Impact of Altered Holiday Plans Due to COVID-19 on Tourist Satisfaction: Evidence from Costa Daurada

Abstract

The COVID-19 pandemic altered the holiday plans of many people. Whether it was due to travel bans or the fear of contracting the infection, people modified, among other aspects, their chosen destination, travel transport, accommodations, length of stay, and activities to be undertaken during the stay. In this context, we aim to disentangle the effect of these changes on tourist satisfaction. Previous research on the effects of COVID-19 on the tourism sector has studied the shrinkage of tourism demand, changes in tourist behaviour and adaptation processes on the supply side. Nonetheless, few works have analysed changes in tourists’ plans. Two main hypotheses have been put forward. First, tourists might be dissatisfied given that they could not attain their holiday expectations. In contrast, the second hypothesis suggests that those individuals who changed their holiday plans might be more satisfied because they diminished their perceived risk of contagion. We have used data drawn from a survey of tourists (N = 2009) who visited Costa Daurada, a very popular Mediterranean coastal destination just after the end of the Spanish lockdown. Then, statistically significant differences in satisfaction levels between the groups that altered their plans and those who did not are assessed by means of Kruskal–Wallis and Wilcoxon Rank Sum tests. Results signal that tourists were not more dissatisfied when they had modified their initial holiday plans. Indeed, the overall satisfaction of those visitors who switched their initial destination to travel to Costa Daurada was slightly lower, and the difference was significant, compared to the ones who were planning to travel there from the very beginning. Satisfaction was not significantly lower for those who changed their holiday plans in the case of the rest of the items analysed (transportation, length of stay, accommodation, and overall activities). On the contrary, in the case of activities, changes apparently contributed to mitigate the risk perception and led to a better tourist experience. Results also suggest that tourists were willing to adapt to a new situation in order not to renounce their holidays. In terms of implications for destination management and stakeholders, the main conclusion is that continuous cooperation and mutual trust are key to adapting to turbulent environments in which risk perception becomes central.

1. Introduction

The spread of SARS-CoV-2 caused a profound depression in activities in the tourism sector throughout the world. The impact was particularly strong during the first months of the pandemic when severe restrictions on mobility, economic activities, and the gathering of people were imposed (Gössling et al., 2021). According to GWI (2020), approximately 50% of American consumers and 38% of British consumers voluntarily cancelled, delayed, or were forced to cancel their initial holiday plans during the COVID-19 pandemic. Nevertheless, in spite of the difficulties faced by the tourism industry, its activities did not come to a complete halt; rather, the industry responded to the demand of visitors who wished to travel notwithstanding the barriers to mobility and the fear of contracting the illness (Roman et al., 2020). Restrictions on mobility, certain activities, and more restrictive capacity limitations, as well as fear of infection, resulted in holiday changes, as the incidence of COVID-19 became a key element in potential tourists’ decision-making processes (Pappas & Glyptou, 2021). As a result, many tourists were obliged to modify, to some extent, their initial plans with changes affecting components of the tourism product such as the choice of destination, accommodation, the length of the stay, the mode of transportation, or the selection of activities. Since little is known about the consequences of these decisions, the present work aims to shed some light on the effect of altering holiday plans on tourist satisfaction. This emerges as a research gap in the extant literature, as the modification of holiday plans could lead to changes in tourist satisfaction that have not been explored in previous research.

In a context where tourists were not having the holidays that they initially intended, the process of tourist satisfaction formation could have been seriously hampered. In this vein, following Oliver (1980) and Castañeda et al. (2007), consumer satisfaction depended on the comparison between the perception of services or products and previous expectations. Within the tourist industry, the comparison takes place between the real experiences that tourists had at the destination and those expectations (Montero & Fernandez-Aviles, 2011). COVID-19 introduced a new scenario in which the expectations attached to the holidays were more difficult to meet, as the initial holiday plans might have been modified in several ways. In this sense, the tourist might have had to change the destination, the mode of transport chosen to reach it, the length of stay, the accommodation, the activities undertaken, and other aspects related to the holiday. These changes, at times, were the result of the restrictions imposed by governments at different levels to deter the expansion of the virus, which obliged visitors to amend some or all of the characteristics of the holiday (Seyfi et al., 2023). Altered plans were also the consequence of fear of the illness, which made visitors more likely to reject certain aspects of the trip or the stay that they deemed risky to their safety (Shin et al., 2022; Matiza, 2022). As a result, modified holiday plans could lead to frustration.

In contrast, some tourists experienced relief, which enhanced their satisfaction due to the feeling of attained safety stemming from the change of plans. This occurred since tourists’ psychological needs were changed as a result of the pandemic (Cheung et al., 2021; Kovačić et al., 2023). In this sense, empirical evidence has signalled that, during the COVID-19 crisis, the feeling of safety was critical for tourist satisfaction (Lu & Atadil, 2021; Mwesiumo & Abdalla, 2023). Moreover, tourists’ priorities before and after the pandemic regarding tourist services changed, and the preference for health safety has gained ground (Mallick et al., 2022; Srivastava & Kumar, 2021; Kum et al., 2024; Mirzaei et al., 2023).

Disruptions like those caused by COVID-19 obliged both the providers of tourism services and the consumers to adapt to the new situation. Within this process of adaptation, the negative effect on satisfaction associated with the frustration caused by the changes to the initial tourists’ plans opposed the positive effect attached to the gain in safety that stemmed from it. Depending on which of these two drivers prevailed, the effect on satisfaction took one direction or the other. The effect of the disruption on the activity of the tourism sector was more intense at the beginning of the pandemic, when knowledge about it was very limited, and the massive production of an effective vaccine was still far off. It must be taken into consideration that the demand of the tourism industry is very sensitive to any disruptions that emerge as a threat to the visitors’ physical health (Yüksel & Yüksel, 2007). As a consequence, the spread of SARS-CoV-2 put some specific types of destinations in a more vulnerable position (Ahmad et al., 2023; Duro et al., 2021; Wickramasinghe & Naranpanawa, 2023). All in all, tourist satisfaction can be seriously damaged if the visitor does not feel sufficiently safe (Alegre & Garau, 2010; Soliman et al., 2024).

Despite the fact that there is a vast literature assessing the different impacts of the pandemic on diverse aspects of the activity of the tourism sector, the number of contributions focused on tourist satisfactions has been relatively limited (Mallick et al., 2022). Indeed, to the best of our knowledge, this is the first work that attempts to evaluate to what extent tourist satisfaction is impacted when tourists change their holiday plans. Previous literature has examined whether the factors that determine tourist satisfaction have switched to different ones during the pandemic, and to a lesser extent, whether there has been an impact of the pandemic on the levels of overall tourist satisfaction. Though, no attention has been paid to the consequences of the alteration of holiday plans. In this context, gaining insight into the consequences of altered visitors’ plans on tourist satisfaction will be of great aid for the managers of firms operating within the tourism industry, as well as destination managers, to properly cope with future disruptions. The results obtained will aid destination managers and firms operating within the tourism sector in settling the priorities to keep their customers satisfied in situations where initial expectations of travelling are no longer feasible. Data generated during the year 2020 are particularly valuable for this object, as the difficulties encountered in enabling a feasible coexistence of tourist activities with the constant threat of contracting the illness and the efforts to deter its spread reached their climax.

Hence, the main aim of this paper is to shed light on whether there were significant differences in the satisfaction levels of tourists who changed their initial travel plans and those who did not in 2020. Two main hypotheses are examined.

H1: 

Those visitors who changed their plans were less satisfied compared to the ones who did not, due to the frustration of the initial plans.

H2: 

Tourists who changed their plans were more satisfied as they felt safer.

The current research endeavour focuses on Costa Daurada, a top coastal tourist destination, as the case study. Costa Daurada is located in Catalonia in the north of Spain, and it is a popular destination for international and domestic tourists. A survey of tourists who visited Costa Daurada was conducted (N = 2009), and tests to assess the significance of satisfaction between tourists have been calculated. Prior to the analysis, a theoretical review on satisfaction and tourism during COVID-19 was carried out. Finally, a discussion of the work and significant findings is presented.

2. Literature Review

2.1. Tourist Satisfaction

Tourist satisfaction is a key factor, not only in the tourism economy but also in general for most economic activities. In this vein, it is conceived as a cornerstone to attain business success in competitive environments (Morgan et al., 1996). In the specific context of the tourism industry, tourist satisfaction is regarded as essential for the survival and future of any tourism product or service (Gursoy et al., 2003; Kozak et al., 2004). Several scholars have found that it has a substantial impact on destination selection, product and service consumption, and the decision to revisit (Kozak & Rimmington, 2000; Wong & Law, 2003; Yoon & Uysal, 2005; Jang & Feng, 2007).

Even though there is no dispute about its importance in the tourist industry, the sources of satisfaction have yielded a diversity of theoretical frameworks (Giese & Cote, 2000). Two main perspectives oppose each other. Following the cognitive approach, satisfaction is the result of a purely cognitive individual process where expectations are compared to performance (Oliver, 1980). Within this framework, the confirmation/disconfirmation paradigm has been widely considered the most suitable framework to analyse the formation of tourist satisfaction (Wirtz et al., 2000). The aforementioned paradigm establishes the existence of some pre-consumption standard that is compared to the perceived performance that the consumer obtains (Mattila & Wirtz, 2000). In contrast, other works have suggested that satisfaction stems basically from emotions. In this vein, Westbrook (1980) signals that it is the result of an emotional evolution, while Engel et al. (1993) suggest that it is an emotional response derived from a consumption experience. Posteriorly, other authors introduced a combined approach, which was referred to as the cognitive-affective view by Rodríguez del Bosque and Martín (2008). According to this point of view, satisfaction is the result of the combination of individuals’ cognitive judgments and consumption experiences (Oliver, 1980, 1993; Bigné et al., 2005). The introduction of emotions into the theoretical framework of tourist satisfaction is fully justified, and more particularly, in the service industry where emotions are a critical part of the experience (Barsky, 2002). The reason lies in the fact that consumers are highly likely to experience affective responses when they interact with the service and the personnel (Zins, 2002).

2.2. The Effect of the Pandemic on Tourist Satisfaction

The number of works that have examined the incidence of the pandemic on tourist satisfaction is relatively limited, and these are most often centred on assessing variations in the main determinants by comparing the situation before and after the spread of the virus. The underlying reasons that account for these changes have received less attention. Srivastava and Kumar (2021) pinpoint three elements that are likely to mediate the determination of satisfaction: the negative health consequences of the disease, the guidelines and advisories issued by health agencies, and media coverage of the pandemic. Years before the outbreak of SARS-CoV-2, Adam (2015) signalled that cognitive risk can damage tourists’ travel and tourism experiences. Cognitive risk can be attached to individuals’ risk perception, which has an emotional impact on the tourist experience, and hence, an impact on tourist satisfaction. It becomes apparent that the threat of infection brought in elements that were likely to influence the tourist experience and, therefore, to impact tourist satisfaction. These factors, which were to some extent beyond the control of tourist destinations, are likely to have mediated the relationship between the determinants of tourist satisfaction and overall tourist satisfaction. Nonetheless, the way these factors have shaped tourist satisfaction has been seldom explored. In this sense, how the affective and cognitive responses to the potential risks attached to travel under the threat of contracting the infection should at least be considered as hypothetical factors to discuss. In some specific cases, like the analysis of the effect associated with changes in holiday plans, no previous research has been found. By contrast, even though the number of contributions is not countless, some researchers have attempted to gauge the direct impact of the pandemic on tourist satisfaction and to what extent the determinants of tourist satisfaction changed because of the spread of COVID-19. With respect to empirical works that have explored the changes in tourist satisfaction scores caused by the pandemic, it must be considered that this type of analysis is not easy to develop, as it requires the availability of comparable satisfaction scores gathered prior to the pandemic as well as once it started. To the best of our knowledge, only Mallick et al. (2022) have made this comparison. Their results show a slight decline in tourist satisfaction after the spread of COVID-19. Many other works have studied whether the determinants of tourist satisfaction changed during the pandemic. Most of the contributions in the analysis of factors that account for tourist satisfaction have used data drawn from customers’ hotel evaluations on Trip Advisor to discern whether there has been a change in the most frequently used words connected to satisfied and dissatisfied customers.

These works concur that there are many elements that were irrelevant prior to the pandemic but became critical for the satisfaction of hotel customers, such as guarantees of health safety (Mallick et al., 2022; Song et al., 2022; Hong et al., 2020). Mallick et al. (2022) analysed the impact of the pandemic on tourist satisfaction by means of a survey launched in Costa Daurada both in 2019 and 2020 and found that, despite the fact that significant determinants of tourist satisfaction were the same in both years, the intensity of their influence on tourist satisfaction had changed. In this sense, there was a non-negligible increase in the importance of safety. These works are relevant, as they unveil significant differences in the tourist satisfaction prior COVID-19 and amid it.

The present contribution focuses on the effects of changes in initial holiday plans on tourist satisfaction. As stated by W. Zhang et al. (2005), under the threat of SARS, these changes might be voluntary or non-voluntary. Voluntary changes of plans could be, for instance, the consequence of either visitors’ fear of contagion or the belief that the restrictions and other prevention measures imposed were going to impinge on the tourist experience. On the other hand, non-voluntary changes would be a direct consequence of the prohibitions implemented (for instance, entry bans imposed by some countries). Departing from the definition of tourist satisfaction, which involves the cognitive and affective comparison between expectations and posterior service performance, when assessing alterations of tourists’ plans because of COVID-19, the tourists’ decision-making process and the implications derived from the changes in holiday plans must be considered.

2.3. The Tourist Decision-Making Process and the Pandemic

Consumer decision-making is an ongoing problem-solving process in the search, purchase, use, evaluation, and disposal of products and services (Belk & Coon, 1993; Valaskova et al., 2015). Schiffman et al. (2010) concluded that consumers decide to buy products not only to solve their problems but also to meet their expectations, which are strongly associated with satisfaction (Jing & Rashid, 2018). From the point of view of the tourism industry, the decision-making process of consumers is complex and involves several factors, such as why, where, when, how, and with whom to travel to seek holiday experiences (activities), along with the length of stay and the budget (Smallman & Moore, 2010). The process becomes even more complicated under circumstances of uncertainty and rapid change (Pappas, 2019).

The decision-making process can be deeply impacted by external disturbances that make the potential visitor modify their intended plans. In this vein, the stress theory ascertains that a public health emergency, such as the COVID-19 pandemic, creates negative emotions and influences cognitive assessment, resulting in excessive avoidance behaviour and blind obedience (Afouxenidis & Chtouris, 2020). In previous crises, changes in tourism behaviour also became apparent. In the context of risk perception, travel avoidance becomes a feasible scenario (Jónsdóttir, 2011; Cahyanto et al., 2016).

Risk perception is a critical concept in this discussion. As synthetized by Cui et al. (2016), the definition of tourism risk perception is determined by the background of scholars, and as a consequence, three different conceptions can be distinguished:

• Tourists’ subjective feelings regarding the negative consequences or negative impact the trip may have (cognitive psychology).
• Tourists’ objective evaluation regarding the negative consequences or negative impact the trip may have (consumer psychology).
• Tourists’ cognitive of exceeding the threshold portion of the negative consequences or negative impact the trip may have (travel safety).

The first two definitions oppose each other, as the first of them is grounded in subjective factors while the second stems from objective ones. At the same time, risk perception depends on internal factors, which centred in the individual, and external factors, mainly information sources such as mass media or travel advisories (E. C. L. Yang & Nair, 2014). During the COVID-19 pandemic, risk perception and risk aversion were again critical predictors of people’s travel intentions (Y. Yang et al., 2020). Williams et al. (2022) conducted an international survey in the five most powerful source tourist markets (USA, China, France, Germany, and the UK) where respondents were asked about the hypothetical holidays that they would have taken in the absence of COVID-19. The authors concluded that people’s intention to take an international holiday was sooner or later affected by the tolerance of COVID situational risk, tolerance of general risk, and perceived competence to manage COVID situational risk. Risk perception and risk aversion were reflected in changes in consumer preferences. Consistent with this view, the empirical evidence provided by Fuchs et al. (2024) shows that higher risk perceptions account for declines in travelling intentions to high-risk destinations, while there are some moderating factors of this perception: social and time distance and limiting interactions to their in-groups. People were willing to travel to destinations with lower tourist densities and better sanitary conditions; tourists also altered their usual trip durations, and domestic travel gained ground in comparison to international travel (Krouskos, 2020; Interface Tourism, 2020; DNA, 2020). It must be taken into account that these works, nonetheless, depart from “hypothetical holiday trips” instead of actual tourist behaviours.

The influence of COVID-19 has driven the preference for avoiding closed spaces. In this vein, empirical evidence has highlighted a growing trend of travelling to the destination by private modes of transport instead of alternatives that involve sharing closed spaces with other people (Ivanova et al., 2021) or spending the holiday in outdoor spaces such as national parks (Seong & Hong, 2021). It has also led tourists to delay their purchase decisions (Hall et al., 2020), given their hesitance regarding late bans on mobility or other sorts of restrictions. Influence on daily routines during the stay is also relevant. Visitors considerably reduced their visits to restaurants, coffee shops, and other entertainment venues (Kourgiantakis et al., 2021). Finally, destination choice and avoidance of certain places were clearly influenced not only by the real incidence of the illness but also by subjective conjectures formed in the collective worldview (Lu & Atadil, 2021).

Tourist behaviour in scenarios like the one brought on by SARS-CoV-2 is likely to be driven by emotions. In this line, studies that delve into the affective roots of visitors’ decisions are particularly interesting. Some studies have focused on the underlying psychological motivations that account for behavioural changes with respect to holiday decisions. According to Kock et al. (2020), the threat of infection makes travellers more xenophobic and thus more reluctant to travel to foreign countries. Many tourists became oversensitive to crowds, as well as more ethnocentric, and developed a preference for group travel, travel insurance, and destination loyalty that built a feeling of safety, thus diminishing the risk perceptions attached to the holiday. Similarly, K. Zhang et al. (2020) found that the circumstances of the pandemic can act as a magnifier of tourists’ negative emotional reactions and risk perception when making decisions.

2.4. Effect of the Alteration of Holiday Plans on Satisfaction

A limited number of works have tried to measure to what extent tourists changed their initial holiday plans due to the pandemic. Altınay Özdemir and Yildiz (2020) stated that a great proportion of potential Turkish tourists who were affected by time and financial restrictions postponed or even cancelled their holidays because of the pandemic during 2020. Some surveys have tried to quantify the proportion of people who changed their initial holiday plans. Cvijanović et al. (2021) found that 46.5% of respondents to a semi-structured questionnaire conducted in September of 2020 had modified, to some extent, their holiday plans as a result of the pandemic. Similar conclusions were reached by Kourgiantakis et al. (2021). Of the respondents of their snowball survey, 46% stated that they had cancelled or modified their initial holiday plans in 2020. Pásztor et al. (2020) found by means of a detailed survey that one-third of responders totally abandoned the idea of having holidays in 2020, 24% switched from an international destination to a national one, 24% changed the country and destination but still travelled to an international destination, and only 10% of the respondents stuck to their initial times and choice of destination.

The hypotheses of the present study are that the main expected outcomes derived from the alteration of the tourists’ holiday plans are, on the one hand, frustration due to the impossibility of having the desired holidays, or on the other hand, relief, as the risk of contracting the illness diminishes as a result of the changes.

Following the tourist satisfaction theories, tourist dissatisfaction should arise when the expected holiday performance is not attained (Oliver, 1980, 1993). Several studies have explored to what extent each of the different dimensions of the products and services that the visitors consume at the destination accounts for overall tourist satisfaction. However, this approach tends to ignore the effect of negative features (Alegre & Garau, 2010). Among the negative features, the impossibility of having the holiday that was initially planned must be considered. In this sense, bad weather encountered by visitors has been identified as an element that obliges them to make changes in their trips, which results in a decline in satisfaction (Hübner & Gössling, 2012; Kim et al., 2017; Becken & Wilson, 2013). The amount of research exploring the impact of holiday plan changes on tourist satisfaction is very scarce. In the specific context of the COVID-19 pandemic, to the best of our knowledge, there is still no study that focuses on the effect of the alterations of holiday plans on tourist satisfaction.

On the one hand, there is a body of research that has provided evidence of the increasing positive influence of safety on tourist satisfaction during the pandemic. Under the threat of COVID-19, positive ratings of hospitality facilities on Trip Advisor and similar platforms were highly dependent on safety (Srivastava & Kumar, 2021; Nilashi et al., 2022; Song et al., 2022; Sun et al., 2022; Yu et al., 2022). The aforementioned studies shared the same methodology, which involved the study of the prevalence of the most frequent words used in customers’ reviews in the ratings of satisfied and dissatisfied guests. There is also empirical evidence of the positive role of safety during the pandemic on satisfaction beyond the specific domain of the hospitality sector. This includes restaurants (Ababneh et al., 2022), airport facilities (Ma et al., 2022), and tourist destinations as a whole (Mwesiumo & Abdalla, 2023; Mallick et al., 2022). This latter work concluded that the influence of safety during the COVID-19 pandemic on tourist satisfaction had doubled compared to the situation prior to the spread of the disease.

In summary, there is evidence of the alteration of tourists’ behaviours and decisions with respect to original travel intentions. There is also evidence of changes in tourist preferences and perceptions. The impact of these changes on tourist satisfaction has not been studied yet, nevertheless.

3. Data

3.1. Study Area

Costa Daurada is a stretch of the Mediterranean coastline located southwest of Barcelona, in the Spanish region of Catalonia. It is a popular mature coastal destination (Domènech et al., 2023). According to data provided by the Costa Daurada Tourism Observatory, the area was visited by over 5.1 million tourists in 2019, who accounted for around 19.7 M overnight stays. These figures sharply dropped in 2020 due to the pandemic. The number of visitors fell to 1.5 M, while the number of overnight stays dropped to 5 M. Tourism activities are mostly centred in the municipalities of Salou, Cambrils, and Vila-seca, which concentrate around 75% of the total tourism capacity of the tourist region. The population of the three coastal cities ranges from 20,000 to 35,000. Moreover, medium-sized cities, Tarragona (132,000 inhabitants) and Reus (104,000), are well connected through public and private transport with the main tourist municipalities. The main top attractions in this area include the beach and the theme park Port Aventura, located between Salou and Vila-seca, which is one of the top five European theme parks (Anton Clavé, 2010); and to a lesser extent, cultural attractions and the Roman remains of Tarragona, which are rated as a UNESCO World Heritage site.

The particular characteristics of the Costa Daurada, as a mature destination with massive coastal tourism, make the case of study very interesting. In fact, it was suitable to divert, due to the aforementioned conditions, demand from other island coastal destinations, which were more vulnerable to the health crisis (Duro et al., 2021).

3.2. Data Collection

The data were drawn from a survey of tourism demand conducted by the Tourism Observatory of Catalonia in 2020. A total of 2009 valid questionnaires were collected by means of interviews with tourists who stayed overnight in Salou, Cambrils, and Vila-seca. The representativeness of the data was ensured by the random selection of the tourists who were interviewed and the choice of the places where the interviews took place, which comprised the main accommodation sites and key attractions. These interview locations had previously been identified by the professional staff of the Tourism of Catalonia to ensure that all the profiles of tourists were represented in the sample. Randomness is guaranteed as the choice of individuals who were interviewed on the street did not follow any specific criteria and was totally random, since no previous stratification or selection criteria were applied. Potential interviewees were simply approached randomly in the streets.

The surveying period comprised the peak tourist season (from June to September), as well as weekends and public holidays during the rest of the year. Obviously, no interviews were conducted until the end of the lockdown, which in Spain ended on 23 June 2020. The tourist season of 2020 started in Costa Daurada with the lifting of the lockdown. At that time, the incidence of COVID-19, measured as the number of cases per 100,000 inhabitants, was close to 0. During the summer months, the number of cases started to grow again, and new preventative measures, such as the compulsory use of face masks, were introduced. Despite the low levels of incidence of infection, some tourists were reluctant to answer the survey, as they perceived the interview could be risky. As a result, the traditional source of a hypothetical selection bias associated with the event of people rejecting to take part in a survey might be increased. Unfortunately, the existence of this bias is hardly measurable, and its mitigation is something almost impossible. The implications of this potential bias would only be significant if the levels of satisfaction of those tourists who refused taking part in the survey had statistically significant values compared to those who answered the questionnaire.

Respondents were asked several questions to allow the gathering of information with regard to their socio-demographic features (gender, age, and place of origin), trip characteristics (length of stay, whether it was the first time they visited the destination, type of accommodation, with whom they were travelling, and the means of transport used to reach the destination). Tourists were also asked to report their level of satisfaction based on a 5-point Likert scale (from 1, very low, to 5, very high) with respect to a total number of 14 elements: cleanliness of public areas, safety, the kindness of locals, accommodation services, public transport, facilities for pedestrians, entertainment and nightlife, green areas, the cleanliness of beaches and the sea, facilities on beaches, price-to-quality ratio, signage, restaurants in general, and the overall degree of satisfaction. These elements are intended to fulfil the diverse dimensions of tourist satisfaction taking into consideration the specific characteristics of the destination. In addition to these items, tourists were asked to assess the degree of overcrowding at the destination. Finally, the questionnaire also included some questions that signalled whether the visitors had changed their travelling plans as a consequence of the pandemic. This change in the holiday plan involved five possible items: destination, transport used to travel to the destination, length of stay, accommodation, and activities undertaken.

3.3. Descriptive Statistics

Table 1. Descriptive statistics of the profile of visitors.

Category	Subcategory	Freq.	Percentage
Gender	Male	1053	0.52
	Female	956	0.48
Repeat Visit	Yes	1816	0.90
	No	193	0.10
Profile	Couples with no children	1104	0.55
	Families with children	672	0.33
	Travelling with friends	93	0.05
	Alone	140	0.07
Origin	Spain	1779	0.89
	France	127	0.06
	Other	103	0.05
Age	0–45	723	0.36
	46–65	844	0.42
	>66	442	0.22
Accommodation	Hotel	480	0.24
	Camping	131	0.07
	Apartment	246	0.12
	Second home	1011	0.50
	Other	141	0.07
No. of nights	1–3	525	0.26
	4–7	577	0.29
	8–15	334	0.17
	>15	573	0.29
Transportation	Public Transport	136	0.07
	Private	1827	0.91
	Flights	46	0.02

presents the descriptive statistics of the variables, which portray the socioeconomic characteristics of the tourists as well as the characteristics of the trip for the whole sample of tourists who visited Costa Daurada in 2020. The variables exhibited in the table are dichotomous, and each sample observation can only be equal to 1 or 0. Therefore, the means should be interpreted as percentages of respondents who gave a specific answer.

Regarding the tourist demographics, Spain was the dominant country of origin (89%), followed by France (6%). Furthermore, 55% of tourists travelled with their partner (couple with no children), while 33% travelled with children. In terms of accommodations, half of the tourists stayed in second homes (50%) and 24% in hotel accommodations. Overall, a majority of the sample spent 4 to 7 nights and over 15 days (29% in both cases). The largest share of tourists was between 46 and 65 years old (42%), followed by those younger than 45 years old (36%). The most common means of arrival to the destination was private transport (91%). Finally, the percentage of tourists who had previously been to Costa Daurada was 90%. Undoubtedly, the pandemic caused profound changes in the profile of visitors (origin or recurrency) compared to previous years (Mallick et al., 2022).

The spread of COVID-19 deeply altered the dynamics of tourism activity in Costa Daurada in 2020. The sudden outbreak of the disease disrupted the demand for visitors in terms of the total number of arrivals and their characteristics. The availability of a very similar survey, which was launched in 2019 in the same territory, allows the comparison between the dominant tourist profiles of both years. In this vein, the percentage of foreign visitors fell significantly (from 57% in 2019 to 11% in 2020); second homes became preponderant (from 22% to 50%); while hotels lost popularity (from 49% to 22%). In terms of the length of stay, the shortest stays (1 night to 3 nights) and the longest ones (>15 nights) gained ground. Similarly, the proportion of visitors who had previously stayed in Costa Daurada also rose. In summary, national tourists who owned property in the area tended to replace international visitors who could not travel due to travel bans or were simply afraid of contracting the illness.

Table 2. Percentage of respondents who altered their holiday plans.

	Freq.	Percentage
Change of destination	460	0.23
Change of length of stay	213	0.11
Change of transport to reach the destination	40	0.02
Change of accommodation	35	0.02
Change of activities	353	0.18

exhibits the proportion of respondents who visited Costa Daurada in 2020 and reported having changed their holiday plans in 2020. A total of 23% of tourists travelled to Costa Daurada instead of their initially intended destination; 11% altered their length of stay, while 18% declared having modified their intentions regarding the activities during the stay. The changes related to the length of stay and accommodation affected 2% of respondents.

Some of the variables that measured the tourists’ degree of satisfaction had to be discarded, given that not all visitors could assess them, and their use would have led to an unacceptable proportion of missing values: accommodation services, public transport, entertainment and nightlife, green areas, and restaurants in general. Thus, the elements that were kept for the analysis included cleanliness of the public areas, safety, kindness of locals, facilities for pedestrians, green areas, signage, cleanliness of beaches and the sea, facilities on beaches, quality-to-price ratio, and overall satisfaction, as shown in

Table 3. Descriptive statistics of the variables of satisfaction.

Variables	Mean	SD	Max	Min
Cleanliness of the public areas	3.87	0.97	5	1
Safety	4.08	0.87	5	1
Kindness of locals	4.23	0.71	5	1
Facilities for pedestrians	4.18	0.87	5	1
Green areas	4.02	0.85	5	1
Signage	4.10	0.73	5	1
Cleanliness of beaches and the sea	4.01	0.94	5	1
Facilities on beaches	3.90	0.98	5	1
Price-to-quality ratio	3.94	0.81	5	1
Overcrowding	2.15	1.11	5	1
Overall satisfaction	4.29	0.68	5	1

.

Overall, and despite the pandemic, the tourists who visited the area in 2020 reported being highly satisfied, as the average level of overall satisfaction was 4.29. The majority of the satisfaction scores were above 4, with the sole exceptions of the price-to-quality ratio, cleanliness of the public areas, and facilities on beaches. With respect to overcrowding, it had an average rate of 2.15. This average is clearly below the results obtained in the survey of 2019, when it was 2.75. The underlying reason for this reduction is the plummeting of the number of visitor records in 2020.

4. Methods

The methods used involved exploratory factor analysis to reduce the number of variables in the analysis and non-parametric analysis in order to unveil significant differences in satisfaction levels between those tourists who changed their initial holiday plans because of the pandemic and those who did not.

4.1. Exploratory Factor Analysis (EFA)

In the first phase, exploratory factor analysis (EFA) was performed to extract and select the destination attributes. This statistical technique allows for the estimation of the relationship between an initial set of variables and constructs (Spearman, 1904). The result is a reduced number of latent factors that account for the former set of initial variables, which enables researchers to simplify complex analyses. In this sense, EFA provides a clear data image and also allows the use of the output obtained in subsequent studies (Field, 2000). Hence, the relationships between the destination attributes are examined in this preliminary stage with the aim of grouping them into a more manageable scale.

4.2. Non-Parametric Analysis (Mann–Whitney U and Kruskal–Wallis)

Secondly, destination attributes extracted from EFA are further utilised to investigate whether there were significant differences in satisfaction levels between tourists who changed their initial holiday plans because of the spread of COVID-19 and those who did not. The Mann–Whitney test is a non-parametric technique that enables researchers to evaluate the significance of differences between two independent groups when the variable is ordinal or continuous (McElduff et al., 2010). Similarly, the Kruskal–Wallis test is an extension of the Wilcoxon Rank Sum test that allows for comparing more than two independent categories. It is also a non-parametric statistical technique for determining if samples originating from the same distribution have statistically significant differences (Kruskal & Wallis, 1952; Corder & Foreman, 2009). We have used non-parametric tests because, in the case of Kruskal–Wallis and Wilcoxon Rank Sum tests, the difference with parametric tests is negligible under conditions of full normality. In fact, parametric tests are capable of performing better only if perfect conditions of normality are guaranteed (Gleason, 2013) and the variances of the subsamples are roughly homogenous (Vrbin, 2022). Also, some sample sizes of our comparisons make it advisable to use non-parametric techniques.

5. Results

5.1. Factor Analysis

In the first phase of the analysis, EFA was conducted by means of the ‘psych’ package (Revelle, 2019) in the R programming language (R Core Team, 2019). The variable named ‘overcrowding’ was removed, as it did not reach the minimum threshold required for significance (0.4). The Kaiser-Meyer-Olkin (KMO) measure of sampling adequacy (Kaiser, 1974) and Bartlett’s test of sphericity (Bartlett, 1951) were applied. The KMO measure was 0.9, hence above the 0.5 cutting value, while the result of Barlett’s test was 0.00. Consequently, the results of both tests were satisfactory.

A clean four-factor structure was extracted in which the factor loadings were all above the 0.4 cut-off value (Stevens, 1992). This structure outperformed other alternative structures with a lesser or larger number of factors in terms of the percentage of total explained variance (66.4%), eigenvalues and the values of Cronbach’s alpha. Regarding the latter, all of them exceeded the threshold of 0.5 (Nunnally, 1975), as shown in

Table 4. Exploratory factor analysis of the destination attributes.

	Factor 1	Factor 2	Factor 3	Factor 4
Safe and security
Cleaning of public areas	0.541	−0.021	0.251	0.043
Security	0.673	0.083	−0.032	0.051
Destination characteristics
Friendliness of the people	0.192	0.524	−0.073	0.022
Pedestrian facilities	0.144	0.503	0.031	−0.032
Green areas—nature	0.112	0.312	0.291	−0.021
Signage	−0.093	0.610	0.124	0.123
Beaches
Cleaning of beaches and the sea	0.042	0.013	0.774	0.000
Equipment on the beaches	0.014	0.121	0.491	0.141
Quality/price ratio
Quality/price ratio	0.011	0.000	0.000	1.000
	KMO	Communality	Eigenvalue	% of variance	C. Alpha
Safe and security
Cleaning of public areas	0.892	0.521	1.051	15.112	0.714
Security	0.893	0.543
Destination characteristics
Friendliness of the people	0.921	0.391	1.324	18.212	0.763
Pedestrian facilities	0.933	0.363
Green areas—nature	0.921	0.424
Signage	0.894	0.510
Beaches
Cleaning of beaches and the sea	0.892	0.642	1.293	17.011	0.721
Equipment on the beaches	0.910	0.434
Quality/price ratio
Quality/price ratio	0.89	0.881	1.132	16.034	1.000

Note: Overall KMO = 0.9, Bartlett’s test of sphericity = chi-square −5567.99; p value = 0; df = 36.

.

The four latent indicators of satisfaction that arose were safety and security, destination characteristics, beaches, and price-to-quality ratio. Each factor’s mean equalled 0. Hence, comparisons will be allowed in the sense that when a specific group has a mean value below 0, it implies that, on average, the members belonging to this group of individuals were less satisfied than the total sample.

5.2. Non-Parametric Tests

In the next stage, the extracted factors were further analysed using the Mann–Whitney U and Kruskal–Wallis tests to assess the significance of the differences in satisfaction levels. Comparisons between two groups were assessed by means of the Mann–Whitney U, while comparisons between larger numbers of groups required the use of the Kruskal–Wallis test.

Table 5. Mann–Whitney U tests of differences in tourist overall satisfaction.

Category	Frequency	Frequency %	Mean	SD	Wilcoxon Rank Sum Test (W)
Change in Destination
Tourists who did not change plans	1549	0.77	0.016	1.007	372,683 *
Tourists who changed plans	460	0.23	−0.055	0.975
Change in Transportation
Tourists who did not change plans	1969	0.98	0.001	0.997	38,998
Tourists who changed plans	40	0.02	−0.025	1.16
Change in Length of Stay
Tourists who did not change plans	1796	0.89	0.003	0.986	189,647
Tourists who changed plans	213	0.11	−0.022	1.116
Change in Accommodation
Tourists who did not change plans	1974	0.98	−0.004	1.004	30,723
Tourists who changed plans	35	0.02	0.245	0.748
Change in Overall Activities
Tourists who did not change plans	1656	0.82	−0.013	0.986	276,339 *
Tourists who changed plans	353	0.18	0.059	1.065

Note: Level of significance of the differences: * Significant at 10%.

presents the results of the Mann–Whitney U test related to overall satisfaction. Overall satisfaction is not a latent variable derived from the factor analysis. For this reason, and with the object of allowing comparisons of the means on the same scale as the factors, the values of overall satisfaction were standardised by subtracting the mean and dividing by the standard deviations in all observations. The table reports whether the level of satisfaction of those tourists who changed their holiday destination was significantly different from that of those who did not. Likewise, the effects of the changes attached to the length of stay, accommodation, mode of transport used, and activities undertaken on overall satisfaction were also assessed and shown in Table 5.

The results reveal that the satisfaction level was slightly lower for those who changed their plans regarding the choice of the destination (−0.055) compared to those who did not ($\overline{x}$ = 0.016). Similarly, the reported satisfaction was also lower for those who changed their plans about the mode of transport selected to travel to the destination (−0.025) and the length of stay (−0.022). Conversely, the level of satisfaction was slightly higher for those who changed their holiday plan with respect to accommodations (0.245) and activities to be undertaken at the destination 0.059). Nevertheless, mean differences in overall satisfaction were just significant, and indeed weakly significant, in the specific cases of the choice of the destination and the activities to be undertaken, while the differences were not significant regarding changes in holiday intentions when dealing with length of stay, transport used to reach the destination, and accommodations.

Tourists who modified their plans might be a heterogeneous group of individuals. In order to examine whether the lack of significant differences in overall satisfaction levels could be attributed to this, the Kruskal–Wallis test was applied. This was used to investigate the differences in overall satisfaction levels within the different groups of tourists who changed their initial plans. For instance, in the particular case of those who changed their destination, it was used to assess whether there were differences in overall satisfaction between those tourists who were planning to travel to other continental Spanish areas, those who were intending to spend their holiday in the Balearic or the Canary Islands, those who were willing to travel to a foreign country, or other options.

Table 6. Kruskal–Wallis test of differences in overall satisfaction of tourists who changed their holiday.

Category	Frequency	Frequency %	Mean	SD	Chi-Squared	df
Intended destination of tourists who changed their travel plans
Continental Spain	164	0.35	0.004	0.939	1.4809	3
Spanish islands	60	0.13	0.011	0.926
Foreign countries	236	0.50	0.018	1.038
Other	14	0.03	−0.396	1.339
Intended length of stay of tourists who changed their travel plans
02–03	9	0.04	0.369	0.699	2.7322	4
04–07	29	0.14	−0.093	1.085
08–14	15	0.07	0.251	0.685
15–30	78	0.37	−0.061	0.954
31–800	82	0.38	0.005	1.088
Intended transport use of tourists who changed their travel plans
Public Transport	27	0.68	0.027	0.776	1.1445	2
Private	5	0.13	0.415	0.699
Flights	8	0.20	−0.351	1.67
Intended accommodation of tourists who changed their travel plans
Hotel	28	0.80	0.014	1.005	2.824	4
Camping	1	0.03	−0.904	NA
Apartment rental	4	0.11	0.085	1.142
Second home	1	0.03	1.074	NA
Other	1	0.03	−0.904	NA

exhibits the satisfaction levels of each of the groups that changed their decision. The conclusion is that no significant variances emerged from any of the different categories, and hence, the lack of significant differences in overall satisfaction between tourists who changed and those who did not change had nothing to do with tourists’ heterogeneity.

Table 7. Mann Whitney U test of differences in satisfaction regarding destination attributes.

Change in Destination
			Safety and security		Destination characteristics
	Fre.	Freq. %	Mean	SD	Mean	SD
Tourists who did not change plan	1549	0.77	0.011	1.741	0.031	2.902
Tourists who changed plan	460	0.23	−0.037	1.698	−0.106	3.034
Wilcoxon Rank Sum test	w		364,154		365,990
			Beaches		Quality/price ratio
	Fre.	Freq. %	Mean	SD	Mean	SD
Tourists who did not change plan	1549	0.77	−0.018	1.745	−0.001	0.989
Tourists who changed plan	460	0.23	0.062	1.717	0.002	1.038
Wilcoxon Rank Sum test	w		346,748		353,505
Change in Length of Stay
			Safety and security		Destination characteristics
	Fre.	Freq. %	Mean	SD	Mean	SD
Tourists who did not change plan	1796	0.89	0.03	1.693	0.032	2.885
Tourists who changed plan	213	0.11	−0.251	2.015	−0.272	3.298
Wilcoxon Rank Sum test	w		202,066		194,121
			Beaches		Quality/price ratio
	Fre.	Freq. %	Mean	SD	Mean	SD
Tourists who did not change plan	1796	0.89	0.041	1.702	0.008	0.99
Tourists who changed plan	213	0.11	−0.346	1.991	−0.071	1.081
Wilcoxon Rank Sum test	w		210,070 **		197,734
Change in Transportation
			Safety and security		Destination characteristics
	Fre.	Freq. %	Mean	SD	Mean	SD
Tourists who did not change plan	1969	0.98	0.0001	1.726	−0.003	2.921
Tourists who changed plan	40	0.02	0.002	2.004	0.129	3.481
Wilcoxon Rank Sum test	w		38,138		36,792
			Beaches		Quality/price ratio
	Fre.	Freq. %	Mean	SD	Mean	SD
Tourists who did not change plan	1969	0.98	0.002	1.74	0.007	0.994
Tourists who changed plan	40	0.02	−0.101	1.689	−0.327	1.233
Wilcoxon Rank Sum test	w		41,931		45,326 *
Change in Accommodation
			Safety and security		Destination characteristics
	Fre.	Freq. %	Mean	SD	Mean	SD
Tourists who did not change plan	1974	0.98	−0.011	1.734	−0.024	2.922
Tourists who changed plan	35	0.02	0.608	1.462	1381	3.196
Wilcoxon Rank Sum test	w		27,248 **		24,947 ***
			Beaches		Quality/price ratio
	Fre.	Freq. %	Mean	SD	Mean	SD
Tourists who did not change plan	1974	0.98	−0.014	1.742	−0.003	0.999
Tourists who changed plan	35	0.02	0.802	1.331	0.18	1.055
Wilcoxon Rank Sum test	w		24,727 ***		31,134
Change in Overall Activities
			Safety and security		Destination characteristics
	Fre.	Freq. %	Mean	SD	Mean	SD
Tourists who did not change plan	1656	0.82	0.022	1.686	−0.018	2.865
Tourists who changed plan	353	0.18	−0.104	1.928	0.083	3.234
Wilcoxon Rank Sum test	w		294,255		280,845
			Beaches		Quality/price ratio
	Fre.	Freq. %	Mean	SD	Mean	SD
Tourists who did not change plan	1656	0.82	0.021	1.707	0.001	0.986
Tourists who changed plan	353	0.18	−0.099	1.88	−0.006	1.066
Wilcoxon Rank Sum test	w		299,169		290,793

Note: Level of significance of differences: * Significant at 10%, ** significant at 5%, *** significant at 1%.

shows the results of the Mann–Whitney U test regarding the differences in the degree of tourist satisfaction for each of the factors that stemmed from the factor analysis: safety and security, destination characteristics, beaches, and price-to-quality ratio. No significant differences related to any of the satisfaction factors are apparent in the case of alterations of holiday plans related to the choice of the destination or the activities to undertake during the stay. In the case of modifications of the holiday plan involving changes in the length of stay, those visitors who altered the duration of their stay were less satisfied with the beaches (−0.346), while tourists who changed their intentions about the mode of transport used to reach the destination were more satisfied with respect to the price-to-quality ratio (0.327). No other significant differences emerged related to the length of stay or the mode of transport selected for travelling. On the contrary, the table signals that tourists who altered their holiday plans related to accommodations were significantly more satisfied with regard to safety and security (0.608), destination characteristics (1.381), as well as beaches (0.802). This result is consistent with previous studies that indicated that accommodations became a very sensitive issue for tourist satisfaction during the pandemic (Srivastava & Kumar, 2021).

Table A1. Result of the Kruskal–Wallis test regarding the satisfaction of tourists who changed their holiday plans.

Category				Safety and Security		Destination Characteristics		Beaches		Quality/Price Ratio
	Frequency	Frequency in %	df	Mean	SD	Mean	SD	Mean	SD	Mean	SD
Intended destination of tourists who changed holiday plans
Continental Spain	164	0.35	3	−0.018	1.547	−0.039	2.714	0.119	1.672	0.01	0.971
Spanish islands	60	0.13		0.111	1.815	0.139	2.694	−0.184	1.654	0.028	0.958
Foreign countries	236	0.5		−0.017	1.835	0.04	3.187	−0.006	1.802	−0.033	1.029
Other	14	0.03		0.026	1.027	−0.813	2.567	−0.515	1.353	0.324	1.066
Kruskal–Wallis Test	chi-squared			0.89339		2.1432		3.5614		2.2868
Intended length of stay of tourists who changed holiday plans
02–03	9	0.04	4	1.002	1.146	1.566	2.991	0.886	1.374	0.261	1.206
04–07	29	0.14		0.154	1.665	0.39	3.572	−0.026	1.909	0.213	0.963
08–14	15	0.07		0.425	1.257	0.277	2.494	0.257	1.717	0.058	1.011
15–30	78	0.37		−0.085	0.685	−0.431	2.936	0.088	1.586	−0.071	1.039
31–800	82	0.38		−0.161	1.769	0.05	2.639	1.852	1.074	−0.047	0.959
Kruskal–Wallis Test	chi-squared			5.2757		5.9644		3.6487		3.5567
Intended transport use of tourists who changed holiday plans
Public Transport	27	0.68	2	0.083	1.711	0.14	2.071	0.134	1.527	−0.12	0.894
Private	5	0.13		0.473	1.643	0.438	2.502	−0.504	0.989	0.125	0.839
Flights	8	0.2		−0.575	2.57	−0.746	5.886	−0.138	2.842	0.326	1.418
Kruskal–Wallis Test	chi-squared			0.53747		0.27646		1.0325		2.49
Intended accommodation of tourists who changed holiday plan
Hotel	28	0.8	5	−0.143	1.643	0.186	3.169	0.13	1.766	0.025	0.975
Camping	1	0.03		−0.6	NA	−2.221	NA	−0.919	NA	−1.273	NA
Apartment rental	4	0.11		0.86	1.139	−0.343	4.244	−0.557	2.316	0.193	1.122
Second home	1	0.03		1.766	NA	2.23	NA	0.413	NA	1.072	NA
Other	1	0.03		−0.6	NA	−3.854	NA	−0.919	NA	−1.273	NA
Kruskal–Wallis Test	chi-squared			3.746		2.6083		2.0769		4.8257

and

Table A2. Result of Mann Whitney U test based on extracted factor of stay period of tourists who changed their travel plans.

Category			Safety and Security		Destination Characteristics		Beaches		Quality/Price Ratio		Overall Satisfaction
	Frequency	Frequency %	Mean	SD	Mean	SD	Mean	SD	Mean	SD	Mean	SD
Stay Period (Shorter/Longer) in changed plans
Shorter than Intended	74	0.35	−0.261	1.625	−0.174	2.90	0.011	1.74	−0.005	0.946	0.063	0.794
Longer than Intended	139	0.65	0.139	1.736	0.093	2.91	−0.006	1.747	0.003	1.031	−0.033	1.095
Wilcoxon Rank Sum test	w		4278 ***		4788.5		5177		4995.5		5150.5

Note: Level of significance of differences: *** significant at 1%.

presented in the annex aim to unveil hypothetical differences in satisfaction levels within groups of tourists who changed their holiday plans. According to the results of Table A1, no significant differences with respect to any of the satisfaction items (safety and security, destination characteristics, beaches, or price-to-quality ratio) arose between those individuals who changed their holiday intentions attached to the choice of the destination, the length of stay, their transport to the destination, and accommodations. Table A2 disentangles whether there were significant differences in the degree of satisfaction between those visitors who had a shorter than initially intended holiday and those who had a longer stay. Results signalled that those who had a shorter holiday than expected were significantly less satisfied with respect to safety and security (−0.261) when compared to those who stayed longer than initially intended (0.139). This result is intimately linked to the choice of accommodation. While shorter stays were more linked to hotels, longer stays were more likely to be linked to second homes or even apartment rentals (Gutiérrez et al., 2020). The latter were perceived as riskier in terms of contagion rather than the former (Aiello et al., 2022). No other significant differences became apparent with regard to the rest of the satisfaction dimensions, including overall satisfaction.

Frustration, supporting H1, and relief, supporting H2, are the drivers of the few significant differences that have arisen. The change of destination, despite the fact that it can be a voluntary decision caused by the risk perception attached to the destination (Fuchs et al., 2024) is the central element of most holiday plans. Consequently, this is the element of the holiday plan that is more likely to cause disappointment if initial expectations are not met. By contrast, higher significant satisfaction levels of tourists who modified their plans regarding accommodation and activities should be associated with a decline in the risk of contracting the disease. Activities and accommodation might not be perceived as important as the choice of the destination in holiday planning. At the same time, the potential sources of contagion might be as perceived closer, given that activities and accommodation could imply contact with other individuals and staying in indoor facilities. The fact that no significant differences have emerged for the majority of satisfaction indicators can be attributed, in general, to the fact that tourist satisfaction was not deeply affected by the alteration of holiday plans. Under the pandemic circumstances, those who travelled to the Costa Daurada for holidays were happy to do so, and were willing to assume a certain degree of risk that could be attenuated with certain decisions (accommodation or activities, for instance).

6. Discussion and Conclusions

The present contribution aimed to disentangle the extent to which tourist satisfaction was affected in visitors who changed their holiday plans because of the incidence of the COVID-19 pandemic, compared to those who did not. Data were drawn from a survey (N = 2009) of tourists who visited Costa Daurada, a top Mediterranean coastal destination during the holiday season of 2020, and hence amid one of the worst periods of lethality caused by the disease. Following W. Zhang et al. (2005), the alterations of the holiday plans under circumstances like the ones caused by COVID-19 could have been, on the one hand, forced decisions based on travel bans or other sort of restrictions that obliged the potential visitors to alter their initial plans. On the other hand, independent decisions were made, driven either by the fear of contracting the disease or by the feeling that under the restrictions imposed; it would not be possible to enjoy the holidays as originally expected.

Two hypotheses were put forward. First, tourists who could not have the holidays they had initially intended should have been disappointed, as their expectations could not be met. According to the theories that provide a theoretical framework to account for the process of tourist satisfaction formation, this situation should have resulted in a decline in satisfaction (Oliver, 1980; Yoon & Uysal, 2005; Giese & Cote, 2000; Wirtz et al., 2000). The second hypothesis stems from the empirical evidence that states that during the pandemic, there was a substantial growth in tourists’ preferences for health and safety during their travels (Mallick et al., 2022; Pásztor et al., 2020). Therefore, visitors who changed their plans could feel a certain relief that could contribute to an increase in the enjoyment of their holidays.

The survey asked visitors to report their overall degree of satisfaction with the holiday, in addition to their satisfaction level with another set of attributes of the destination. The set of satisfaction items was reduced by means of an exploratory factor analysis (EFA) used in four latent constructs: safety and security, destination characteristics, beaches, and price-to-quality ratio. The average degree of satisfaction of those who had changed their holiday plans was compared to those who did not alter theirs. The significance of the differences was assessed using the Mann–Whitney and Kruskal–Wallis tests.

Tests signalled that, even though scarce, there were some significant differences in terms of satisfaction levels. These are mainly related to the choice of destination (lower satisfaction if plans were changed) and activities and accommodation (higher satisfaction if plans were changed). More specifically, regarding overall satisfaction, those visitors whose initial destination was not Costa Daurada were less satisfied than those who planned to travel there and did not modify their intentions because of the pandemic. Conversely, those visitors who altered their activities during the stay were more satisfied than those who did not. Regarding the different satisfaction attributes that were considered, the most outstanding results were associated with accommodations. Tourists who changed their initially planned accommodations were more satisfied than those who did not, with respect to destination safety and security, destination characteristics, and beaches. In addition, those who changed their plans with respect to the mode of transport chosen to reach their destination were more satisfied than those who did not, with respect to the price-to-quality ratio. Finally, another interesting result was that significant differences in terms of the satisfaction associated with safety and security emerged between those who had a shorter holiday than initially expected and those who had a longer than initially planned stay. The former were less satisfied than the latter.

According to these results, the evidence obtained is too weak to support any of the hypotheses of this work. Hence, in opposition to the literature that states that the failure in fulfilling the initial holiday expectations should have a negative impact on tourist satisfaction, the modification of the plans would not lessen tourist satisfaction. Nevertheless, there was a relative prevalence of the significance of the differences that supported the hypothesis that changing the intended holiday plans resulted in a certain relief for tourists, and as a consequence, it was a driver for a higher level of tourist satisfaction compared to those who did not alter their plans. This conclusion must be taken with caution and cannot be generalised. Nevertheless, the results obtained concurred with previous evidence that showcased accommodations as a particularly critical element for tourists’ decisions during the worst periods of the pandemic (Pappas & Glyptou, 2021; Del Chiappa et al., 2022; Sánchez-Sánchez et al., 2024), as well as for tourist satisfaction (Nilashi et al., 2022; Jafari et al., 2023).

It stands out that in some cases differences in the degree of satisfaction occurred in destination attributes that do not have a tight relationship with the type of decision. For instance, those who changed their intended accommodation reported a significantly higher level of satisfaction with the beaches in comparison to those who did not. Hence, individuals’ emotional responses, as stated by Engel et al. (1993), should play a role in the formation of tourist satisfaction. Consistent with the concept of cognitive risk put forward by Adam (2015), the valuation of the tourist experiences will vary depending on the perception of the risk of contagion. Those tourists who replaced their original tourist plan for alternatives, which were deemed ‘safer,’ were more likely to enjoy their stay. On similar grounds, and in the specific context of COVID-19, K. Zhang et al. (2020) concluded that tourists’ negative emotional responses can be augmented by the fear of contracting the illness.

On the other hand, the evidence obtained is too scarce and weak to support the hypothesis that those tourists who modified their holiday plans were less satisfied for not being able to meet their initial expectations. It must be considered that the great majority of people had to go through totally unexpected life experiences in 2020 that put them under grave stress and exhaustion associated with the fear of contagion, and the resulting isolation caused social restrictions (Bao et al., 2020; Rania & Coppola, 2022). The loosening of mobility and activity restrictions opened the door for many of these people to have a holiday, even though it was within the new context of the ‘new normal’. The opportunity to enjoy a holiday is, in fact, a chance to heal from past stress and anxiety (de Bloom et al., 2011), and this reasoning is fully applicable to the consequences of the COVID-19 pandemic (Buckley, 2023; Naidoo et al., 2023; Buckley & Westaway, 2020).

In summary, the idea of having a holiday was sufficiently appealing, despite the difficult circumstances, and the alteration of the original plan was considered by most people as the least of the inconveniences. Tourists were willing to alter their holiday plans to be able to experience their holidays. The consequence was that destination managers and tourism businesses had to adapt to a new complex, and above all, uncertain scenario that allowed tourists to stay and cope with the risks associated with the threat of the spread of the virus (Rivera, 2020; Robina-Ramírez et al., 2021). Many managerial risks ensued and ranged from favouring the spread of the virus if the measures were not strict enough (Qiu et al., 2020) to leading towards visitors’ dissatisfaction on the other extreme if they were too stringent (Davras & Durgun, 2022).

Overall, tourists adapted their holiday plans to the new situation; therefore, the tourism industry was forced to readapt its products and services as well to remain in operation. The potential diminutions of tourist satisfaction were more likely to derive from malfunctions in the process of adaptation rather than directly from the mere modification of the expected holidays. The complexities attached to this process of adaptation were characterised by decision-making amid a situation of chaos control (Davras & Durgun, 2022), where many agents (visitors and service providers) were making decisions at the same time; these decisions not only impacted the outcomes of the others but also had an influence on others’ decisions.

Even though tourists were, in general, little affected by having to modify their plans, the challenge of keeping the levels of tourist satisfaction faced by the tourism sector was far from easy. Within crises that affect tourism demand, safety is a critical issue, but tourist satisfaction is contingent on tourists’ safety perception (Du et al., 2009). The equilibrium between safety, safety perception, and enjoying a pleasant holiday was a demanding wager, especially in a multistakeholder environment. Under these circumstances, the coordination between destination managers and service providers emerges as something essential. The management of tourist destinations had to cope with a diversity of agents pursuing the same goal: adapting to a landscape shaped by mobility restrictions, various prevention measures, significant shifts in visitor profiles, and the need to modify initial holiday plans. Hence, for tourist activity, close cooperation between stakeholders is required in situations characterized by the need of adaptation and dynamism (Tasci & Boylu, 2010). The design and implementation of specific actions in a turbulent environment require the establishment of a continuous dialogue where all the stakeholders of the destination feel adequately represented. Cooperation should contribute to enhancing agility, flexibility, and organizational performance (Jamal & Budke, 2020). This sort of collaborative strategy should be helpful in mitigating potential individual irresponsible behaviours at the firm level that are likely to yield adverse outcomes for the rest of the agents. In this line, it is important to adopt a standpoint of mutual collaboration, considering that the agent that might be responsible for the implementation of certain actions will not be the one that benefits most of them. Though, the damage if the actions are not implemented probably will be shared by all the stakeholders. Therefore, mutual trust is key to implementing measures (Handayani et al., 2022). Previous research points out that destination manager organizations and local governments should oversee leading this dialogue (Robina-Ramírez et al., 2022; Shrestha & L’Espoir Decosta, 2023; Gori et al., 2021). The implicit risks attached are slow decision-making processes and the lack of flexibility associated with situations where many actors confront their ideas at the same time (Mattessich & Johnson, 2018). Therefore, proper designs of the mechanisms of coordination, collaboration, and cooperation must be accepted by all parties. In summary, under difficult and changing environments affecting the operations of the tourism sector, the creation of committees at the destination level where all stakeholders feel adequately listened to and represented can be a successful tool to reach wide agreements, assess the effectiveness of the actions implemented, facilitate communication flows between the agents involved, and build trust. Additionally, flexibility and agility in the decision-making processes is imperative.

The present study is not exempt from limitations that must be taken into consideration. First, it can be argued that the survey uses data from tourists who stayed overnight at a specific tourist destination. There is a process of selection, which was particularly severe in 2020, since those individuals who decided to cancel their holidays and stay at home were not part of the sample. The profound alteration of the profile of the tourists who travelled to the area reflects this issue. Likewise, there is not information on those tourists who changed their minds about travelling to Costa Daurada and chose an alternative destination. While it is true that the lack of information can be a source of bias, the results obtained provide valuable information as to whether the visitors of a destination might be dissatisfied due to not meeting their initially desired holiday plans, nonetheless. Second, given that tourist satisfaction is in part the consequence of emotions, and these are very sensitive to risk perception, results might be influenced by the incidence of contagions at a specific moment in time as well as the media coverage about them.

These considerations open the door to future research, which might assess how the changes of holiday plans have affected tourist satisfaction in different destinations and during different times of the pandemic. On top of that, future studies should explore, beyond the COVID-19 pandemic, what are the reasons that lead tourists to change their plans and to what extent, and their implications for tourist satisfaction. In this sense, the present work has delved into the effects of a health crisis altering tourists’ plans. Further disturbances attached to external factors such as weather conditions, as well as failures of service providers at different levels (transport, accommodation, activities, food, and beverages) that oblige the visitor to modify the initial plan can be an interesting object of analysis. In this vein, given that tourist services are complex in the sense that they are highly dependent on the different components of their supply chain (Robina-Ramírez et al., 2022), it will be worthwhile to assess to what extent the level of tourist satisfaction related to each of the items that are significant determinants of overall destination satisfaction is affected when one or more of the items do not meet tourists’ expectations.