduction enhances travelers’ willingness to pay a price premium for travel offerings. The study focuses on risk perceptions of travel destination cities. It is devoid of country contexts and the role of related travel advisories. This should be featured in future research applications of PMT and destination marketing.

Future research on perceived destination risk might also involve destination image. Destination image was studied as a moderator of the impact of destination risk on travel intentions or behavior (Hsiao-Yun, 2018). Alternatively, destination risk has been studied as a mediator of the impact of destination image on attitudinal or behavioral travel outcomes (Makhdoomi and Baba, 2019). Several studies have related risk information processing to travel risk appraisal (Aro et al., 2009; Jonas et al., 2011; Zhou-Talbert et al., 2020). However, destination image measurement may be problematic for some travelers, especially if they have given little thought to a sojourn at the destination.

The present research failed to explore and possibly incorporate other significant personal traits associated with threat appraisal. Anxiety is a central trait which might have been incorporated to advantage. This is because anxiety affects the usage of cognitive resources under threatening stimuli. Additionally, anxiety serves to exaggerate threat probabilities and increase subjective costs (Mitte, 2007). There is a need for research that determines effects of anxiety on travel risk perceptions.

Control variables that might have been employed in the analysis—but were not—include destination framing effects. These might have been introduced to further explain variance in travel risk perceptions. An example would be message framing of travel destinations by stimulation or tranquility would differentially affect sensation seeking tourists.

Further, the study omitted measurement of tourist behaviors closely associated with destination intention: relations with travel agents, response to tourism promotions, and group versus individual travel arrangements. Neither did it capture intended durations of a tourism destination visit.

Protection motivation theory can be extended to explain much more about travel destination decisions. Coping appraisal can be explored alongside of threat appraisal. Both represent major dimensions of protection motivation. Coping appraisal is based on self-perceptions of efficacy and projected benefits and costs of travel decisions.

Future research on health-related threats and destination marketing should include the impact of travel advisories. Travel advisory information which might be useful in the threat appraisal process can probe advisory acceptance resulting from disseminating information to travelers. Furthermore, even when assimilated, advisory information about travel destination risk may be extremely unstable (Tsang et al., 2018).

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) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Melvin Prince

Appendix 1

Appendix 2

Appendix 3

References

Aro

Vartti

Schreck

, et al. (2009) Willingness to take travel-related health risk—a study among Finnish tourists in Asia during the avian influenza outbreak. International Journal of Behavioral Medicine 16(1): 68–73.

Bandura

(1983) Self-efficacy determinants of anticipated fears and calamities. Journal of Personality and Social Psychology 45(2): 464–469.

Boshoff

(2002) Service advertising: an exploratory study of risk perceptions. Journal of Service Research 4(4): 290–298.

Byrnes

Miller

Schafer

(1999) Gender differences in risk taking: a meta-analysis. Psychological Bulletin 125(3): 367–383.

Caber

Gonzalez-Rodriguez

Albayrak

, et al. (2020) Does perceived risk really matter in travel behavior? Journal of Vacation Marketing 26(3): 334–353.

Carrera

Caballero

Munoz

, et al. (2014) Construal level as a moderator of the role of affective and cognitive attitudes in the prediction of health risk intentions. British Journal of Social Psychology 53(4): 773–791.

Cui

Liu

Chang

, et al. (2016) An overview of tourism risk perception. Natural Hazards 82(1): 643–658.

Dijkstra

(2010) Latent variables and indices: Herman-Wold’s basic design and partial least squares. In: Vesposito

Chin

Henseler

, et al. (eds) Handbook of Partial Least Squares; Concepts Methods and Applications. Berlin, Germany: Springer-Berlin, pp. 23–46.

Dijkstra

Schermelleh-Engel

(2014) Consistent partial least squares for nonlinear structural equation models. Psychometrika 79(4): 585–604.

10.

Ferrer

Klein

WMP

Avishal

, et al. (2018) When does risk perception predict protection motivation for health risk? A person-by-situation analysis. Plos One 13(3): 1–15.

11.

Frijda

Zeelenberg

(2001) Appraisal: What is the dependent? In: Scherer

Schorr

Johnstone

(eds) Appraisal Processes in Emotion: Theory, Methods, Research. New York: Oxford University Press, pp. 20–36.

12.

Goodman

Cryder

Cheema

(2013) Data collection in a flat world: the strength and weaknesses of Mechanical Turk samples. Journal of Behavioral Decision Making 26(3): 213–224.

13.

Gowen

Fillpowicz

Ingram

(2019) Chronotype mediates gender differences in risk propensity and risk-taking. Plos One 14(5): 1–13.

14.

Hernandez

Wright

Rodrigues

(2015) Attributes versus benefits: the role of construal levels and appeal types on the persuasiveness of marketing messages. Journal of Advertising 44(3): 243–253.

15.

Hong

Faedda

(1996) Refinement of the Hong psychological reactant scale. Educational and Psychological Measurement 56(1): 173–182.

16.

Hsiao-Yun

(2018) The impacts of perceived risk in senior travel: exploring the moderating roles of destination image and involvement. Global Review of Research in Tourism Hospitality and Leisure Management 4(1): 1–10.

17.

Jonas

Mansfeld

Paz

, et al. (2011) Determinants of health risk perception among low-risk taking tourists traveling to developing countries. Journal of Travel Research 50(1): 87–99.

18.

Law

(2006) The perceived impact of risks on travel decisions. International Journal of Tourism Research 8(4): 289–300.

19.

Lermer

Streicher

Sachs

, et al. (2015) The effect of construal level on risk taking. European Journal of Social Psychology 45(1): 99–109.

20.

C-Y

Tsai

, et al. (2015) The impact of extraversion and sensation-seeking on tourist role. Society for Personality Research 43(1): 75–84.

21.

Liberman

Trope

(1998) The role of feasibility and desirability considerations in near and distant future decisions: a test of temporal construal theory. Journal of Personality and Social Psychology 75(1): 5–18.

22.

Lowenstein

Weber

Hsee

, et al. (2001) Risk-as-feelings. Psychological Bulletin 127(2): 267–286.

23.

Makhdoomi

Baba

(2019) Destination image and travel intention of travelers to Jammu and Kashmir: the mediating effect of risk perception. Journal of Hospitality, Application and Research 14(1): 35–56.

24.

Mandrik

Bao

(2005) Exploring the concept of general risk aversion. Advances in Consumer Research 32: 531–539.

25.

Massara

Scarpi

Porcheddu

(2020) Can your advertisement go abstract without affecting willingness to pay? Product-centered vs. lifestyle content in luxury brand print advertisements. Journal of Advertising Research 60(1): 28–38.

26.

Mills

Reyna

Estrada

(2008) Explaining contradictory relations between risk perception and risk taking. Psychological Science 19(5): 429–433.

27.

Milne

Orbell

Sheeran

(2002) Combining motivational and volitional interventions to promote exercise participation: protection motivation theory and implementation intentions. British Journal of Health Psychology 7(2): 163–184.

28.

Mitte

(2007) Anxiety and risk decision-making: the role of subjective probability and subjective costs of negative events. Personality and Individual Differences 43(2): 243–253.

29.

Norman

Boer

Seydel

(2005) Protection motivation theory. In: Conner

Norman

(eds) Predicting Health Behaviour: Research and Practice with Social Cognition Models. Maidenhead Bekshire, UK: Open University.

30.

Panno

Donati

Milioni

, et al. (2018) Why women take fewer risks than men do: the mediating role of state anxiety. Sex Roles 78(3-4): 286–294.

31.

Gibson

Zhang

(2009) Perceptions of risk and travel intentions: the case of China and the Beijing Olympic games. Journal of Sports and Tourism 14(1): 43–67.

32.

Quick

Stephenson

(2008) Examining the role of trait reactance and sensation seeking on perceived threat, state reactance, and reactance restoration. Human Communication Research 34(3): 448–478.

33.

Quick

Scott

Ledbetter

(2011) A close examination of trait reactance and issue involvement as moderators of psychological reactance theory. Journal of Health Communications 16(6): 660–679.

34.

Roehl

Fesenmaier

(1992) Risk perceptions and pleasure travel: an explanatory analysis. Journal of Travel Research 30(4): 17–26.

35.

Rogers

(1975) A protection motivation theory of fear appeals and attitude change. Journal of Psychology 91(1): 93–114.

36.

Rogers

(1983) Cognitive and psychological processes in fear appeals and attitude change: a revised theory of porotection motivation. In: Cacioppo

Petty

(eds) Social Psychophysiology: A Source Book. London UK: Guilford, pp. 153–176.

37.

Schroeder

Pennington-Gray

Kaplanidou

, et al. (2013) Destination risk perceptions among US residents for London as the host city of the 2012 Summer Olympics games. Tourism Management 38(c): 107–119.

38.

Sheeran

Harris

Epton

(2014) Does heightening risk appraisal change people’s intentions and behavior? A meta-analysis of experimental studies. Psychological Bulletin 140(2): 311–343.

39.

Slevitch

Sharma

(2008) Management of perceived risk in the context of destination choice. International Journal of Hospitality and Tourism Information 9(1): 85–103.

40.

Sonmez

Graefe

(1998) Influence of terrorism risk on foreign tourism decision. Annals of Tourism Research 25(1): 112–144.

41.

Sunstein

(2003) Terrorism and probability neglect. Journal of Risk and Uncertainty 26(2-3): 121–136.

42.

Tan

(2017) The effects of temporal psychological distance on word of mouth for information about destination image attributes. Behaviour and Information Technology 36(11): 1101–1110.

43.

Tanner

Hunt

Eppright

(1991) The protection motivation model: a normative model of fear appeals. Journal of Marketing 55(3): 36–45.

44.

Trope

Liberman

(2010) Construal-level theory of psychological distance. Psychological Review 117(2): 440–463.

45.

Tsang

NKF

Wong

Prideaux

(2018) An evaluation of the effectiveness of travel advisories with a specific focus on Hong Kong’s travel alert system. Journal of Vacation Marketing 24(4): 307–323.

46.

Crompton

(1992) The roles of perceived inhibitors and facilitators in pleasure travel destination decisions. Journal of Travel Research 30(3): 18–25.

47.

Verma

Sharma

(2013) Construct validation of self-efficacy scale. Journal of Soft Skills 7(4): 15–26.

48.

Vincent-Wayne

(1999) Consumer perceived risk: conceptualization and models. European Journal of Marketing 33(1/2): 163–195.

49.

Wang

Liu-Lastres

Ritchie

, et al. (2019) Travellers’ self-protections against health risks: an application of the full protection motivation theory. Annals of Tourism Research 78: 102743.

50.

Weber

Blais

Betz

(2002) A domain-specific risk-attitude scale: measuring risk perception and risk behaviors. Journal of Behavioral Decision-Making 15(4): 263–290.

51.

Zhou-Talbert

Lee

Vielot

, et al. (2020) Awareness, knowledge, and travel-related risk factors for Zika virus among Latinos attending a federally qualified health center in rural North Carolina. North Carolina Medical Journal 81(1): 1422.