Role of Non-Pharmaceutical Interventions for COVID-19 in Cruise Tourists’ Decision-Making Process: An Extended Model of Goal-Directed Behavior

: This study developed an extended model of goal-direct behavior (EMGB) to investigate the role of non-pharmaceutical interventions (NPIs) in predicting latent cruise travelers’ decision-making process in the COVID-19 pandemic context to foster the growth of a sustainable cruise business. An online survey was conducted for Korean respondents with experience of being on a cruise, and a total of 288 valid data were collected. The proposed model was examined using SmartPLS 3.0. Results show that in general, antecedents of MGB affected the desire, which in turn inﬂuenced cruise travelers’ behavioral intentions. Desire was also found to affect NPIs, which inﬂuenced behavioral intentions. Results reveal that the perception of COVID-19 affected NPIs and behavioral intention during COVID-19. Findings provide academia with theoretical implications and cruise managers with practical implications.


Introduction
The outbreak of COVID-19 has seriously affected the world. A series of national control measures (e.g., travel restrictions, border shutdown) have been implemented globally to impede the pandemic, and tourism is one of the hardest-hit sectors. According to the UNWTO, in 2020, an unprecedented recession was witnessed with the loss of international tourist arrivals (1 billion), international tourism receipts (US$ 1.3 trillion), and global GDP (over US$ 2 trillion), setting back international tourism by 30 years [1]. At the same time, the global cruise tourism industry was also affected greatly by the COVID-19 pandemic. The Cruise Lines International Association (CLIA) claimed that almost all the cruise lines in the world have been shut down since the COVID-19 virus spread on the Diamond Princess ship in February 2020 [2]. Data from the World Health Organization (WHO) indicated that in this incident, a total of 706 cruise passengers were infected and six died [3]. This incident has resulted in reputational damage to the cruise industry and has also shed light on the high risk of COVID-19 transmission on a cruise. According to the Centers for Disease Control and Prevention (CDC), people from all over the world gathered and participated in activities together in the cabin rooms with poor ventilation during the days and weeks on board, which accelerated the spread of the COVID-19 virus [4]. In a situation where the accurate effects on the global cruise industry remain unclear, cruise operators have struggled with heavy financial losses including huge costs of maintenance and employment because of the COVID-19 pandemic.
As tourism restarts slowly in an increasing number of countries, cruise operators are planning for a phased and controlled resumption of operations in 2021 [5,6]. Previous research on health and the pandemic, coupled with limited practical experience on cruise lines, has suggested that non-pharmaceutical interventions (NPIs) could help prevent Sustainability 2021, 13, 5552 2 of 15 humans from contracting COVID-19 while traveling [5]. The NPIs refer to measures to impede the transmission of COVID-19 without the use of drugs [7], such as wearing a mask, keeping a social distance, improving personal hygiene practices, etc. A new NPIs framework for cruise operators has been established by CLIA based on the recommendations of health experts, local and regional governments, and the resumed cruise lines [5] with the aim of mitigating the risk of COVID-19 during cruise trips and to facilitate sustainable resumptions of the global cruise industry. For example, cruise companies have reinforced ventilation facilities by changing the air circulation structure on board, making COVID-19 tests mandatory for all passengers before boarding, and operating the facilities on the ship with physical distancing during the cruise tour. In addition, the safety of these procedures was proven with objectivity and reliability by obtaining a COVID-19 safety quarantine approval from an international classification ship company [5]. To date, over 200 successful sailings have been successfully carried out around the world with these measures in place and support from local governments [5].
Although most of the cruise operators are ready to restart with more health-protective measures [5,6], customers' attitudes toward the new NPIs framework and their intentions to take cruise trips, which are essential to operators' marketing and business strategies in the near future, remain unclear because of the limited practice and research on NPIs measures on a cruise. Therefore, research is necessary for the cruise sector to comprehend and predict potential travelers' behaviors, particularly when new service arrangements (e.g., NPIs) or barriers (e.g.,  appear. Based on a literature review [8][9][10][11], a model of goal-directed behavior (MGB) was found to be useful in understanding individuals' decision-making process when traveling in certain situations. Previous studies have stated that during the pandemic, individuals' travel behaviors were very likely to be influenced (cancel or change) by the perceived risk of the disease on the grounds of human instinct for self-preservation [12,13]. Additionally, adaptive behavior (e.g., NPIs) to a specific tourism barrier (e.g., H1N1 pandemic) can modify tourists' intention to take a tour during the pandemic period [14,15].
However, the important measures for NPIs in COVID-19 research on the effects of NPIs on cruise travelers' behavioral intentions are limited. Thus, this study fills this gap by developing a research framework using an extended model of goal-directed behavior (EMGB) that incorporates new variables of cruise operators' NPIs and travelers' personal NPIs, as well as perception of COVID-19, into the original MGB constructs. The new EMGB model in the context of cruise can be used to better understand the latent cruise travelers' decision-making process and identify the predictability of the two NPIs to their cruise travel intentions during and after the COVID-19 pandemic. Findings of this study will contribute to extending not only the research framework of the social-psychological behavior of cruise tourists but also sustainable management of the global cruise industry in the COVID-19 pandemic.

Model of Goal-Directed Behavior (MGB)
As a representative socio-psychological theory to understand an individual's decisionmaking process, Perugini and Bagozzi [16] developed the MGB, which was extended from the theory of reasoned action (TRA) and the theory of planned behavior (TPB) [17]. MGB posits that attitude, subjective norm, positive anticipated emotion, negative anticipated emotion and perceived behavioral control affect desire, which in turn affects behavioral intention [15]. Attitude refers to one's favorable or unfavorable evaluation of performing a specific behavior [18], while subjective norm refers to one's perceived social pressure, such as perceptions of relevant groups (e.g., family, friends, peers, colleagues) on performing a specific behavior [17]. Perceived behavioral control refers to one's confidence to conduct a specific behavior [17]. Positive anticipated emotion and negative anticipated emotion refer to the expected affective response to achieve a specific goal or fail to achieve the goal [19]. Desire refers to a motivational state of mind related to the emotion of longing or hoping Sustainability 2021, 13, 5552 3 of 15 for a person, object, or outcome [16,20]. Behavioral intention refers to one's willingness to perform a specific behavior [19]. Lee et al. [21] suggest that MGB can perform better than TRA and TPB when analyzing tourist's behavior.

Perception of COVID-19
According to the health belief model (HBM) [22], individuals' perception of a specific disease comprises perceived susceptibility and perceived severity of the disease. The former is defined as one's subjective assessment of contracting the disease, while the latter refers to one's subjective assessment of the seriousness of the disease and its potential consequences, including physical, psychological, or social negative effects [22]. Regarding a specific disease, the HBM suggests that high perceived susceptibility and strong perceived severity are more likely to motivate individuals to adopt actions to prevent it. Similar views were discussed in Alifano et al. 's [23] research, which introduced the effects of the COVID-19 pandemic on European society in physical, psychological and social perspectives. They claimed that perception of COVID-19 infection risk and health threat could positively intensify individuals' appropriate risk-avoidance behaviors, for example, cancelling their travel plans [14], or adopting NPIs in their trips [15]. These risk-avoidance behaviors are essential to the tourism industry during the pandemic.
COVID-19 is a novel infectious disease that can be transmitted rapidly from humans and objects with the virus through respiratory droplets, with a two-week incubation period [24]. Anyone can be infected and show symptoms such as fever, headache, sore throat, and even difficulty breathing, and elderly people and children are at high risk of infection [24]. People infected with COVID-19 can be treated with medicine but may be at the health risk of sequelae [24]. In the context of tourism, researchers have conducted studies pertaining to individuals' perception of infectious diseases (e.g., 2009 H1N1 influenza, Malaria, Dengue, Zika, Chikungunya, West Nile, COVID-19) to predict tourists' travel intention [14,15], tourists' health-protective behaviors [25], and destination residents' support for tourism development [26].

Non-Pharmaceutical Interventions (NPIs) for COVID-19
WHO suggests that individuals could protect themselves against COVID-19 through pharmaceutical (i.e., vaccination) and non-pharmaceutical interventions [7]. As the most effective way to stop COVID-19 transmission, vaccination has been started in some countries, but it remains unavailable for most people because the limited production capacity cannot meet the global demand, and some countries may not be able to afford enough vaccines [14,27]. Moreover, the patents of vaccines are held in just few pharmaceutical firms and research institutes in few countries in the world. If these patents of vaccines are not shared with other firms and countries, it is difficult to produce a worldwide large-scale vaccine [28,29]. Therefore, NPIs for COVID-19 are more widely used among people. NPIs are related to measures to interrupt or reduce the transmission of COVID-19 without using vaccines and medicines, which have been proven to be effective, available, and affordable in early stages of the COVID-19 outbreak [30]. Van Bavel et al. [31] also assert that the effectiveness of the behavioral response to COVID-19 through NPIs can be supported by social and behavioral science. WHO [7] classified NPIs for COVID-19 into four types, namely, personal protective measures (face masks, hand hygiene, and respiratory etiquette), physical distancing measures (school/workplace measures and closures and avoiding crowds), environmental measures (surface and object cleaning), and travel-related measures (internal travel restrictions, travel advisories, entry and exit screening, and border closures). Although some of the NPIs (e.g., travel restrictions, border closures) have led to unprecedented worldwide economic recessions, their contributions to halting the spread of COVID-19 cannot be ignored [7]. Lee et al. [15] have provided evidence that NPIs played an important role in tourists' decision-making process in the 2009 H1NI background. Recently, Das and Tiwari [14], who conducted a research following Lee et al.'s study [15] in the COVID-19 background, obtained consistent results. Considering the characteristics of the cruise tour, this study investigated NPIs related to cruise tours from the perspectives of travelers and cruise operators. Travelers' NPIs (T-NPIs) refer to personal protective measures, whereas operators' NPIs (O-NPIs) refer to procedures, including COVID-19 testing for all travelers and crew before embarkation, elevated sanitation measures, medical services, social distancing on board, protected ashore visits at destinations, the COVID-19 prevention guidelines for passengers, and other safety measures from the time of booking through the cruise itinerary and at disembarkation [5,7].

Relationships among Constructs in MGB
The MGB suggests that desire is considered as the most proximal cause of intention among the antecedents of behavioral intention [16], while other antecedent constructs (e.g., attitude, subjective norm, positive anticipated emotion, negative anticipated emotion, and perceived behavioral control) affect behavioral intention through desire. In the past 20 years, previous studies have widely used the MGB to explain tourists' decisionmaking process in various tourism settings, such as the Korean wave [32], cultural heritage [9], cruises [33], overseas tourism [34], casinos [19], and mega-sporting events [8].
Han et al. [35] used it in the context of environmentally responsible museum tourism and found that visitors' attitude, subjective norm, positive anticipated emotion, negative anticipated emotion, and perceived behavioral control positively affected their desire, which in turn positively affected behavioral intention. The consistent results were confirmed by Meng and Choi [10] in their study on slow tourists' behavior. Attanasi et al. [36,37], in their studies of mass-gathering events, found that individuals' behavioral intention was positively influenced by subjective norm and their positive anticipated emotion. Recently, Lee et al. [9] examined Korean Hanok visitors' intention formations in which attitude, subjective norm, positive anticipated emotion, negative anticipated emotion were found to have positive effects on desire, which influenced behavioral intention. Furthermore, Choe et al. [8] explored Rio 2016 to identify US people's behavioral intention towards Rio 2016 and found that attitude, subjective norm, positive anticipated emotion, negative anticipated emotion, and perceived behavioral control had significant influences on potential tourists' desire to attend the event, and that desire affected their behavioral intention in two aspects: behavioral intention to attend the event physically and to watch the game via media platforms. In this study, negative anticipated emotion used expressions with double negatives (e.g., "If I cannot take a cruise tour, I will be disappointed"), thus indicating a positive effect on desire. This study also investigates cruise tourists' behavioral intention during COVID-19 and after the COVID-19 pandemic separately. Based on previous research, the following hypotheses are postulated:   Hypothesis 6b. Desire positively affects behavioral intention after COVID-19.

Relationships among Perception of COVID-19, NPIs, and Behavioral Intention
The health belief model (HBM) [22] suggests that individuals' perception of a disease is positively related to their health-protective behaviors (e.g., pharmaceutical interventions, NPIs). Based on the HBM, individuals who perceive a high risk of being affected by a disease are more willing to take actions to protect themselves from developing the condition, including personal health-protective behaviors and organizational or social health-protective behaviors. Lee et al. [15], in their work on Koreans' international travel intention during the 2009 H1N1, found that individuals' perception of the 2009 H1N1 had a positive effect on their willingness to adopt NPIs. Das and Tiwari [14] also found consistent results in India that domestic and international tourists who perceived higher risk of the COVID-19 were more likely to adopt NPIs. Thus, this study postulates the following hypotheses: Hypothesis 7a. Perception of COVID-19 positively affects O-NPIs.

Hypothesis 7b. Perception of COVID-19 positively affects T-NPIs.
In addition, during the pandemic period, tourism activities are believed to be at high risk because they represent a series of cross-regional activities, where the virus can be spread by tourists [15]. Reisinger and Mavondo [13] revealed that tourists' decision-making process was associated with their travel risk perception and that their travel intention may decrease with the increase in perceived risk of a pandemic [38]. Furthermore, Das and Tiwari [14] and Lee et al. [15] claimed that individuals' perception of a specific pandemic had a negative effect on their behavioral intention through NPIs. And their studies [14,15] also revealed that NPIs had positive influence on travelers' behavioral intention. Therefore, the following hypotheses are postulated: Hypothesis 8a. Perception of COVID-19 negatively affects behavioral intention during COVID-19. Hypothesis 9d. T-NPIs positively affect behavioral intention after COVID-19.

Relationships between Desire and NPIs
The behavior change theory suggests that coping plans for anticipated barriers can help improve individuals' behavioral intention toward a specific goal [39][40][41]. According to the self-determination theory, human behavior may be motivated intrinsically by basic psychological needs (e.g., interest, desire, enjoyment) [42,43]. Self-determined motivation could directly predict individuals' efforts for goal achievement with the regulation of human autonomy [44,45]. In other words, individuals with stronger desire are more active in their goal-achievement process, for example, adopting much more proactive approaches to deal with the obstacles. Therefore, travelers' desire to take cruise trips is assumed to have a positive effect on their willingness to adopt the NPIs. Therefore, the following hypotheses are postulated: Based on the literature and hypotheses, a research model is proposed as depicted in Figure 1.

Measures
Attitude, subjective norm, positive anticipated emotion, negative anticipated emotion, perceived behavioral control, and desire in the MGB were assessed with 3 items, each adapted from Lee et al.'s study [15]. Behavioral intention was measured in two dimensions (behavioral intention during COVID-19 and after COVID-19) with four items each, based on Lee et al.'s study [15]. PC was measured using five items adapted from Das and Tiwari's [14] and Lee et al.'s [15] studies. NPIs were measured in two dimensions (O-NPIs and T-NPIs), where O-NPIs consisted of four items adapted from WHO [7] and CLIA [5], and T-NPIs consisted of three items adapted from Lee et al.'s study [15]. All items mentioned above were measured on a five-point Likert-type scale, where 1 is strongly disagree and 5 is strongly agree. Sociodemographic variables were also included in this study. The questionnaire was developed in English and then translated into Korean, and then backtranslated into English to ensure consistency with the original measurement items. Two experts in the area of cruise tourism were invited to review whether these measurement items are appropriate in the cruise context. To ensure the readability of the questionnaire, this study also conducted a pre-test with 30 tourists with experience of being on cruises. Some minor wording changes (e.g., tense) were implemented for clarity.

Data Collection and Analysis
The target population of this study were Koreans over 20 years old who had taken a cruise tour in the past. Online surveys were conducted using Google Forms from 10 October 2020 to 10 February 2021. Given that the target sample is a small population compared to the general tourists in South Korea, the survey was completed with the help of a travel agency that specializes in cruise tours, the Korea Cruise Lines Association, and 10 staff from three cruise companies in Korea. The survey link was distributed through the travel agency and the cruise staff to their guests and through the Korea Cruise Lines Association to its members. A total of 314 samples were collected, but 26 responses were

Measures
Attitude, subjective norm, positive anticipated emotion, negative anticipated emotion, perceived behavioral control, and desire in the MGB were assessed with 3 items, each adapted from Lee et al.'s study [15]. Behavioral intention was measured in two dimensions (behavioral intention during COVID-19 and after COVID-19) with four items each, based on Lee et al.'s study [15]. PC was measured using five items adapted from Das and Tiwari's [14] and Lee et al.'s [15] studies. NPIs were measured in two dimensions (O-NPIs and T-NPIs), where O-NPIs consisted of four items adapted from WHO [7] and CLIA [5], and T-NPIs consisted of three items adapted from Lee et al.'s study [15]. All items mentioned above were measured on a five-point Likert-type scale, where 1 is strongly disagree and 5 is strongly agree. Sociodemographic variables were also included in this study. The questionnaire was developed in English and then translated into Korean, and then back-translated into English to ensure consistency with the original measurement items. Two experts in the area of cruise tourism were invited to review whether these measurement items are appropriate in the cruise context. To ensure the readability of the questionnaire, this study also conducted a pre-test with 30 tourists with experience of being on cruises. Some minor wording changes (e.g., tense) were implemented for clarity.

Data Collection and Analysis
The target population of this study were Koreans over 20 years old who had taken a cruise tour in the past. Online surveys were conducted using Google Forms from 10 October 2020 to 10 February 2021. Given that the target sample is a small population compared to the general tourists in South Korea, the survey was completed with the help of a travel agency that specializes in cruise tours, the Korea Cruise Lines Association, and 10 staff from three cruise companies in Korea. The survey link was distributed through the travel agency and the cruise staff to their guests and through the Korea Cruise Lines Association to its members. A total of 314 samples were collected, but 26 responses were eliminated because of insincere responses (e.g., pattered responses, outliers). As a result, 288 responses were used to test the hypotheses, representing 91.7% of the response rate. Power analysis of the sample size was carried out using the G*Power 3.1 program recommended by Hair et al. [46]. Testing results showed the sample size of this study (n = 288) was acceptable because the proposed model achieved a common statistical power of 80% for detecting R 2 values above 0.25 at the significant level of α = 5% [46,47].  Table 1 presents the respondents' demographic characteristics. The proportion of males (48.3%) were somewhat similar to that of females (51.7%). About 25.7% of the respondents were in the age group of 40-49, followed by the respondents aged above 59 (24.3%), 30-39 (23.3%), 50-59 (20.5%), and 20-29 (6.3%). A total of 77.1% of the respondents had at least a university degree, and most of the respondents (66.0%) were married. Over half (54.9%) of the respondents had a monthly income of more than 4 million Korean Won. Moreover, 24.7% of the respondents were office workers, followed by self-employed (16.3%).

Measurement Model
As mentioned earlier, the proposed research model was estimated with PLS-SEM using SmartPLS 3.0. According to Hair et al. [49,50], the model fit of the measurement model can be assessed using criteria such as factor loading (>0.7), composite reliability (CR) (>0.7), Cronbach's α (>0.7), and average variance extracted (AVE) (>0.5). For the discriminant validity, the square roots of AVE values should be greater than the correlations of corresponding constructs [51]. Table 2 shows that all factor loadings were above the cut-off value of 0.7 and AVE values were above the required criteria of 0.5, thereby confirming convergent validity of the measurement model. Cronbach's α and CR were also above the required criteria of 0.7, thereby supporting the reliability of the model. Table 3 shows that all square roots of the AVE values were greater than the correlations of the corresponding constructs, which confirms discriminant validity [51].

Indirect and Total Effects
By running the upper bootstrap procedure, the indirect and total effects among the constructs were assessed as well. Table 5 shows that behavioral intention during COVID-19 was most affected by desire (0.

Conclusions and Implications
During a pandemic, finding an appropriate way to promote the sustainable development of the cruise industry is important. In addition to international travel restrictions, concerns about COVID-19 infection during travel is one of the main factors that have negative effects on individuals' travel intentions [7]. Although the non-pharmaceutical interventions (NPIs) for COVID-19 could prevent infection effectively in research and practice, what kind of role the NPIs play in potential cruise travelers' decision-making processes remains unclear. Hence, to address this study gap, the current study developed a research framework using an EMGB and found that NPIs played significant roles in predicting cruise travelers' behavioral intentions. Results have also shown that among the five predictors in the original MGB, attitude, subjective norm, negative anticipated emotion, and positive anticipated emotion significantly affected desire, which coincided with previous research [9,15]. Desire and perception of COVID-19 had significant direct and indirect effects on travelers' behavioral intentions during COVID-19 via O-NPIs and T-NPIs. T-NPIs and desire had significant direct effects on their travel intention after COVID-19, while perception of COVID-19 and O-NPIs had no significant direct influence on travel intention after COVID-19. A plausible explanation is that the O-NPIs may be thought of as unnecessary by tourists after COVID-19 due to its tedious work, extra expenditures, and inconveniences during travel. Although some apprehensions remain, T-NPIs will be effective enough to protect tourists at the time. The findings of this study provide significant theoretical and practical implications for the sustainable management of the cruise industry during a pandemic crisis.

Theoretical Implications
To the authors' knowledge, the current study was the first attempt to investigate the influence of NPIs for the pandemic on behavioral intention from two perspectives: operators and travelers. The important role T-NPIs play in travelers' decision-making processes was confirmed in previous research [14,15], whereas O-NPIs were highlighted but not incorporated into research models. Thus, this study sheds light on tourism literature by extending NPIs research in the pandemic context.
Differing from Das and Tiwari's [14] and Lee et al.'s [15] works, this research first confirmed that perception of COVID-19 had a significant and negative direct influence on behavioral intention during COVID-19 in an EMGB, confirming the findings of Reisinger and Mavondo [13]. Additionally, based on previous literature, new paths between desire (independent variable) and the two NPIs factors (dependent variable) were established and confirmed with significant results. More precisely, individuals' desire to take cruise tours was found to have significant positive impacts on their support for O-NPIs and willingness to adopt T-NPIs. The findings provided evidence that the new constructs and new paths integrated into the proposed model were appropriate and valid. Therefore, the current research has made contributions to theory broadening in the field of the MGB. Further, this will contribute to extending our knowledge of how NPIs play a role in predicting cruise tourists' behavioral intentions in the pandemic and post-pandemic situation.
Moreover, desire was found to be a powerful predictor of both behavioral intention during COVID-19 and after COVID-19, coinciding with findings in earlier studies [34,54,55]. Figure 2 shows that O-NPIs had a stronger impact than T-NPIs on individual's travel intention amid COVID-19. Nevertheless, the opposite result was found when predicting travel intention after COVID-19. In this aspect, this study suggests that the impacts of T-NPIs and O-NPIs are different at different stages of the pandemic, providing a significant empirical reference for future tourism research.

Practical Implications
Since this study is the first one to focus on the effects of NPIs on cruise travel intentions in the context of the COVID-19 pandemic, the findings provide several practical implications for cruise operators, travelers, and tourism policymakers.
First, as effective health-protective behaviors, the NPIs were found to be significant to the cruise industry in the context of COVID-19. The findings in this study revealed that potential cruise travelers were more likely to express a high degree of travel intentions through the adoption of T-NPIs and choosing a cruise that implements effective NPIs. Therefore, this study provides evidence that supports supporting cruise operators to restart operations by implementing stringent NPIs during the pandemic. A variety of promotions and educational activities regarding NPIs are necessary to help reduce potential customers' apprehension of the COVID-19 infection during the trip and stimulate their travel intention. A previous study found that the financial burden caused by personal NPIs may decrease individuals' willingness to adopt NPIs [14], and thus operators can encourage T-NPIs adoption by providing free masks and sanitizers. Meanwhile, cruise managers and crews should implement these NPIs procedures strictly in order to offer tourists the safest possible tour experience, which can help improve customers' brand trust. Previous studies have shown that tourist experience and brand trust are positively associated with word of mouth and loyalty [56][57][58], which are essential to cruise industry sustainability. Additionally, considering travelers' experiences, O-NPIs procedures should also be aligned with the stage of the pandemic. While O-NPIs had a significant effect on tourists' travel intentions during COVID-19, O-NPIs had a non-significant effect on travel intention after COVID-19 with negative sign. This finding suggests that cruise operators should simplify NPIs procedures after the pandemic.
Second, given that this study highlights desire as a powerful antecedent of potential travelers' behavioral intentions and NPIs, cruise operators should stoke their desire for cruise trips. Cruise operators need to build a positive and safe image by introducing NPIs procedures in video advertisements and promotions to enhance individuals' anticipated emotions and attitudes, which can further shape their desire for a cruise, especially during the period of a pandemic, when tourism settings with full security remain limited.
Third, government agencies should formulate relevant policies to supervise the implementation of cruise O-NPIs (e.g., imposing short-time shut-down on cruise line operations when infection incidents occur). Furthermore, health authorities should provide immediate assistance when a suspected case is identified.

Limitations and Future Research Directions
This study conducted a survey focusing on Koreans who had the experience of taking cruise tours. Thus, this study was limited to one country and thus, future researchers should test the study's research model to assure generalizability. Since it is difficult to reach a target population in the COVID-19 situation, the sample size (288) of this study is slightly small. Considering the importance of information for potential customers in the cruise marketing process, future research is recommended to incorporate people with no cruise experience. People from different cultures and socioeconomic backgrounds should also be considered to identify the factors affecting individuals' perception of the pandemic and NPIs. Future research should also use in-depth interviews with destination residents and government agencies to extract other unique items from the sub-dimensions of NPIs. Additionally, 77.1% of the respondents had a university degree, which might be high proportion. This might be attributable to the fact that people with a higher level of education are more likely to take cruise tours and also accept an invitation to participate in online surveys.