Destination image during the COVID-19 pandemic and future travel behavior: The moderating role of past experience

This study investigates the effects of cognitive destination image shaped by media during the COVID-19 pandemic on willingness to support and post-pandemic travel intention. Drawing upon the concept of cognitive destination image and through an online self-administered survey, the effects of four factors including trust, crisis management, healthcare system, and solidarity on travel behavioral intention are compared based on tourists’ prior experience of a given destination. To achieve this aim, ten countries with different coping strategies, numbers of positive cases and mortality rate were studied. A total number of 518 useable questionnaires were collected from the prospect international tourists who followed news related to COVID-19 for one of the selected countries and plan to travel in the future. Partial least squares – structural equation modeling and multi-group analysis were used to test the model and hypotheses. The results showed the high predictive power of the model on post COVID-19 travel behavioral intention. The findings revealed the strong and positive effects of trust and healthcare system on behavioral intention of respondents without past experience to visit a destination, whereas the effect of solidarity on behavioral intention was identified much stronger for the prospect tourists with past experience of visiting a destination. This research provides unique theoretical contributions by investigating the effects of trust, crisis management, healthcare system, and solidarity shaped by media during COVID-19 outbreak as the components of cognitive destination image on future behavioral intention across past experience of visiting a destination. This study also provides insights on post-crisis recovery factors affecting travel behavioral intention and demand.


Introduction
The novel coronavirus COVID-19 has profoundly affected the world economy. The tourism industry has been among the most severely affected sectors due to the unprecedented restrictions on mobility (Gössling et al., 2021;Rastegar, Seyfi, & Rasoolimanesh, 2021;Rastegar, Higgins-Desbiolles, & Ruhanen, 2021;Seyfi et al., 2020). Based on UNWTO estimates, the COVID-19 pandemic has caused an approximate 1.1 billion decline in international tourist arrivals with 100-120 million jobs at risk as a result (Hall & Seyfi, 2021;UNWTO, 2020). Countries' different responses and coping strategies during the COVID-19 pandemic have received extensive media coverage which potentially affect their destination image (Gössling et al., 2021). Prior research suggests that tourist behaviour and destination image are significantly influenced by tourist perceptions of safety and risk (e.g. Casali et al., 2020). Additionally, the influence on country and destination image by news media framing can play a significant role in understanding travel intentions. For instance, Chemli et al. (2020) investigated the influence of intra-pandemic media coverage and found that, as the primary source of information during a crisis, the media greatly affect potential travellers' destination awareness. Yang et al. (2021) also found that media coverage surrounding COVID-19 has negatively affected China's destination image and potential visitors' travel intentions in some markets. Other studies have investigated the effects of COVID-19 on tourist's perceived risk and its impact on their future travel behaviour. For instance, the studies of Xie et al. (2021) and Jin et al. (2021) on Chinese travellers; Pappas (2021) and Pappas and Glyptou (2021) on adult residents of Athens; and Perić et al. (2021) on Serbian travellers, all highlight the significant influence of the COVID-19 on potential tourist's perceived risk and post-pandemic travel intention and decision-making.
Many studies on COVID-19 have attempted to identify the major predictors of post-pandemic travel intentions, including people's intrapandemic risk perceptions (Li et al., 2020;Neuburger & Egger, 2020;Sánchez-Cañizares et al., 2020); destination health-risk images (Bhati et al., 2020;Rastegar, Seyfi, & Rasoolimanesh, 2021;Yang et al., 2021); travel fears and anxiety (Zenker et al., 2021;Zheng et al., 2021); and perceived knowledge (Han et al., 2020). Although such studies provide an assessment of the impact of intra-pandemic destination image and subsequent post-pandemic travel intention, significant research gaps nevertheless remain. To the best of the authors' knowledge, no prior study has investigated the differences between the perceptions of tourists with past experience and without past experience of a given destination and post pandemic travel intention. The accumulated knowledge of prior experience strongly influences the tourism decision-making process (Casali et al., 2020) and the likelihood of visitation (Milman & Pizam, 1995). In other words, tourists who have previously visited a destination often have a more comprehensive knowledge foundation for their attitudes (Casali et al., 2020;Milman & Pizam, 1995).
Research also suggests that destination familiarity gained through prior visits to the destination influences long-term relationships and stimulates the visitor to return (Tsai, 2012). In their studies, De Nisco et al. (2015) suggested that past experience generally improves destination image for visitors, which also reflects similar observations by Campo and Alvarez (2014). Hence, experience can explain differences in behaviour and shape the image that a tourist holds of a destination (Chen & Lin, 2012) and potentially moderates the relative influence of various kinds of antecedents on behavioural intentions as expressed in the intention to travel and the willingness to support a destination (Tan & Wu, 2016). Familiarity and prior experience with a destination has also been identified as a factor in early return to a destination following disasters, such as earthquakes . Hence, it can be concluded that prior experience could be an important determinant of post-pandemic travel behavioral intention. Additionally, the majority of studies on COVID-19 and tourism tend to be case-or country specific, while a cross-country analysis is absent from the extant literature. Thus, the interplay between the triggers of post pandemic travel and subsequent behavioural intent needs to be researched further from differing theoretical perspectives through a cross-country analysis. The latter helps to reveal connections between media framed destination image through the national responses dealing with pandemic and the post-pandemic travel intention to the investigated countries.
To address the above-mentioned gaps, through a cross-country analysis of ten countries with different response strategies, case numbers, and mortality rates, this study develops and empirically tests an integrated model. The proposed model links trust, crisis management, health care system and solidarity, which have been shaped by media to investigate the effects of cognitive perceived image of a destination on behavioral intentions and willingness to visit a destination through the moderating role of tourists' prior experience of selected destinations. The outcomes of this research help to understand the shaping factors of potential tourists' desire to visit a destination and its associated behavioural intent which are of vital significance in developing effective tourism marketing and management strategies in post crisis recovery processes.

Destination image and past experience
The concept of a tourist destination image has been highly influential in destination development and marketing since the 1970s (Crompton, 1979;Gunn, 1972;Oliveira & Huertas, 2019). Destination image plays a major role in understanding tourist behavioral intentions and decision-making (Afshardoost & Eshaghi, 2020;Karl et al., 2020;Tan & Wu, 2016). While some scholars measure destination image using the two components of cognitive and affective (Tan & Wu, 2016), recent literature is more inclined towards conceptualizing destination image as a multi-faceted construct based on cognitive, affective, and conative components (Afshardoost & Eshaghi, 2020). The cognitive component relates to an individual's knowledge and belief (Pike & Ryan, 2004) or perception and attitude (Tan & Wu, 2016) towards a destination. The affective component refers to feelings and emotions an individual might have towards a destination (Tan & Wu, 2016), while the conative components can outline the desired future situation or active consideration of a place as a potential travel destination (Pike & Ryan, 2004). Both the cognitive perception and the affective appraisal made by the tourist following the previous visit can shape the future selection of a place (Chen & Lin, 2012). This underlines the key role of cognitive perception and affective appraisal in a place's image which also reflects the observations of Chen and Lin (2012) who argue that past experience is a key determinant of destination image.
During crises, such as COVID-19, when tourists cannot freely travel and may be more exposed to media coverage than usual, destination image is projected by both mass and social media. Tourist perceptions can therefore be shaped by negative and misleading information or media broadcasting others' reactions and fear towards the crises (Zheng et al., 2021). Presenting such images of a destination can influence people who have never visited a destination (Chew & Jahari, 2014;Zenker et al., 2019). In such circumstances, the cognitive image of a destination held by tourists provides an opportunity to investigate how destination familiarity or past experience can affect future travel intentions. This is particularly important as cognitive destination image is related to individual familiarity or knowledge of a destination derived from past experience (Pike & Ryan, 2004).
Past experience in this study refers to a tourist's destination familiarity gained through prior visit to a destination. The moderating effect of past experience have been examined in different contexts which reflect the key role of personal knowledge for previous visitors/repeaters (De Nisco et al., 2015;Tsai, 2012). The latter also affect perceptions of a place, their formation search process, and ultimately affecting their revisit intention and WOM publicity (Campo & Alvarez, 2014;De Nisco et al., 2015). Consequently, this study proposes a model in which cognitive destination image formed by past experience moderates the effect of trust, healthcare system, solidarity and crisis management on behavioral intention.

Intention to travel and willingness to support a destination
Experience is one of the most powerful predictors of the behavioural intent of a tourist (Coudounaris & Sthapit, 2017). Ajzen and Fishbein (2000) refers to behavioral intentions as the perceptions of individuals about what they expect to do in a given situation. Travel intention is "the subjective probability of whether a customer will or will not take certain actions that are related to a tourist service" (Moutinho, 1987, p. 11). This reflects an individual's intent to travel or commitment to travel (Jang et al., 2009).
Tourist's impressions of the travel experience and perceptions of the destination are the most reliable source of information when it comes to revisitation and WOM communications (Coudounaris & Sthapit, 2017). Tourists' behaviors, including their selection of destinations to visit, subsequent evaluation of destination decisions and future behavioral intentions, are related to willingness to revisit or support a destination (Coudounaris & Sthapit, 2017). A positively held image of a destination can trigger desire to visit and influence tourists' willingness to support a destination (Chen & Tsai, 2007) by revisiting and recommending the destination to others (Chen & Tsai, 2007;Prayag et al., 2017). Such studies suggest that a positively held destination image plays an important role in the public perception of a particular country. Hall (2010) and Avraham (2015) highlight that major crises, such as warfare and epidemics, can affect destination image triggering negative desires to visit such destinations.
The media's portrayal of different national responses to COVID-19, morality rate and infections is believed to affect the image of a country. The findings of a Pew Research Center poll showing public perceptions of the USA in 13 countries highlight that the American handling of COVID-19 has influenced international public opinion toward the U.S. (Wike et al., 2020). Thus, the current study sought to investigate how national responses dealing with the pandemic can potentially affect destination visits post-COVID-19. Morgan and Hunt (1994, p. 23) define trust as "one party's confidence in an exchange partner's reliability and integrity". Trust has been viewed as an effective tool to minimize uncertainty and the perception of risk (Han & Hyun, 2015). Trust has also been viewed as one of the essential factors affecting travelers' intention. The connection between individuals' trust and their behavioural intent has been assessed in various contexts (e.g. Chen et al., 2011) including tourism and hospitality (Abubakar et al., 2017;Kim et al., 2011). In their study, Han and Hyun (2015) found that travelers tend to visit destinations they feel trustworthy and reliable. Abubakar et al. (2017) reached similar conclusions and their findings highlighted the important effect of destination confidence on revisit intent. The perception of trust has been viewed as a critical element in the likelihood of visitation (Abubakar et al., 2017). Past experience is also considered as an antecedent to making decision about future travel (Tsai, 2012). Past experience also affects feelings of trustworthiness in relationships (Abubakar et al., 2017). During COVID-19, public trust in government pandemic management measures has been crucial to their acceptance and implementation (Fancourt et al., 2020). Based on the preceding discussion, the following hypotheses are proposed. H1 a : There is significant difference for the effect of trust on intention to travel to a destination after COVID-19 crisis between tourists with past experience and without past experience. H1 b : There is significant difference for the effect of trust on willingness to support a destination between tourists with past experience and without past experience.

Crisis management, intention to travel and willingness to support
The negative impacts of crises on tourism are well recognized (Chew & Jahari, 2014;Huan et al., 2004;Li et al., 2018). Disasters and crises can negatively affect destinations by significantly reducing arrivals and expenditure (Hall et al., 2004). For example, studies show a significant decrease in the number of tourists in Africa after an Ebola outbreak (Cahyanto et al., 2016) or in China after SARS (Wen et al., 2005). Similarly, public fear during the COVID-19 pandemic and travel being a high-risk and often difficult activity have led to significant reduction in travel demand (Zheng et al., 2021). While governments worldwide took preventive actions to combat the virus, effective crisis management can play a significant role in success of recovery efforts. In post crisis recovery, government policies, effective positive communication, and new tourism product may be effective in restoring destination image (Avraham, 2015).
Tourists' perceived risk and travel fear can significantly affect individuals' destination choices and travel behaviors (Kozak et al., 2007). Such perceptions can be shaped by negative information about the pandemic in the social media (Zheng et al., 2021) or media broadcasting others' reactions, fear and panic towards pandemics (Zheng et al., 2021). Misleading information and imagery can particularly influence those who have not visited a destination before (Chew & Jahari, 2014;Zenker et al., 2019). This is especially important as tourists with low familiarity of a destination tend to rely more on external information sources (Kozak et al., 2007). In addition, tourists consider their travel to be risky in unfamiliar environments, while they feel more secure in familiar destinations (Lepp & Gibson, 2003). In examining destination values for tourists visiting Vietnam, Chi et al. (2020) found that destination familiarity positively moderates the effects of destination awareness and travel intentions.
Destination image recovery is a tourism-related priority of many nations post crisis. Destination image can be improved when stakeholders at a destination work towards promoting a positive image which in turn affect tourists' perception and travel intention (Avraham, 2015;Pappas & Papatheodorou, 2017). However, as destinations are intangible products, past experience, tourists' knowledge, and information can potentially lead to greater certainty (Chi et al., 2020). Similarly, destination familiarity is believed to positively affect destination image (Baloglu, 2001;Hernández Maestro et al., 2007). Past experience has also been shown to directly affect tourists risk perceptions of a destination and their travel intentions at uncertain times such as the COVID-19 pandemic (Hassan & Soliman, 2021;Karl et al., 2020). Given the preceding discussion, the following hypotheses are proposed: H2 a : There is significant difference for the effect of crisis management on intention to travel to a destination after COVID-19 crisis between tourists with past experience and without past experience.
H2 b : There is significant difference for the effect of crisis management on willingness to support a destination between tourists with past experience and without past experience.

Healthcare system, intention to travel and willingness to support
COVID-19 has demonstrated the importance of a strong healthcare system (WHO, 2020). The COVID-19 outbreak therefore raises significant questions concerning the efficacy of national healthcare systems particularly in low-and middle-income countries with limited resources (Arora et al., 2020). While countries such as as China, Iran and Italy experienced largescale community transmission, national COVID-19 response and effective healthcare systems in countries such as Australia and New Zealand were important factors in reducing the pandemic's impacts and portraying a safe destination image (Wike et al., 2020).
Different studies have highlighted the importance of health destination image for the competitiveness of tourism destinations (Becken et al., 2017). The likelihood of presence of any risk, injury, or sickness while visiting a destination can negatively influence tourists perceived risk and destination image (Chew & Jahari, 2014). This is particularly important in the case of healthcare systems, as trust in reliability and quality assurance of a destination with positive health image can directly influence the future travel intentions (Abubakar & Ilkan, 2017). Healthcare system and the quality of health services at a destination are considered as public infrastructure influencing destination image. For example, Moreno-González et al. (2020) in examining the formation of health-related destination image found that perceived health risk factors affect both destination image and overall health expectation in the destinations. The quality of healthcare system in a destination is therefore a high priority for many tourists (Crouch, 2010) which strongly influence destination attractiveness and travel intentions (Abubakar & Ilkan, 2017).
The role of information broadcasted by mass media, social media WoM or EWoM is fundamental in influencing public opinion (Zarezadeh et al., 2019) and perceived safety of destinations (Marine-Roig & Huertas, 2020). It is particularly important for non-visitors as they usually have limited destination familiarity with any knowledge instead gained from media, other individuals or the Internet (Tan & Wu, 2016). In contrast, past experience or familiarity with a destination directly influences the image an individual has of a destination and can affect his/her choice of destination (Chen & Lin, 2012). Prior knowledge and experience are therefore considered to play an important role in shaping a more favorable destination image and judgment and feeling regarding safety at a destination (Karl et al., 2020). Given the extant literature, the following hypotheses are proposed: H3 a : There is significant difference for the effect of healthcare system on intention to travel to a destination after COVID-19 crisis between tourists with past experience and without past experience.
H3 b : There is significant difference for the effect of healthcare system on willingness to support a destination between tourists with past experience and without past experience.

Solidarity, intention to travel and willingness to support
During the COVID-19 outbreak, the UNWTO urged world nations to act in solidarity with each other by stressing the importance of international dialogue and cooperation (UNWTO, 2020;West-Oram, 2021). Such solidarity was shown to play a significant role in combating the COVID-19 pandemic (UN, 2020;West-Oram, 2021). Solidarity can be constructed via the development of common social identities among those sharing the same experience during crises (Arab-Zozani & Hassanipour, 2020; Tomasini, 2021). During the COVID-19 pandemic, solidarity plays a critical role in encouraging people to support health and social interventions by assisting those in need and sharing resources (Arab-Zozani & Hassanipour, 2020). There were also cases when a lack of solidarity was seen when people protested against restrictions or abused healthcare workers (Arora et al., 2020).
Understanding tourist's feeling of solidarity with residents and destinations is regarded as essential for effective tourism planning and destination marketing (Stylidis et al., 2020). The relationship between tourists and a destination and its residents can be explained through emotional solidarity (Woosnam, 2011;Woosnam et al., 2015). Understanding such relationship offers substantial "insights into tourists 'destination image and future behavioural intentions" (Stylidis et al., 2020, p. 2). In addition, emotional solidarity is "a significant predictor of how people think or behave in relation to tourism" (Joo et al., 2020, p. 340).
Tourist's emotional connections to destinations affect future travel intentions (Yuksel et al., 2010). Stylidis et al. (2020) argue for the importance of emotional solidarity for destinations in term of its effect on tourists' relationship to a destination and its residents with respect to destination image. The tourists' destination image cannot be studied without considering tourists feelings towards or their relationship with residents (Tsai, 2012). Familiarity also directly influences tourists' destination image (Baloglu, 2001) and behavioral intentions (Ha & Jang, 2010). Here we can conclude that prior or potential visitors have different degrees of emotional solidarity with destinations and residents that can affect their travel intentions through destination image. Based on the preceding discussion, the following hypotheses are proposed: H4 a : There is significant difference for the effect of solidarity on intention to travel to a destination after COVID-19 crisis between tourists with past experience and without past experience.
H4 b : There is significant difference for the effect of solidarity on willingness to support a destination between tourists with past experience and without past experience. Fig. 1 shows the conceptual framework of this study.

Research methodology
This study applied a quantitative method using online selfadministered method through social media to collect data.
The pilot and pre-test were conducted by interviewing with five experts to check the items, and also distributing of the questionnaire among 35 respondents and check the reliability using Cronbach's alpha. The results showed the Cronbach's alpha greater than 0.8 for all constructs indicating acceptable reliability from pilot test. The respondents were also asked about their past experience in visiting the selected country and were therefore the collected data were divided into two groups for analysis and hypothesis testing based on their answers.

Data collection
The data were collected from respondents who followed COVID-19 news about ten specific countries (China, South Korea, Italy, Germany, Iran, USA, Sweden, UK, New Zealand, and Turkey) and plans to visit those countries in the future. The ten countries were selected because of their different approaches in handling the COVID-19 outbreak and their various levels of infections and mortality rates, which may influence the images of these destinations (Pew Research Centre, 2020). At the beginning of the questionnaire, respondents were asked to identify the country that they followed most with respect to COVID-19 related news, other than their country of origin and country of permanent residency. Then, the respondent was transferred to the section related to selected country to answer the questions. Given that the data were obtained from prospective tourists for particular destinations, and so the respondents could be from different countries, and the researchers do not know about origin language of respondents, the questionnaire was administered in English. This study used social media (e.g. Facebook, LinkedIn, Twitter) to distribute the online questionnaire and collect the data, which has been recommended in recent literature as the most effective approach to collect online data with an average response rate of 54% in previous studies (Ali et al., 2021). The data for this study were collected from May to August 2020.
A total number of 542 questionnaires were completed from the respondents who followed the news about COVID-19 for ten selected countries, however only 518 questionnaire who answered the question about previous visit were included in this research. Out of 518 respondents, 286 (55%) had not previously visited the selected country and 232 (45%) respondents had already visited the selected destination. Two hundred and sixty respondents (50.2%) were male, 254 (49%) were female, and the rest did not identify their gender. The majority of respondents (360) belonged to two age groups; 25-34 and 35-44 groups and had college and university level education (466). To check the Common Method Variance (CMV), two recommended methods for PLS-SEM were applied, namely the full collinearity variance inflation factor (VIF) (Kock, 2015), and the correlation matrix procedure. To check the CMV using fill collinearity VIF, the literature recommends a threshold of 5 for a composite-based approach (Kock & Lynn, 2012). According to the correlation matrix procedure, the value of correlation between constructs should be lower than 0.9, to demonstrate that the CMV is not an issue in the research model. The results showed the full collinearity VIF for all constructs ranged 1.537-4.51, and correlation between constructs was less than 0.9, indicating the model free of CMV.

Analytical technique
A combination of partial least squaresstructural equation modeling (PLS-SEM) and multi-group analysis (MGA) was applied to assess measurement model and structural model and test developed hypotheses. The PLS-SEM approach was adopted because the study is predictionoriented research which aims to predict future behavioral intentions of prospective tourists to support and travel to a destination post COVID-19. In addition, this study aims to compare the effects of antecedents on dependent variables using MGA, which is a non-parametric analysis, and PLS-SEM is recommended as a non-parametric approach (Hair et al., 2019;. SmartPLS 3.0 software package  has been used to perform PLS-SEM and MGA. We assessed reliability validity and predictive power of model using PLSpredict for two groups separately (Hair et al., 2019). Prior to perform MGA to compare the path coefficients between two groups and hypothesis testing, we tested measurement invariance using the measurement invariance for composite (MICOM) approach (Henseler et al., 2016;Rasoolimanesh et al., 2017). In order to test measurement invariance, we checked the configurational invariance, compositional invariance, and equality of mean and variance (Gannon et al., 2021). To establish, the partial measurement invariance, which is the requirement for MGA, the configurational invariance, and compositional invariance should be established (Henseler et al., 2016). To perform MGA, and test hypotheses, we applied two most conservative non-parametric approaches the Henseler's MGA  and the permutation test (Rasoolimanesh et al., 2017).
The adequacy of data for performing the analysis was checked using power analysis and G*Power software (Faul et al., 2009). A minimum sample of 191 was required for each group (e.g. respondent with and without previous visit to a destination) to get a power of 0.95. In addition, based on Reinartz et al. (2009), a minimum of 100 is sufficient to perform PLS-SEM. Therefore, our sample for each group (286 without previous visit vs. 232 with previous visit) is sufficient to run the analysis. Table 1 shows the results of the descriptive analysis for the items to measure each construct, as well as skewness and kurtosis. The results show the values of skewness of all items between − 1 and 1, and the value of kurtosis between − 2 and 2 indicating the distribution of data close to normal distribution (Hair et al., 2017) and no issue of normality. Although the PLS-SEM is a non-parametric approach and can handle non-normal data, recent literature recommends to avoid very highly non-normal distributed data, which is not the case in this study based on the values of skewness and kurtosis (

Assessment of measurement model
In the first step of model assessment using PLS-SEM, the measurement model of six reflective constructs namely trust, crisis management, healthcare system, solidarity, travel intention, and willingness to support a destination, were assessed using reliability and validity criteria . The composite reliability (CR), rho_A, and average variance extracted (AVE) should be greater than 0.7, 0.7, and 0.5 respectively to establish reliability and convergent validity (Hair et al., 2019). The results presented in Table 2 show acceptable reliability and convergent validity for all constructs using the CR, rho_A and AVE for the data from two groups (e.g. respondents with and without past experience to visit a destination).
To establish discriminant validity, two most conservative approaches namely the Fornell-Larcker criterion and the heterotrait-monotrait (HTMT) ratio have been applied (Hair et al., 2017;Henseler et al., 2015). According to Fornell-Larcker criterion, the square root of AVE of each construct should be greater than the correlation with any other construct in the model, and the value of HTMT ratio should be less than 0.9 to establish discriminant validity (Hair et al., 2019;Henseler et al., 2015). Table 3 and Table 4 show acceptable discriminant validity based on both criteria and for data from two groups (e.g. respondents with and without past experience to visit a destination).
The measurement invariance should be tested between two groups before assessment of structural model and hypothesis testing. The results of measurement invariance testing using the MICOM approach have been presented in Table 5. The results show partial measurement invariance by establishing configural and compositional invariance, which is a requirement to perform MGA (Hair et al., 2019;Henseler et al., 2016). However, the results of MICOM could not support full measurement invariance, which there are significant differences between mean and variance of some constructs for the data from two groups (e.g. respondents with and without past experience to visit a destination). Therefore, the assessment of the structural model can be reported for two groups separately.

Assessment of structural model and hypothesis testing
This study aims to predict the future behavioral intention of tourists based on their perceptions about destination shaped by media. Therefore, this study is a prediction-oriented study, and prediction power is assessed for two groups. To assess predictive power of model, both insample approach using the values of R 2 and Q 2 for endogenous constructs and the out of sample approach using the PLSpredict have been applied (Hair et al., 2019). The results identified the value of 0.558 and 0.361 for R 2 , and 0.452, and 0.292 for Q 2 of willingness to support a designation and travel intention respectively for the data from respondents with past experience to visit a destination, the values of 0.516 and 0.317 for R 2 and 0.410, 0.278 for Q 2 respectively for the respondents without past experience to visit a destination. The results demonstrated acceptable in-sample predictive power for both groups (Hair et al., 2017). Using out of sample PLSpredict approach, the predictive power was assessed for two groups, and the results showed high predictive power of model to predict willingness to support a designation and travel intention for two groups by assessing the value of Q 2 predict for the items of each construct and comparing the root mean squared error (RMSE) between PLS-SEM and linear model (LM) (Danks & Ray, 2018). Table 6 shows the results of PLSpredict for two groups indicating high predictive power for the model based on data from two   groups. Table 7 shows the results hypothesis testing and MGA to compare the path coefficients between two groups using two most conservative nonparametric approaches namely the Henseler's MGA ) and the permutation test (Rasoolimanesh et al., 2017). The results supported H1a and H1b and showed significant difference for the effects of trust on willingness to support a destination and travel intention, which these effects are significantly higher for the respondents without past experience to visit a destination. The results of this study could not support the effects of crisis management on willingness to support a destination and travel intention for both groups, and so the differences between two groups (H2a, H2b). The findings of this study demonstrated the significant differences of the effects of healthcare system on willingness to support a destination and travel intention, supporting H3a and H3b. The results showed positive and greater effects of healthcare system on willingness to support a destination and travel intention for the respondents without past experience to visit a destination. Finally, the results strongly supported H4a and H4b, showing significant differences for the effects solidarity on willingness to support a destination and travel intention between two groups, which the highest effects belong to the respondents with past experience to visit a destination.

Discussion
This study aimed to predict travel intention and willingness to support a destination of prospect tourists for the post COVID-19 era, based two groups of respondent's image of a destination (e.g. respondents with and without past experience to visit a destination). This study considers the cognitive aspect of destination image, shaped by the media during COVID-19 outbreak period, and based on the ways that countries (i.e., destinations) managed COVID-19. In this study, the effects of trust, crisis management, healthcare system, and solidarity, as the components of destination image on future travel intention, and willingness to support a destination have been examined and compared between two groups. The results of this study showed the high predictive power of model for both groups (e.g. respondents with and without past experience to visit a destination).
In addition, the results showed significant difference and highest effects of trust on willingness to support a destination and travel intention, for the respondents without past experience to visit a destination. These results highlight the important role of trust to the government of destinations for future travel, for the first-time tourists, and when the tourists have no past experience to travel to a destination. This reflects the results of previous studies (e.g. Abubakar et al., 2017;Han & Hyun, 2015;Tsai, 2012). One possible explanation for this is that those with prior knowledge of a destination gained through a prior visit have certain trust and this minimizes uncertainty and the perception of risk in future travel (Han & Hyun, 2015). This is consistent with the findings of other studies which found that repeat visitors tended to have a stronger intention to return and word-of mouth communications (Li et al., 2008). This highlights that a trustworthy tourist destination has strong competitive advantage compared to other destinations.
At the same line, the results also highlighted the significant differences and greatest effects of healthcare system on willingness to support a destination and travel intention, for the respondents without past experience to visit a destination, indicating the importance of healthcare system for first-time tourists to visit a destination. This is in line with previous studies which also highlight that trust in destination services play a significant role in destination attractiveness and affect travel intentions (Abubakar & Ilkan, 2017). Our findings support the argument that quality of healthcare system in a destination is a top priority for tourists (Crouch, 2010) particularly for non-visitors as they usually have limited knowledge and familiarity with the destination and may therefore be more risk averse (Tan & Wu, 2016). Here we can argue that past experience can positively influence the image an individual has of a destination and their choice of destination (Chen & Lin, 2012), particularly at times of crisis when media broadcasts may generate fear and panic (Hall, Prayag, et al., 2020. The results also demonstrated the importance of solidarity for the respondents with past experience to visit a destination, to predict their behaviour to support and travel to a destination for post-COVID-19 era. The findings identified strong effects solidarity on willingness to support a destination and travel intention for the respondents with past experience compared to the respondents without past experience of visiting a destination. This supports previous studies that argued the significant role of emotional solidarity between tourists and residents in shaping a favorable destination image (Stylidis et al., 2020) and affecting future travel intentions (Yuksel et al., 2010). Our study confirms that previous vitiators to a destination can have different degrees of emotional solidarity with a destination and its residents which can play a significant role in their future travel intentions. However, the lack of such emotional solidarity in individuals with no past experience can weaken the willingness to visit a destination post pandemic.

Theoretical contributions
Taken collectively, the theoretical contributions of this study are three-fold. First, it develops an integrated framework to address the questions on whether and how trust, solidarity, healthcare system and crisis management triggers future travel intention post pandemic, providing empirical evidence and a theoretical basis for the effect of these antecedents in a global pandemic. Prior research supports the significant association between intra-pandemic perception and postpandemic travel intention (Li et al., 2020). However, the predictors of future travel intention and the role of media have not been investigated in prior research and this makes this study's theoretical contribution unique. Second, it extends literature on the moderating effects of past experience linked to COVID-19 which has not been assessed in prior research. This is particularly important from the theory-building perspective as the results of this study showed the high predictive power of model for both groups of respondents. While prior research suggests that past experience is a key determinant of destination image during crises, our research shows that how cognitive aspect of destination image can be projected by both mass and social media signifying the effect of the components such as trust, healthcare system, and solidarity on future travel intention, for individuals with and without past experience. Third, the extant studies on COVID-19 and tourism tend to be country-specific, and a cross-country analysis has been neglected. As a pioneer study, this research used cross-country data of ten countries to reveal connections between media framed destination image through the national responses dealing with pandemic and the post-pandemic travel intention to those countries.

Practical contributions
Encouraging domestic and international tourism activity is the ultimate goal in post pandemic recovery strategies for many destinations. However, perceived health risk has negatively influenced tourists' perceptions, destination image and travel intentions. The identification of appropriate interventions in response, such as destination media profiling and recovery marketing, will play critical roles in mitigating the effects of tourists perceived risk (Kim et al., 2021;Matiza, 2020). Therefore, the findings of this study provide important managerial implications and essential insights particularly for DMOs to address the current evolving tourists' perceptions and destination image during the COVID-19 pandemic.
As attested by this research, media framing of a crisis can immediately affect the image of a destination particularly for individuals with no past experience or those who have not visited the destination before. Given the magnitude of the COVID-19 pandemic and its extensive media coverage, image restoration should be a key priority for DMOs through marketing and advertising programs, media relations and crisis communication techniques which can demonstrate a reliable and trustworthy destination. In addition, the results showed that trust and solidarity are strong predictors of individuals' future travel intention. Transparent communications regarding the pandemic can potentially improve the destination reputation and enhance individuals' perceived trust in a destination. As noted by the OECD (2020), fostering public trust in government and ensuring effectiveness need to be guided by the principles of transparency, integrity, accountability, and stakeholder participation. At national level, officials are suggested to leveraging public communication to counteract disinformation and support policy. Public communication is a key element of such agenda and is required to be handled through digital platforms and channels and media markets through which information is framed and delivered (see also Li et al., 2020).
Effective communication and dissemination of risk messages by destinations can help tourists to perceive and cope with risks more appropriately, while inadequate communication may amplify perceived risk thereby inducing lasting negative impacts on destination images and intentions to travel (Sano & Sano, 2019). Therefore, the agendasetting and media communication of risk messages are of great importance in crisis responses and risk interventions, which are critical for destinations in preserving their images.
As a pioneer study, this research attempted to investigate the moderating effect of potential tourists' prior experience on behavioral intentions in the contexts of the COVID-19 pandemic. Past experience is a powerful factor in moderating the effects of trust, healthcare system, and solidarity on travel behavioral intention. The findings of this study suggest that destination managers are required to develop different strategies to target individuals with and without past experience. Previous experience and destination familiarity will lead to a greater feeling of safety and confidence in destination choice and may potentially therefore become one of the first market segments targeted in post-COVID promotions to aid destination recovery. However, in contrast, individuals with no previous experience or visit to the destination will require higher level of assurance and greater confidence to visit a destination. It is particularly important in times of crisis such as the COVID-19 pandemic when assurance can enhance the formation of positive destination image and raise the probability of future travel intentions.

Limitations and directions for future research
The empirical results of this study should be considered in the light of some limitations which constitute potential lines of research for the future. First, data gathered from an online survey has coverage difficulties similar to other surveys (Hwang & Fesenmaier, 2004). Despite our efforts to distribute the survey through different channels (including social media platforms) which led to an acceptable response rate, the use of an online survey for this study may have resulted in a selection bias for respondents as most online surveys do (Hwang & Fesenmaier, 2004). Future research is suggested to expand the scope of investigation by alternative data collection methods such as qualitative and longitudinal research methods. Longitudinal methods are suggested for future research to examine what factors would influence tourists' intention during and after the COVID-19 outbreak. Furthermore, the relationship between responding to the survey and the specific COVID-19 infection cases and mortality rate at that time which may have affected the responses of respondents.

Appendix 1. List of Adapted items
Trust TR1 I trust the information and communication provided by the government of destination A about the infection and mortality rate of COVID-19 TR2 I have admired the openness and transparency of the government of destination A since the beginning of the COVID-19 outbreak TR3 I would characterize the government of destination A as honest in their response to COVID-19 TR4 My level of trust in the information and communication provided by the government of destination A has increased since the start of the COVID-19 outbreak.

CM1
I admire the timely and early diagnosis and treatment of Covid-19 by the authorities of destination A CM2 I admire the forceful and rapid response to COVID-19 in destination A CM3 I admire the effective management of the supply chain risk and disruption (shortage of food and medicine) in destination A Health Care System HCS1 I believe the healthcare system of destination A is reliable and robust HCS2 I admire the timely and fast tracking of those exposed to COVID-19 in destination A HCS3 I admire the high-capacity for COVID-19 testing in country A Solidarity SOL1 I believe the government of destination A cared about its people during COVID-19 crisis SOL2 The government of destination A did the best to relieve and decrease the emotional and mental impacts of COVID-19 on affected people SOL3 In destination A, all groups of people help and care about each other during COVID-19 crisis SOL4 The government and people of destination A were close and together to overcome the COVID-19 crisis SOL5 People help each other in different ways financially and non-financially to relieve and decrease the emotional and mental impacts on affected people Willingness to Support a Destination WSD1 I would encourage my friends and relatives to travel to destination A after COVID-19 crisis WSD2 I say good things about destination A on social media WSD3 I would promote this destination to help tourism recovery WSD4 If the destination agency/someone from the destination asked me to promote the destination, I would do all my efforts to do promote the destination.

TI1
If given the opportunity, I am willing to travel to destination A after COVID-19 TI2 I am planning to travel to destination A after COVID-19 in the near future TI3 The likelihood of my travel to destination A is high